Real Estate Sales Agents
41-9022.00Median wage $52,830/yr193,370 employed (US)Rank #182 of 923 scored · top 20% by substitution
Rent, buy, or sell property for clients. Perform duties such as study property listings, interview prospective clients, accompany clients to property site, discuss conditions of sale, and draw up real estate contracts. Includes agents who represent buyer.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution40
Exposure35
Augmentation70
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
33 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
3%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%35
panel mean rating 2.4/5 → substitution pressure 35/100
Technical feasibility todayw 20%35
panel mean rating 2.4/5 → substitution pressure 35/100
Cost vs. human wagew 15%42
panel mean rating 2.7/5 → substitution pressure 42/100
Adoption barriersw 20%inverted — strong barriers lower the score50
panel mean rating 3.0/5 (barrier strength) → substitution pressure 50/100
Sector adoption velocityw 10%38
panel mean rating 2.5/5 → substitution pressure 38/100
Task breakdown (33 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
Generate lists of properties that are compatible with buyers' needs and financial resources.
89CI 79–100 · exposure 87 · augmentation 88 · importance 4.7/5 · click for rater detail
Generate lists of properties that are compatible with buyers' needs and financial resources.
89| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Major real estate platforms and brokerages are actively deploying AI-driven property matching and list generation in production; adoption is deep in the information-rich, digitized real estate sector. Agents and firms are already using these tools, signaling rapid mainstream adoption. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Real estate search platforms have deeply integrated automated listing/matching tools for over a decade, with near-universal consumer and agent adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI property-list generation substantially amplifies agent productivity by instantly filtering inventory to buyer specs, freeing agents to focus on relationship-building, negotiation, and nuanced guidance. The agent remains central while AI eliminates manual list-building drudgery. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-driven listing tools significantly speed up an agent's ability to identify and present relevant properties, though agents still add value through negotiation and personalized advice. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems can reliably filter and rank properties against structured criteria (price, location, features, square footage) using MLS data and buyer profiles, achieving substantial time savings over manual list curation. However, the final judgment of true compatibility—especially subtle lifestyle or investment fit—often requires human contextual knowledge, limiting full end-to-end automation to approximately 70–80% of the work. |
| Task automatability | claude-sonnet-5 | 5/5 | Filtering property databases by criteria like price, location, and features is a structured search/query task that current AI and even simple software already automate well, especially with MLS/API integration. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Real estate sales is competitive and lightly regulated; no legal requirement mandates human curation of property lists, and no significant liability barrier prevents AI automation of list generation. Customer preference for human guidance on final selection exists but does not prevent automated list generation. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or legal requirement mandates a human agent personally generate property lists; consumer-facing tools already do this without an agent involved. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | An AI system integrated into an MLS platform can generate filtered lists for thousands of buyers at near-zero marginal cost per list, whereas a human agent manually curating lists incurs full wage cost. The AI cost is orders of magnitude lower per equivalent output. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated search and filtering costs are trivial compared to an agent's hourly time manually compiling listings, making AI dramatically cheaper per query. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Production-grade real estate platforms (Zillow, Redfin, MLS integrations with AI filtering) routinely generate curated property lists based on buyer parameters at scale. These systems perform reliably in deployed settings, though some edge cases and non-standard buyer needs still require manual refinement. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Real estate platforms (Zillow, Redfin, MLS-integrated CRMs) already generate compatible property lists automatically at scale for millions of users today. |
Compare a property with similar properties that have recently sold to determine its competitive market price.
67CI 59–75 · exposure 67 · augmentation 100 · importance 4.6/5 · click for rater detail
Compare a property with similar properties that have recently sold to determine its competitive market price.
67| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Real estate technology adoption is rapid and broad; MLS systems, Zillow, Redfin, and agency management platforms widely deploy automated comparables and valuation estimates. Most U.S. agents routinely use AI-generated price recommendations, though as decision support rather than replacement. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate has adopted valuation tools like Zestimate and CMA software widely, but full automation of pricing judgment remains a human-supervised pilot-to-production transition rather than complete replacement. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI dramatically augments agent productivity by instantly surfacing relevant comparables, price trends, and valuation ranges that would take hours to compile manually. Agents remain in the loop for final pricing strategy, but the task's research and analysis phase is substantially accelerated. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI-powered comp analysis tools substantially speed up and improve the accuracy of an agent's market pricing work, letting agents focus on client-specific judgment and negotiation strategy. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI can automate significant portions of this task—gathering comparable sales data, calculating price metrics, and generating baseline valuations—but the final pricing determination requires market judgment, local knowledge, and negotiation context that current systems handle inconsistently. This reaches roughly 50% time-saving potential with AI-assisted workflows. |
| Task automatability | claude-sonnet-5 | 4/5 | Comparative market analysis relies on structured data (comps, sale prices, features) that AI tools can pull and analyze quickly, generating price recommendations with minimal human input beyond final judgment calls on qualitative factors.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While no legal barrier prevents AI from generating market analyses, real estate custom and liability concerns mean agents must review and sign off on valuations; lender appraisals still require licensed appraisers by regulation. These oversight and licensing requirements create moderate friction to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human perform the comparison itself, though agents typically must sign off on pricing recommendations delivered to clients under real estate practice norms. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-powered valuation tools cost a fraction of a human appraiser or experienced agent's labor per property analyzed. Integration costs are minimal for agents already using MLS systems. The cost per comparable analysis is likely 10–100× cheaper than human-delivered equivalent reports. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated CMA generation via software subscriptions costs a small fraction of the agent time it would take to manually research comps, though some oversight and local judgment still add cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Multiple deployed products (Zillow, Redfin, CoreLogic, MLS platforms) perform automated valuation models and comparable property analysis in production at scale. However, these tools are typically advisory rather than definitive; real estate professionals still validate and adjust recommendations, so reliability is high for data gathering but lower for final pricing calls. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Zillow, Redfin, and CMA-generating tools (e.g., within CRMs like kvCORE, Cloud CMA) already automate comp pulling and valuation estimates in production, though agents still refine and validate outputs. |
Prepare documents such as representation contracts, purchase agreements, closing statements, deeds, and leases.
62CI 62–62 · exposure 66 · augmentation 88 · importance 4.9/5 · click for rater detail
Prepare documents such as representation contracts, purchase agreements, closing statements, deeds, and leases.
62| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Real estate is moderately digitized with some automation pilots (e.g., e-signature platforms, template tools), but widespread deployment of AI document generation remains uneven; traditional forms and attorney involvement slow adoption in many markets. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate is a moderately digitized industry with widespread adoption of e-forms and transaction software, but full AI-driven contract drafting is still emerging rather than deeply embedded. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI draft generation substantially speeds document preparation for agents and enables rapid iteration on templates and clause variations, allowing humans to focus on negotiation and customization rather than form-filling. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI and template-based software dramatically speed up drafting and reduce errors in routine paperwork, letting agents focus on negotiation and client relations while staying in the loop for final review. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can generate most document templates, fill in standard fields, and produce first drafts of contracts and closing statements with high accuracy using existing forms and data. However, final review and customization for jurisdiction-specific nuances and negotiated terms typically require human judgment, preventing full end-to-end automation at the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | Document drafting from templates using standardized fields (party names, price, dates, property details) is well-suited to current AI/document-automation systems, though final agreements still need review for jurisdiction-specific and negotiated terms. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate documents must comply with state and local laws, and some jurisdictions require attorney review or agent sign-off, creating regulatory friction. Liability concerns around document errors and non-standard clauses create organizational reluctance despite technical capability. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Many jurisdictions require licensed agents or attorneys to prepare or review certain real estate contracts and deeds, creating moderate legal/liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Document generation by AI costs pennies to dollars per document versus several hundred dollars in human attorney or agent time for the same output; the cost difference is substantial and measurable. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Software-based document generation is very cheap per transaction compared to agent time spent manually drafting paperwork, though some oversight cost remains. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed legal document automation tools (e.g., LawGeex, Zuva, general LLMs) can draft and populate standard real estate documents reliably, but error rates remain material in complex or non-standard transactions, and production use remains inconsistent across agencies. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Transaction management platforms (e.g., dotloop, DocuSign, zipForm) already auto-populate many standard forms, but full contract generation with legal accuracy across jurisdictions still requires agent/broker or attorney review, so it's not fully autonomous in production. |
Coordinate appointments to show homes to prospective buyers.
62CI 46–79 · exposure 55 · augmentation 75 · importance 4.3/5 · click for rater detail
Coordinate appointments to show homes to prospective buyers.
62| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Real estate is a digitized, information-heavy sector with high competitive pressure. CRM platforms and scheduling automation are already widely adopted in brokerage firms and agencies, with active integration of AI scheduling assistants in production environments. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate has moderate digitization; scheduling tools and CRMs with AI features are increasingly used, but adoption varies widely by brokerage size and agent tech-savviness. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assistants augment agents by handling routine scheduling, follow-up reminders, and availability suggestions, freeing agents to focus on relationship-building and property knowledge. Agents remain in the loop for complex negotiations or special requests, boosting their overall productivity. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI scheduling assistants and CRM automation significantly reduce back-and-forth communication burden, letting agents focus on higher-value client interactions. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Calendar scheduling, contact management, and appointment logistics can be largely automated via AI agents that integrate with email, calendar systems, and CRM platforms. However, negotiating buyer preferences, handling last-minute changes, and managing complex multi-party coordination still require some human judgment, preventing a full 5 rating. |
| Task automatability | claude-sonnet-5 | 2/5 | AI scheduling tools can coordinate calendars and send reminders, but negotiating buyer/seller availability, seller access, and lockbox logistics still requires human judgment and relationship management.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few hard regulatory or legal barriers exist; appointments are not legally restricted to human schedulers. The main friction is agent preference for human touch and fear of losing customer relationship control, but these are soft organizational barriers, not compliance-level constraints. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically for scheduling showings, though agents remain the primary point of contact for security and liability reasons around home access. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI scheduling costs per appointment (API fees, inference) are orders of magnitude cheaper than the loaded cost of a human receptionist or agent time spent on calendar coordination. The ROI is strongly in favor of automation for routine bookings. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI scheduling tools are cheap and widely available, but agents still spend time on exceptions and personal touches, so net cost savings are moderate rather than order-of-magnitude. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products (calendar AI assistants, real estate CRM scheduling tools, chatbots) reliably handle appointment booking at scale in production. Minor limitations exist around complex edge cases and regional variations, but the core task is performable by mature systems today. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Scheduling assistants (e.g., AI calendar bots, CRM automation) are deployed in real estate practice today for appointment coordination, though full end-to-end handling including access permissions still often needs agent involvement. |
Evaluate mortgage options to help clients obtain financing at the best prevailing rates and terms.
58CI 54–62 · exposure 58 · augmentation 75 · importance 3.5/5 · click for rater detail
Evaluate mortgage options to help clients obtain financing at the best prevailing rates and terms.
58| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Mortgage technology adoption is active in fintech and digital lenders, but traditional real estate brokerage remains moderately digitized; pilots of AI rate-shopping tools exist, but broad production displacement of this advisory task in agencies is still developing. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate and mortgage industries have adopted AI-driven rate comparison and pre-qualification tools moderately, with fintech disruption ongoing but many agents still relying on traditional lender relationships. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI mortgage comparison tools significantly assist agents by instantly surfacing rate options, terms, and eligibility scenarios that would take hours to gather manually, allowing agents to focus on client fit and negotiation rather than data collection. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools significantly speed up gathering and comparing mortgage rates and terms, letting agents present more options faster and with better data-backed insights to clients. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI can rapidly compare mortgage rates, terms, and eligibility across multiple lenders using current market data and client financial information, delivering significant time savings. However, the task still requires human judgment on client risk tolerance, life circumstances, and final recommendation, preventing a full end-to-end automation rating. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can quickly compile and compare mortgage rates, terms, and eligibility scenarios, but tailoring advice to a client's specific financial situation and negotiating with lenders still requires human judgment and relationship management.tests.rrationale placeholder |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate agents are not required by law to perform mortgage evaluation, but regulatory oversight of loan origination and disclosure requirements (TRID, NMLS) means formal loan officers and compliance sign-off remain necessary. Client preference for human guidance and state licensing variations create moderate friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While agents aren't licensed loan officers, discussing financing terms can brush against mortgage broker/lending regulations in some jurisdictions, and clients often expect a trusted human intermediary for major financial decisions. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Mortgage comparison APIs and AI tools are relatively cheap to integrate and run, with minimal per-transaction costs compared to the human labor time spent manually gathering quotes and analyzing options. The cost advantage is substantial for routine cases. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated rate-comparison and pre-qualification tools cost very little per use compared to the time an agent would spend manually researching lender options, though oversight and verification still add some cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Multiple fintech products and lender platforms offer automated mortgage comparison and rate-shopping tools, but they operate within limited lender networks and often require human review for complex cases or non-standard situations. Production deployments exist but with meaningful scope limitations. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Mortgage comparison tools and calculators are widely deployed and reliable for basic rate/term comparisons, but comprehensive personalized financing advice combining creditworthiness, lender relationships, and negotiation is not yet handled end-to-end by deployed products. |
Review property listings, trade journals, and relevant literature, and attend conventions, seminars, and staff and association meetings, to remain knowledgeable about real estate markets.
57CI 50–64 · exposure 42 · augmentation 75 · importance 4.3/5 · click for rater detail
Review property listings, trade journals, and relevant literature, and attend conventions, seminars, and staff and association meetings, to remain knowledgeable about real estate markets.
57| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Real estate is moderately digitized; brokerages increasingly use CRM systems and market analytics platforms, but adoption of AI-driven competitive intelligence specifically for this knowledge-maintenance task remains in pilot/early-adoption phases rather than widespread production use. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate is a moderately digitized sector with growing use of AI market tools, but broad reliance on traditional networking and events keeps overall adoption middling. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI excels at filtering, summarizing, and organizing vast volumes of listings and market data in real time, materially amplifying an agent's ability to stay current without reading full documents or attending every seminar, while the agent applies judgment to decisions. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools substantially help agents by aggregating and summarizing market data, trends, and news, saving significant research time even though in-person components remain. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Current AI can summarize property listings and literature efficiently, but market knowledge requires contextual judgment about local economic trends, buyer psychology, and competitive dynamics that demand human interpretation. The task mixes routine information gathering (automatable) with experiential synthesis (not fully automatable to the 50% threshold). |
| Task automatability | claude-sonnet-5 | 3/5 | AI can summarize listings, market reports, and journals efficiently, but attending conventions/seminars and networking-based knowledge acquisition cannot be automated end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing or legal requirement prevents using AI to assist with market research; real estate agents retain autonomy to use AI tools or human methods freely, though industry conventions and professional judgment expectations create light organizational friction. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or regulatory requirement mandates a human specifically perform market research or literature review; it's a personal professional development task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Information aggregation and summarization via AI costs a fraction of the human time spent reading journals and attending events; however, the human must still interpret and contextualize the curated summaries, limiting pure cost displacement. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | AI-driven market research and content summarization tools are much cheaper than an agent's time spent reading and attending events, though attendance costs remain human-borne. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | AI tools exist to summarize trade journals and parse property listings at scale, but no deployed system reliably synthesizes multi-source market intelligence into actionable agent knowledge without human curation and validation. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | AI research assistants and market summary tools are deployed and used by agents today, but they don't replace in-person events or nuanced local market synthesis reliably. |
Solicit and compile listings of available rental properties.
52CI 50–55 · exposure 50 · augmentation 75 · importance 3.0/5 · click for rater detail
Solicit and compile listings of available rental properties.
52| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Real estate has adopted CRM and listing aggregation tools widely, but agent-driven outreach and manual curation remain common; automation is present but unevenly deployed and often human-supervised. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate is moderately adopting AI tools for listing management and lead generation, but overall sector digitization and agent-level AI adoption remain uneven compared to fully digital-native industries. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-powered listing aggregation, filtering by criteria, and data enrichment (property details, neighborhood info) substantially boost an agent's ability to surface and organize available rentals quickly, maintaining human judgment on client fit and quality. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can significantly speed up compiling, organizing, and updating rental listings, and can assist with outreach templates and prospecting, meaningfully boosting agent productivity while the agent still manages relationships. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Web scraping and database aggregation of rental listings can be automated, but compilation typically requires human verification of property details, contact information accuracy, and filtering by local regulations—meaning roughly 40–60% of the work could be automated with setup. |
| Task automatability | claude-sonnet-5 | 3/5 | Gathering and compiling rental listing data can be substantially automated via scraping, CRM integration, and AI-driven data entry, but soliciting new listings requires relationship-building and persuasion that AI cannot fully replicate. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate markets often require local licensing for certain activities, MLS access restrictions, and client trust in agent selection—these create some friction, though property listing compilation itself is not legally restricted to licensed agents in most jurisdictions. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically blocks AI from compiling listings, though agents’ relationships with property owners create some organizational and trust-based friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | API costs and data aggregation infrastructure are moderate, but human verification and curation overhead remain significant; costs approach parity with a part-time rental coordinator rather than a full agent. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated data compilation is cheap, but the solicitation component still requires human agent time and relationship management, keeping blended costs roughly comparable to full manual work. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Tools exist to aggregate rental listings from multiple sources (APIs, web scraping), but real estate platforms require access control and consistent human oversight to ensure accuracy and legal compliance; deployed systems exist but typically support rather than fully replace the task. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like CRM automation tools, listing aggregators, and AI-assisted outreach exist and are used in production, but they handle compilation better than the solicitation/relationship aspect, so overall reliability is mixed. |
Contact previous clients for prospecting of referral business.
51CI 25–77 · exposure 50 · augmentation 88 · importance 4.4/5 · click for rater detail
Contact previous clients for prospecting of referral business.
51| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains largely small-firm and individually-agent-based, with slow digital transformation; while CRM adoption is common, active AI-driven prospecting agents are not yet standard practice in the sector. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate is a moderately digitized industry with growing CRM/AI marketing tool adoption, but many agents still rely on personal calls and relationships, limiting full automation uptake. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can effectively assist by drafting personalized outreach templates, scoring past clients by likelihood to refer, suggesting optimal contact times, and preparing client-specific talking points—all of which enhance an agent's efficiency while the agent retains the critical relationship and closing role. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI tools significantly help agents draft personalized outreach, segment past clients, and time follow-ups, greatly boosting referral prospecting productivity while agents retain relationship control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft personalized outreach messages and identify prospect lists, the task fundamentally requires relationship-building communication and human judgment about timing, tone, and existing client relationships—elements that cannot be reliably automated end-to-end with 50% time savings at equal quality today. |
| Task automatability | claude-sonnet-5 | 4/5 | Outreach messages, follow-up emails, and CRM-driven referral campaigns can be largely automated with AI-generated personalized communications and scheduling, though final personal touch is often added by the agent. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Real estate sales is relationship-driven with informal personal networks; clients often expect human contact from their known agent, and conversion depends on trust and rapport that clients specifically value—creating both regulatory (fiduciary duty to serve clients personally) and market friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or regulatory requirement mandates a human perform client outreach; marketing communication is not tightly restricted for real estate agents. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for CRM and email automation cost $50–200/month, but a single agent's network contact work is relatively low-cost per unit (a few minutes per call/email); the AI infrastructure does not yield a cost advantage sufficient to replace the agent's direct relationship engagement. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated email/SMS campaigns cost a small fraction of an agent's time spent personally calling or messaging past clients, though some human oversight and personalization still add cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some CRM systems integrate AI-assisted message drafting and scheduling, but no deployed product reliably performs genuine prospecting outreach to previous clients at scale; most tools are limited to templated follow-ups or require heavy human oversight to maintain relationship authenticity. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | CRM tools with AI-driven email/SMS drip campaigns and personalized outreach are widely deployed in real estate today (e.g., Follow Up Boss, kvCORE) and reliably generate and send referral-prospecting messages. |
Promote sales of properties through advertisements, open houses, and participation in multiple listing services.
51CI 46–55 · exposure 50 · augmentation 75 · importance 4.6/5 · click for rater detail
Promote sales of properties through advertisements, open houses, and participation in multiple listing services.
51| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Larger brokerages and platforms (Zillow, Redfin) have adopted AI-assisted listing and ad tools, but most independent and small-firm agents still rely on manual methods, placing adoption in the pilot-to-early-production range rather than deep, universal deployment. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate is a moderately digitized sector with growing use of AI marketing tools and listing platforms, but broad, deep agent-level adoption is still uneven and pilot-stage in many brokerages. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-generated property descriptions, market analysis, targeted ad suggestions, and virtual tour automation substantially boost agent productivity in promotion tasks, allowing agents to focus on client relationships and negotiations while AI handles routine content creation and scheduling. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI substantially speeds up drafting listings, creating marketing materials, and optimizing ad targeting, meaningfully boosting agent productivity while the agent still manages relationships and open houses. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI can automate property listing creation, ad generation, and scheduling open houses with significant time savings, but the task requires human judgment on market positioning, negotiation strategy, and client relationship building that AI cannot fully replicate today. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can generate ad copy, social posts, and listing descriptions and automate MLS syndication, but open houses and in-person promotion require human presence, capping full end-to-end automation.: |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate sales remain relationship-dependent and often require state licensing for agents; however, the promotional elements (ads, listings, open house logistics) face lighter regulatory barriers than the sales transaction itself, creating modest friction but not hard legal blocks. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement blocks AI-generated marketing content, though fair housing advertising rules and MLS compliance create some regulatory friction around AI-generated claims. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | While AI reduces costs for ad generation and listing management, the integrated oversight, customization, and human verification needed means total cost per sale remains comparable to or only marginally cheaper than traditional agent effort. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI content generation and ad optimization tools are cheap relative to agent time, but physical promotion (open houses) still requires paid human labor, keeping overall cost comparable rather than order-of-magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Products exist for automated listing descriptions, property ads, and virtual tours, but deployed systems require substantial human oversight for quality and brand consistency, and no single product reliably handles the full promotional workflow end-to-end. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Products like AI listing description generators, virtual staging, and automated MLS syndication tools exist and are used, but adoption is partial and quality/accuracy still requires agent review. |
Answer clients' questions regarding construction work, financing, maintenance, repairs, and appraisals.
50CI 39–61 · exposure 38 · augmentation 75 · importance 4.3/5 · click for rater detail
Answer clients' questions regarding construction work, financing, maintenance, repairs, and appraisals.
50| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Real estate firms, particularly brokerages and large agencies, operate in a digitized information-heavy sector with strong profit incentives and are already deploying chatbots and AI-assisted CRM tools for client communication. Production adoption is visible in mid-to-large firms, with clear momentum toward automation of routine Q&A. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a moderately digitized but relationship-driven, fragmented small-business sector where AI adoption for direct client advising remains in early pilot stages. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly augment a salesperson's productivity by rapidly drafting answers to common questions, pulling appraisal data, summarizing financing options, and flagging maintenance or repair considerations, allowing the agent to focus on relationship-building and complex negotiations. The human remains in the loop for final client interaction and judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can quickly draft answers, summarize inspection/appraisal reports, and provide talking points, meaningfully speeding up agents' ability to respond to client questions while they verify accuracy. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI can handle routine questions about financing, maintenance, and general construction via Q&A systems and can retrieve appraisal information reliably. However, nuanced client inquiries often require contextual knowledge of local market conditions, specific property histories, and personalized loan scenarios that demand human judgment, limiting full end-to-end automation to roughly 50% of interactions. |
| Task automatability | claude-sonnet-5 | 2/5 | AI chatbots can answer general questions about financing or maintenance, but nuanced, client-specific advice on construction, repairs, and appraisals often requires local knowledge and judgment that current systems can't fully replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing requirement mandates a human answer client questions, and no hard regulatory barrier prevents AI chatbots from handling routine inquiries. Customer preference for human contact and organizational friction around trust in AI answers introduce moderate friction, but these are soft barriers rather than legal prohibitions. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement to answer general questions, though agents may have some liability exposure for misinformation on financing or structural issues, creating moderate caution around fully automating advice-giving. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference and simple integration for question-answering is substantially cheaper than paying a salesperson to spend time researching and responding to routine client inquiries; oversight is minimal for straightforward Q&A. The cost per question answered is likely 10–100× lower than human labor, though integration and fallback mechanisms add overhead. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI can cheaply handle routine informational queries, but complex or liability-sensitive questions still require agent review, keeping blended costs roughly comparable to human-only handling for this specific task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Chatbots and AI assistants deployed in real estate today can answer factual questions about financing and maintenance with reasonable accuracy, but errors in jurisdiction-specific regulations, appraisal interpretation, and construction details remain common. Mature production systems exist but require significant human oversight and cannot reliably handle edge cases or novel scenarios without agent intervention. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Real estate chatbots and AI assistants exist and can handle FAQ-style queries, but reliable, accurate handling of technical construction/appraisal questions in production is limited and error-prone. |
Contact utility companies for service hookups to clients' property.
47CI 30–65 · exposure 45 · augmentation 63 · importance 3.1/5 · click for rater detail
Contact utility companies for service hookups to clients' property.
47| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate brokerage remains moderately digitized with legacy workflows; while CRM adoption is common, end-to-end automation of utility coordination is not yet a focus area for major platforms or agents, so actual production deployment is minimal. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a relationship-driven, moderately digitized sector where such administrative automation is emerging but not yet widely deployed at scale for this specific sub-task. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by drafting utility contact templates, extracting and organizing property details for submission, tracking service status, and reminding agents of deadlines—all useful support functions that raise agent efficiency without fully automating human negotiation and verification. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI assistants can already draft utility contact lists, auto-fill forms, or place calls via voice agents, meaningfully reducing agent time spent on this coordination task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | The task involves phone or email contact with utility companies to arrange service hookups, which requires navigating variable workflows, explaining client details, and coordinating timing. While AI could draft communications or prepare information, completing the full task end-to-end with 50% time savings would require reliable automated interaction with proprietary utility systems that vary by provider—currently undeployed at scale. |
| Task automatability | claude-sonnet-5 | 4/5 | This is a routine, scripted communication task (calling/emailing utility providers, providing address and account details) that AI voice/chat agents can largely handle with minimal judgment required. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Utilities require verified account ownership and identity verification before service activation, and some providers prefer or require direct human contact from the property owner or a licensed agent. There is customer preference and procedural friction, but no strict legal barrier preventing AI-initiated outreach. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement dictates a human must do this; main friction is utility companies' own phone systems and occasional need for account verification tied to a real person. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | An AI solution would need to integrate with multiple utility company systems, maintain updated contact methods and requirements, and handle exceptions; the integration and oversight overhead likely exceeds the cost of a human agent making a few phone calls or emails. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | An AI calling/automation agent costs a small fraction of an agent's or assistant's hourly time for a simple coordination task like this. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No mature product reliably automates utility service hookup coordination across multiple utility companies with their different intake processes, documentation requirements, and verification rules. Narrow pilot systems may exist, but production deployment of fully autonomous utility contact and hookup scheduling is not standard in the industry. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | AI phone agents and scheduling bots exist and are deployed in some real estate and property management workflows, but utility companies' own IVR/human systems and inconsistent APIs create friction and error rates. |
Investigate clients' financial and credit status to determine eligibility for financing.
47CI 40–54 · exposure 42 · augmentation 75 · importance 3.6/5 · click for rater detail
Investigate clients' financial and credit status to determine eligibility for financing.
47| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Real estate is moderately digitized; some brokers use automated pre-qualification tools and credit integrations, but adoption is uneven and many agents still perform manual financial vetting; not yet the standard like in pure lending sectors. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Mortgage/lending sector has adopted automated underwriting significantly, but real estate agents as an occupation show slower, uneven AI adoption for this specific credit-investigation function since it's often outsourced to lenders. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can substantially assist agents by automating credit pulls, flagging documents, and summarizing financial profiles, allowing agents to focus on client communication and approval decisions rather than data hunting and basic analysis. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered credit check tools and financial pre-screening platforms meaningfully speed up an agent's ability to assess client eligibility and refer them appropriately, even though final financing decisions remain with lenders. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist with data retrieval and basic credit scoring from public records and APIs, but the task requires judgment about client eligibility, interpretation of complex financial situations, and discrete authorization decisions that currently demand human oversight and discretion. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can pull credit reports, verify income documents, and flag eligibility issues, but nuanced judgment about borderline cases and communicating with lenders still requires human involvement.4 A significant fraction can be automated but not the full workflow end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Fair lending regulations, equal credit opportunity laws, and lender/broker liability create compliance oversight requirements, though no explicit licensing bars automation; agents remain responsible for ensuring fair and accurate eligibility determinations, limiting full substitution. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Financial eligibility determination is heavily regulated (fair lending laws, credit reporting regulations) and typically falls under licensed loan officers/underwriters, not real estate agents directly, creating strong institutional and legal barriers to full automation by the agent. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-driven credit and financial data services are inexpensive compared to the time an agent would spend gathering and manually analyzing financial documents; the per-task cost is well below typical agent hourly loaded cost. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated credit and financial verification services are cheap relative to manual review, especially compared to an agent's time, though integration and compliance oversight add some cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Credit-checking and financial screening tools exist and are deployed in finance, but in real estate sales they are typically human-supervised; automated systems handle data collection but agents must review and make final eligibility determinations due to liability and nuance. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Loan origination and mortgage pre-qualification software with automated underwriting (e.g., Fannie Mae's Desktop Underwriter) is deployed at scale in lending, but real estate agents themselves typically rely on third-party lenders/brokers rather than performing this analysis with AI tools directly. |
Secure construction or purchase financing with own firm or mortgage company.
41CI 25–57 · exposure 45 · augmentation 75 · importance 2.6/5 · click for rater detail
Secure construction or purchase financing with own firm or mortgage company.
41| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Mortgage and fintech sectors are actively piloting AI-driven underwriting and document automation, but production deployment remains concentrated in larger lenders and technology-forward firms. Mid-market and smaller real estate operations still rely heavily on manual processes, indicating middling overall velocity. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate and mortgage industries have been slower to adopt AI agents for transactional financing tasks compared to sectors like finance-tech or professional services, with pilots more common than production use. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI substantially assists loan officers and real estate agents by automating document collection, initial compliance screening, and data entry, freeing them to focus on borrower consultation and exception resolution. This augmentation meaningfully raises productivity on routine transactions while keeping humans in control of final decisions. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can meaningfully assist agents by pre-screening loan options, automating document collection, and flagging financing issues, improving efficiency while humans retain the client-facing negotiation role. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Securing financing involves document review, underwriting workflows, compliance checks, and communication tasks that AI systems can substantially automate. Current AI can extract data from applications, validate requirements against lending guidelines, flag issues, and generate initial financing summaries; however, final approval decisions and complex exception handling typically require human judgment, limiting full end-to-end automation to ~70–80% time saving. |
| Task automatability | claude-sonnet-5 | 2/5 | This task involves relationship-building, negotiation with lenders, and navigating client-specific financial situations that require judgment and interpersonal coordination beyond current AI capabilities to fully replace.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong regulatory barriers apply: mortgage lending is heavily regulated by federal (CFPB, OCC) and state authorities, and final loan decisions must be made or signed off by a licensed loan officer or underwriter. Compliance risk and liability create hard friction against full automation without human authorization. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Mortgage origination is heavily regulated (licensing, disclosures, fair lending laws), requiring licensed professionals to structure and approve financing, creating strong legal barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-driven document processing and preliminary underwriting screening cost a fraction of the labor required for manual application review and initial compliance checks. At scale, AI inference plus lightweight oversight is substantially cheaper than the loaded cost of a mortgage specialist or loan officer managing the same volume. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Human loan officers and agents still dominate this process; AI can cut some paperwork costs but the overall financing arrangement still requires human-mediated services, keeping cost savings modest. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Several fintech and mortgage automation platforms deploy AI for document processing and initial underwriting assessment in production, but adoption remains unevenly distributed and error rates on edge cases remain material. Mature systems exist in larger mortgage firms, but the broader real estate market still relies on significant manual review. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tools exist for mortgage pre-qualification and document processing, but no deployed product autonomously secures financing arrangements end-to-end for a real estate transaction. |
Arrange for title searches to determine whether clients have clear property titles.
37CI 25–50 · exposure 38 · augmentation 63 · importance 3.9/5 · click for rater detail
Arrange for title searches to determine whether clients have clear property titles.
37| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains relatively traditional in its back-office operations; while some title companies have digitized workflows, the core task of determining clear title remains bound to licensed professionals and has not seen rapid AI displacement. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate transaction management software adoption is moderate, with many brokerages using digital transaction platforms, though full automation of title arrangement is not yet universal. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist title professionals by automating document gathering, flagging potential defects, and organizing search results, meaningfully speeding up the examination process while the licensed professional retains judgment on clearance. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enabled transaction management tools significantly streamline ordering, tracking, and follow-up on title searches, freeing agents to focus on client-facing work. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Title search initiation and basic document retrieval can be partially automated, but determining title clearance requires legal interpretation, judgment about exceptions, and resolution of competing claims—tasks where current AI lacks reliable decision-making authority and still requires human legal oversight. |
| Task automatability | claude-sonnet-5 | 3/5 | The coordination and ordering of title searches (contacting title companies, submitting requests, tracking status) can be automated via workflow tools, but interpreting title issues and client communication still requires human involvement., though the core task is largely administrative and routing-based. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Title examination is heavily regulated; most jurisdictions require a licensed title agent or attorney to conduct and sign off on title searches, creating a hard legal barrier to full automation of clearance determination. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement prevents automating the arrangement itself, but title searches must ultimately be performed by licensed title companies/attorneys in many states, creating institutional friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Professional title searches and examination still require licensed title agents or attorneys; AI might reduce document processing costs modestly, but the core liability-bearing work remains human-dominated and labor costs remain significant relative to AI overhead. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated ordering systems reduce agent time modestly, but title search itself still incurs third-party fees and human review costs comparable to current processes, limiting large cost savings. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Title search services exist and some legal tech platforms offer document automation, but actual clearance determination and liability for missing defects remains a legal function performed by title companies and attorneys, not fully delegated to AI systems in production. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Title companies and real estate platforms already use software to order and track title searches automatically, but agents still manually initiate and manage the process with title companies in most transactions today. |
Interview clients to determine what kinds of properties they are seeking.
36CI 34–39 · exposure 25 · augmentation 75 · importance 4.4/5 · click for rater detail
Interview clients to determine what kinds of properties they are seeking.
36| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains a relationship-driven, human-centric sector with slow AI adoption; while some brokerages pilot chatbots for lead qualification, most production workflows still rely on agent interviews and human judgment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a traditionally low-digitization, relationship-driven sector where AI adoption for client interviews remains in early pilot stages, not widespread production use. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully augment agents by suggesting follow-up questions, summarizing client preferences in real time, flagging likely property matches, and automating notes—substantially boosting agent productivity while the human agent remains in control of the interaction and relationship-building. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can help agents prepare questions, summarize client communications, and organize preferences from forms/emails, meaningfully boosting agent efficiency while the agent still leads the interview. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can conduct structured interviews via chatbots and gather basic property preferences, the task requires building rapport, reading nonverbal cues, understanding nuanced client needs, and adapting questions dynamically—capabilities where current AI falls far short of delivering 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | While chatbots can gather basic preferences (budget, location, bedrooms), truly interviewing clients to uncover nuanced needs, build rapport, and interpret unstated priorities remains largely a human relational task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate firms face moderate friction: client preference for human interaction, liability concerns if AI misunderstands needs, and the need for human judgment to qualify and nurture leads, though no formal licensing barrier applies specifically to the interview task itself. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human specifically for this conversational step, though client trust and relationship-building customs create moderate friction against pure automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | Chatbot infrastructure and oversight are modestly less costly than a human agent's fully loaded wage, but the gap is narrow given the need for human followup and the risk of losing deals due to poor information gathering. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | A chatbot-based intake form is cheap to run, but since it only partially substitutes for the task, the cost comparison is muddied; combined human+AI intake is roughly comparable in cost to agent time saved. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Deployed chatbots can collect basic information, but production systems lack the conversational depth and judgment needed for effective client discovery; most real estate agents still rely on human-led interviews because AI solutions produce material gaps in understanding and client satisfaction. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some real estate platforms use intake forms or basic chat-based questionnaires, but no deployed product reliably conducts a full client-needs interview replacing agent judgment and trust-building. |
Advise sellers on how to make homes more appealing to potential buyers.
34CI 30–39 · exposure 25 · augmentation 63 · importance 4.3/5 · click for rater detail
Advise sellers on how to make homes more appealing to potential buyers.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While real estate is digitizing rapidly, actual deployment of AI for seller advisory remains limited to early pilots and supplementary tools; most agents still rely on experience and human judgment, with slow organizational shift toward AI integration. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate remains a relationship-driven, moderately digitized sector where AI tools are used for marketing and photos but advisory tasks still lean heavily on human agents. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist by surfacing comparable home features, suggesting staging ideas based on market data, and identifying repair priorities, thereby accelerating an agent's advisory process while the human maintains final judgment and client relationship. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools (virtual staging, market comps, generative design suggestions) meaningfully help agents formulate and visualize appealing improvements, enhancing their advisory capability. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can analyze market trends and generate generic staging advice, advising sellers on home appeal requires understanding individual property nuances, local market subtleties, and subjective aesthetic judgment that current systems cannot reliably deliver end-to-end at the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate generic staging and repair suggestions from photos or descriptions, but effective advice depends on local market knowledge, in-person walkthroughs, and buyer psychology that current systems can't fully replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate sales is lightly regulated at the task level; however, market custom and client preference for human relationship-building, combined with agent commission structures and licensing requirements for some advisory functions, create moderate friction to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically covers this advisory task, though sellers strongly prefer trusted human relationships and localized judgment, creating moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | The integrated cost of AI advisory systems (including data gathering, model inference, and human oversight of recommendations) remains comparable to or exceeds the cost of a brief human consultation with a real estate agent. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-generated generic advice is cheap, but since it can't fully substitute for the human judgment needed, sellers still pay for agent expertise, making cost comparison roughly comparable when quality is held constant. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some AI tools exist for basic home staging suggestions and property analysis, but deployed products lack the contextual sophistication and reliability needed to replace human advisors; most remain proof-of-concept or narrow-scope applications rather than production systems at scale. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some staging/renovation AI tools and chatbots offer generic tips, but no deployed product reliably replaces an agent's in-person, contextualized advisory role at scale. |
Conduct seminars and training sessions for sales agents to improve sales techniques.
34CI 30–38 · exposure 25 · augmentation 75 · importance 2.9/5 · click for rater detail
Conduct seminars and training sessions for sales agents to improve sales techniques.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Real estate firms are adopting AI for content generation and asynchronous learning modules at a moderate pace, but live seminar automation remains limited. Most adoption falls into hybrid models (AI-drafted content delivered by human instructors) rather than full AI-led seminars. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a moderately digitized sector with slow, uneven AI adoption in training functions, though e-learning and AI content tools are gradually being piloted. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can significantly augment human trainers by generating personalized training materials, analyzing agent performance data to highlight improvement areas, and creating interactive practice scenarios or role-play simulations. This enhances trainer productivity and customization while keeping the human instructor central to live delivery. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can significantly enhance training preparation by generating scripts, role-play scenarios, presentation decks, and personalized coaching materials, boosting trainer productivity substantially. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can generate training content and slides, but conducting live seminars requires real-time interaction, addressing audience questions, adapting tone and pacing—skills that depend heavily on human judgment and presence. Current systems cannot replicate the full interactive facilitation at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate training content and materials, but delivering live seminars and interactive training sessions requires human presence, adaptability, and interpersonal engagement that current AI cannot fully replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Sales teams and organizations often prefer human instructors for live training due to cultural fit, trust, and the assumption that experienced agents deliver credibility. However, there are no hard legal barriers to using AI-assisted or AI-led training, only organizational preference and custom friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human trainer, but organizational culture and preference for live, relationship-based coaching in sales training creates moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Creating comprehensive training content via AI is cheap, but integrating it into live seminars still requires human facilitation or sophisticated interactive AI systems. The all-in cost (content generation plus live interaction oversight) remains comparable to hiring a human trainer for ongoing sessions. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cheaply assist in content creation, but the actual delivery of training still requires a human trainer, so overall cost savings are limited unless training shifts fully to AI-generated video/asynchronous formats. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI can draft training materials and scripts, no deployed product reliably conducts live seminars end-to-end. Some organizations use AI for asynchronous training content, but interactive facilitation by AI remains in pilot stages with significant limitations in audience engagement and adaptive response. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | AI tools exist for creating presentations, scripts, and e-learning modules, but no deployed product independently conducts live sales training seminars for real estate agents at scale. |
Arrange meetings between buyers and sellers when details of transactions need to be negotiated.
33CI 16–50 · exposure 17 · augmentation 63 · importance 2.8/5 · click for rater detail
Arrange meetings between buyers and sellers when details of transactions need to be negotiated.
33| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Real estate technology (CRM, scheduling integration) is moderately adopted, with many agents using productivity tools, but full meeting-arrangement automation remains uncommon in practice. Pilot projects are more common than widespread production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a relationship-driven, moderately digitized sector where AI adoption for scheduling exists but negotiation facilitation is still handled by humans; overall sector adoption is slow to moderate. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI scheduling assistants and CRM tools already assist agents by proposing time slots, flagging conflicts, and drafting meeting summaries, materially raising their throughput. Agents remain in the loop to approve and adjust based on negotiation strategy, making this a strong augmentation scenario. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help by suggesting meeting times, drafting communications, and summarizing negotiation points, meaningfully aiding the agent's workflow around the meeting. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Scheduling meetings can be partially automated (calendar coordination, time-slot proposals), but the task requires understanding negotiation stakes, buyer/seller preferences, and responding to counteroffers—elements that currently demand human judgment. AI cannot reliably handle the interpersonal negotiation dynamics that determine when and how meetings should occur. |
| Task automatability | claude-sonnet-5 | 1/5 | Arranging and facilitating in-person negotiation meetings between two parties requires relationship management, trust-building, and real-time human judgment that current AI cannot perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No legal or licensing requirement mandates that a licensed agent must personally arrange a meeting; however, real estate industry norms and customer expectations for human relationship-building create moderate friction. Clients often prefer agents to handle these communications directly. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement to schedule a meeting, but real estate agents often have fiduciary duties and clients expect personal representation during negotiations, creating moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automated scheduling systems (including integration with CRM and email) cost far less than a real estate agent's fully-loaded hourly rate. However, the task still requires human oversight to validate meeting appropriateness and handle exceptions, preventing the highest rating. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI scheduling assistants are cheap for calendar coordination, but the negotiation-facilitation portion still requires an agent's paid time, so overall cost savings are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Calendar-scheduling AI and chatbots exist in production (e.g., Calendly, assistant bots), but they typically handle straightforward logistics, not the complex negotiation sequencing required in real estate transactions. Products deployed today lack reliable understanding of negotiation context and deal-specific constraints. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product independently arranges and conducts negotiation meetings between buyers and sellers; scheduling tools exist but the substantive negotiation coordination remains human-led. |
Review plans for new construction with clients, enumerating and recommending available options and features.
33CI 30–35 · exposure 25 · augmentation 75 · importance 3.5/5 · click for rater detail
Review plans for new construction with clients, enumerating and recommending available options and features.
33| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains a relationship-heavy, human-centric sector with slow digital adoption; most agencies still rely on human agents for client consultations, and automation pilots remain rare in this specific task. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a traditionally low-digitization, relationship-driven sector where AI tools are used for marketing and lead-gen more than for client-facing plan consultations, so adoption in this specific task is slow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can assist significantly by auto-generating plan summaries, comparing option features side-by-side, and flagging popular upgrades or conflicts, enabling agents to brief clients faster and more comprehensively while the agent maintains the advisory role. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can help agents quickly summarize floor plans, generate visualizations, compare options, and prepare talking points, meaningfully boosting the agent's efficiency and knowledge during client interactions. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can enumerate construction options and features from technical documents or databases, but the task critically requires client dialogue to understand preferences, constraints, and trade-offs—a consultative judgment component that AI cannot reliably perform end-to-end today without substantial human direction and override. |
| Task automatability | claude-sonnet-5 | 2/5 | This task requires reading physical/digital blueprints, understanding client preferences in real time, and providing nuanced spatial and aesthetic judgment during in-person or interactive walkthroughs, which current AI cannot fully replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | There is some friction from customer preference for human relationship-building during major purchasing decisions (real estate), and agents are typically responsible for legal accuracy of option recommendations, creating liability concerns that inhibit full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human for this specific plan-review conversation, but client trust, relationship-based sales dynamics, and customization needs create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | An AI system capable of independent plan review and recommendation would require significant customization per development project, ongoing oversight, and integration with CRM systems, making per-task cost comparable to or higher than a sales agent's fractional time on this activity. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI could cheaply generate options summaries, but the human relational and advisory component still requires an agent's time, so overall cost savings versus the human-led interaction are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While AI systems can retrieve and summarize construction plans and options (e.g., via document processing), no deployed product reliably performs the full consultative review—matching client needs to options—without human sales agent involvement and error correction. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some tools can generate feature lists or renderings from plans, but no deployed product reliably conducts the full interactive client consultation and recommendation process in production real estate settings. |
Contact property owners and advertise services to solicit property sales listings.
32CI 32–32 · exposure 25 · augmentation 75 · importance 4.3/5 · click for rater detail
Contact property owners and advertise services to solicit property sales listings.
32| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Real estate agents have begun adopting AI-powered CRM tools, lead-scoring systems, and message drafting; however, production-level autonomous prospecting remains limited. Pilot adoption is common, but agents still perform the critical relationship work themselves. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate is adopting AI tools for marketing, lead generation, and CRM automation at a moderate pace, though core sales/listing solicitation remains largely human-driven. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI substantially assists this task by generating personalized prospecting lists, drafting tailored outreach messages, scheduling follow-ups, and identifying high-probability leads. Agents using AI tools significantly increase contact volume and speed while managing the relationship and closing work themselves. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI significantly boosts productivity by generating property descriptions, targeted ad copy, personalized outreach emails, and lead scoring, letting agents focus more time on relationship-building and closing. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft outreach messages and identify contact lists, the personalized persuasion needed to solicit listings—requiring negotiation, relationship-building, and trust—remains largely dependent on human judgment and rapport. Current AI systems cannot reliably close listing agreements at scale without human-led follow-up and negotiation. |
| Task automatability | claude-sonnet-5 | 2/5 | Generating outreach messages or ads can be AI-assisted, but the actual solicitation, persuasion, relationship-building, and closing of listing agreements requires human judgment, trust-building, and negotiation that AI cannot fully replace end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate sales operates in a competitive, unregulated prospecting environment with no legal barrier to automation of contact and messaging. However, consumer expectations for human trust and relationship-building, plus brokers' preference for agent accountability, create organizational friction to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement blocks AI from drafting marketing content, but real estate transactions involve licensed agent representation, fiduciary duties, and strong client preference for personal trust, creating moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | While AI-powered outreach tools reduce some labor, the sales agent's compensation is heavily commission-tied, and AI cannot yet replace the high-touch relationship work that generates listings. Integration and oversight costs offset marginal labor savings on the prospecting phase alone. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cheaply draft ads or cold-outreach templates, but human agents still must do calls, meetings, and relationship management, so overall cost savings are partial rather than a full substitute. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI can assist with message generation and contact identification, but deployed products lack the ability to autonomously conduct the full prospecting conversation including objection handling and contract negotiation. Most real estate CRM systems offer AI-aided communication drafting, not autonomous solicitation. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | CRM tools and AI-generated marketing copy exist and are used by agents, but no deployed product autonomously contacts owners, pitches services, and secures listings reliably at scale. |
Display commercial, industrial, agricultural, and residential properties to clients and explain their features.
31CI 25–37 · exposure 30 · augmentation 75 · importance 4.1/5 · click for rater detail
Display commercial, industrial, agricultural, and residential properties to clients and explain their features.
31| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While some real estate firms use virtual tours and online property platforms, the core task of displaying and explaining properties to clients in real-time remains human-driven in most markets. Adoption of AI-assisted tools is slow relative to fully digital sectors. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate has adopted AI tools (virtual tours, chatbots, listing generation) moderately, but the core showing/explaining task remains largely human-driven with adoption concentrated in marketing support. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI augmentation is strong here: generative descriptions, virtual staging, 3D tours, and data-driven property recommendations all measurably assist agents in preparing presentations and engaging clients, while agents remain central to relationship-building and closing. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI substantially aids agents via virtual staging, 3D tours, automated property descriptions, and client-matching tools that improve efficiency while the agent still conducts client interactions. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can generate property descriptions and virtual tours, the task requires live in-person interaction, reading client preferences, and responding to dynamic questions—elements that current AI cannot fully automate without significant human involvement. The core activity (displaying and explaining to a specific client) remains dependent on human presence. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical property tours and in-person feature explanation require presence and interpersonal engagement that current AI cannot fully replicate, though virtual tours and AI-generated descriptions can offset some portions., |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Real estate sales involve fiduciary duties, liability for property misrepresentation, and client trust that typically require a licensed human agent to conduct showings and explain features. Regulatory frameworks and consumer expectations create high barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement mandates a human show a property, but liability for property access, safety, security, and client trust create meaningful friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Virtual tour platforms and property description tools are relatively inexpensive, but integrating them into a seamless client experience still requires significant human agent labor. The cost of AI systems plus mandatory agent oversight does not yet undercut agent wages for this bespoke service. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Virtual staging/tour tools are cheap relative to agent time for marketing, but actual property showings still require human presence, keeping overall cost comparable to or higher than pure AI substitution. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Virtual tour technology and AI-generated property descriptions are in production, but they serve as supplements to agent interaction rather than end-to-end replacements. VR walkthroughs and chatbots exist but lack the contextual judgment and personalized client engagement that defines this task. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Virtual tour platforms and AI-generated property descriptions exist and are used, but they supplement rather than replace agent-led showings, especially for higher-value or complex properties. |
Locate and appraise undeveloped areas for building sites, based on evaluations of area market conditions.
30CI 30–30 · exposure 25 · augmentation 50 · importance 2.3/5 · click for rater detail
Locate and appraise undeveloped areas for building sites, based on evaluations of area market conditions.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains highly localized and relationship-driven; adoption of AI for site appraisal is slower than in finance or information sectors. Most firms use data tools supplementarily but continue traditional scouting and appraisal practices. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate remains a relationship-driven, moderately digitized sector; AI tools are used for lead generation and marketing but land-site evaluation workflows show slow, limited AI integration in practice. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by surfacing market trends, zoning data, comparable sales, and development potential via analytics and mapping, helping agents screen candidates faster. However, the human still drives the final appraisal and site selection decision. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can meaningfully assist by aggregating zoning data, market trends, and comparable sales, helping agents narrow down site options and speeding preliminary market analysis. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can analyze market data, property records, and zoning information to support site evaluation, but the task requires on-site judgment about soil conditions, neighborhood character, development potential, and local relationships that current systems cannot reliably assess end-to-end. Human appraisal of undeveloped land remains essential. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can support data gathering and comparables analysis, but locating undeveloped sites and forming appraisal judgments requires physical site visits, negotiation context, and local knowledge that current AI cannot fully replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Real estate transactions and development approvals often require licensed agent involvement and legal signoff, and clients typically prefer in-person relationship and local expertise. Regulatory requirements and customer preference create moderate friction to pure automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Formal appraisals often require licensed appraisers, and real estate transactions involve fiduciary and disclosure obligations, creating moderate regulatory and liability friction against pure automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-based market analysis and mapping tools are relatively cheap, but the human expert must still visit sites, conduct negotiations, and validate findings. The total cost-per-appraisal remains comparable to or higher than traditional methods when integration and human oversight are included. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While data tools reduce some research time cheaply, the human agent's site visits, negotiation, and judgment-heavy appraisal work still dominate cost, keeping AI cost savings modest relative to the full task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Data analytics and GIS tools exist to support market analysis, but no deployed product reliably performs complete site appraisal and locating suitable undeveloped areas without substantial human field work and local expertise. Tools assist rather than perform the task end-to-end. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Products exist for property data aggregation and automated valuation models (AVMs), but these are narrow in scope and not reliable for undeveloped land appraisal, which lacks standardized comparables and requires nuanced judgment. |
Confer with escrow companies, lenders, home inspectors, and pest control operators to ensure that terms and conditions of purchase agreements are met before closing dates.
28CI 25–30 · exposure 25 · augmentation 63 · importance 4.7/5 · click for rater detail
Confer with escrow companies, lenders, home inspectors, and pest control operators to ensure that terms and conditions of purchase agreements are met before closing dates.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains fragmented across local markets, independent brokerages, and regulatory jurisdictions. While some large brokerages use transaction management software, deep automation of escrow coordination is not widely adopted or measured in production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a traditionally slow-adopting, relationship-driven, physically-anchored sector where transaction coordination software is used but full AI-driven agent coordination is not yet mainstream. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by tracking document deadlines, flagging missing items, drafting routine correspondence, and summarizing inspection/appraisal reports—useful augmentation that raises agent productivity without removing the human from judgment and compliance authority. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can draft status update communications, track deadlines, flag missing documents, and summarize inspection reports, meaningfully assisting agents while they retain responsibility for negotiation and final decisions. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could draft routine coordination messages and track documents, the task fundamentally requires judgment about whether terms are met and real-time negotiation with multiple external parties—activities that still require human authority and relationship management today. Partial automation of communication logging and scheduling is possible, but end-to-end execution falls well short of 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | This involves coordinating multiple parties, tracking contingencies, and negotiating timing/issues that arise, which requires real-time judgment and relationship management beyond simple document processing.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Real estate transactions are heavily regulated; escrow companies and title firms are themselves licensed and have legal obligations. The coordinating agent must act on behalf of the buyer/seller and bear reputational and liability risk—creating strong friction against full automation and a requirement for human sign-off on compliance. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Real estate transactions often require licensed agent involvement for certain communications and there's liability exposure in miscommunication about contract terms, creating moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Integration costs, human oversight, and the need for domain-specific setup (APIs to escrow/lender systems, custom rule engines) remain substantial compared to the wage of an agent or coordinator performing routine coordination tasks. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While software can reduce administrative overhead, human agents still need to make calls, negotiate resolutions to inspection issues, and manage relationships, so all-in AI cost savings are modest rather than order-of-magnitude. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably performs multi-party escrow coordination and compliance verification independently. Tools exist for document management and CRM, but actual verification of contract terms against third-party deliverables and exception handling remain manual and require human judgment. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Transaction management software and CRM tools exist to track deadlines and automate reminders, but no deployed product independently confers with escrow, lenders, inspectors, and pest control to resolve issues before closing. |
Inspect condition of premises, and arrange for necessary maintenance or notify owners of maintenance needs.
28CI 20–35 · exposure 20 · augmentation 50 · importance 3.2/5 · click for rater detail
Inspect condition of premises, and arrange for necessary maintenance or notify owners of maintenance needs.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate is moderately digitized and adoption of AI for property assessment is still in pilot stages. Most agents rely on manual walkthroughs and third-party inspectors; systematic AI-driven premise inspection in production remains rare outside a few high-tech brokerages. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate remains a relatively low-digitization, in-person-heavy sector; while some CRM and smart-home sensor tools are emerging, broad AI-driven inspection adoption is minimal. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted photo analysis and automated defect flagging can help agents prioritize what to inspect closely and generate preliminary reports, improving their workflow efficiency. However, the human agent must remain the decision-maker and verifier, limiting transformative augmentation. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help agents draft maintenance notices, track issues via checklists/apps, and even assist with photo analysis to flag concerns, moderately boosting efficiency even though physical inspection itself remains manual. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI-powered computer vision can identify some visible damage or wear from photos or video, comprehensive premise inspection requires physical presence, tactile assessment (foundation cracks, water damage, structural integrity), and contextual judgment about severity and urgency. Current systems cannot reliably perform the full inspection end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical inspection of a property's condition requires on-site presence, sensory judgment, and often walking through rooms, which current AI cannot perform autonomously; arranging maintenance involves coordination but still needs a human decision-maker on-site first. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Real estate sales and property management are heavily regulated with liability concerns; a human agent or licensed inspector typically must legally sign off on property condition disclosures and maintenance recommendations to protect both broker and client. Failure to disclose known defects carries legal consequences, creating a hard barrier to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically for inspecting premises, though real estate agents' fiduciary duty and liability for missed defects create some caution around fully automating this without human oversight. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI vision tools exist but require expensive hardware (thermal imaging, 3D scanners), integration with property management systems, and substantial human oversight to validate findings and prioritize maintenance. The all-in cost approaches or exceeds a field inspector's hourly rate for a single property. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI cannot replace the physical inspection component, so cost comparison mostly applies to the administrative follow-up (notifying owners, scheduling), where AI is cheap but represents only a small slice of the full task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Deployed products exist for visual defect detection in images, but they remain narrow in scope and require significant human validation. No production system reliably performs autonomous premise inspection without expert human review and on-site verification. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical property inspections independently; some computer-vision tools assist with photo-based defect detection but are not standard production tools for agents doing this task. |
Rent or lease properties on behalf of clients.
28CI 25–30 · exposure 25 · augmentation 63 · importance 2.7/5 · click for rater detail
Rent or lease properties on behalf of clients.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate has slow digitization compared to finance or tech; most transactions still rely on human agents for trust and local knowledge. Pilots of autonomous leasing exist but penetration in production remains shallow; adoption is primarily in lead capture, not transaction closure. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a traditionally slow-adopting, relationship-driven, physically-grounded sector where AI use is mostly limited to marketing and lead generation pilots rather than deep transactional automation. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by automating property search filtering, generating comparable market analyses, drafting lease terms, and scheduling showings—all of which raise agent productivity. However, the human agent remains central to negotiation and closing, so augmentation is partial rather than transformative. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI meaningfully assists agents with drafting listings, screening tenants, scheduling, market analysis, and answering inquiries, significantly boosting productivity while the agent remains central to the transaction. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with property matching, scheduling, and initial communication, finalizing rentals requires negotiation, legal judgment, fraud detection, and contextual understanding of tenant-landlord dynamics. The human agent remains essential for building trust, handling exceptions, and closing deals—no off-the-shelf system achieves ≥50% time savings end-to-end at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Renting/leasing involves relationship-building, negotiation, property showings, and client-specific judgment that current AI cannot fully replicate end-to-end, though listing, screening, and communication sub-tasks can be automated. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | State licensing of real estate agents, fiduciary duty requirements, anti-discrimination law oversight, and state-specific lease regulations create hard adoption barriers. Fair Housing Act compliance and liability for wrongful refusal mean a licensed human must ultimately sign off on tenant selection and lease execution. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Real estate agents often need licensing to lease/rent properties legally, and fiduciary duties, contracts, and liability for lease terms create moderate barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for lead generation and property matching are affordable, but integrating fraud detection, legal document generation, and human oversight adds cost. The loaded wage for a real estate agent includes significant commission and benefits; current AI+oversight remains more expensive than narrow task automation alone. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI tools can cut costs for lead qualification and paperwork, but human agents still must handle showings, negotiations, and relationship management, keeping overall cost comparable to or only modestly cheaper than human-only processes. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Property listing and matching systems exist (Zillow, Apartments.com), but no deployed product reliably handles the full leasing workflow: tenant screening, price negotiation, lease document customization, and legal compliance verification. Most systems are narrow filters rather than autonomous closers. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Products exist for tenant screening, chatbots for inquiries, and listing syndication, but no deployed system reliably handles full lease negotiation, showings, and closing on behalf of a client at scale. |
Advise clients on market conditions, prices, mortgages, legal requirements, and related matters.
27CI 25–29 · exposure 25 · augmentation 63 · importance 4.2/5 · click for rater detail
Advise clients on market conditions, prices, mortgages, legal requirements, and related matters.
27| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains a relationship-driven, licensed profession with slow digital transformation relative to information-sector tasks. While market data tools are adopted, independent AI advisors have not meaningfully penetrated production workflows in the sector. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate remains a relationship-driven, moderately digitized sector with slow, fragmented AI adoption at the agent level despite proptech tools proliferating. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist agents by providing instant market comparables, mortgage calculators, legal requirement summaries, and document drafting support, raising their efficiency in research and proposal preparation while the agent retains advisory authority. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI substantially aids agents by quickly synthesizing market data, comparables, and mortgage information, letting agents deliver more informed advice faster while retaining the client relationship. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can retrieve and summarize market data, mortgage rates, and legal requirements, advising clients requires contextual judgment, negotiation sensitivity, and fiduciary responsibility that current systems cannot reliably deliver end-to-end. AI may assist in data gathering but cannot replace the human advisor role. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can generate general market summaries and answer common questions, but personalized advice integrating client-specific financial situations, negotiation strategy, and local nuance still requires human judgment and relationship trust. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Real estate agents typically hold state licenses and are legally responsible for advice given to clients; liability and fiduciary duty create substantial regulatory barriers. Clients expect and often prefer direct human relationships with licensed professionals for major financial decisions. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Real estate agents are licensed professionals and legal/financial advice carries liability exposure; many jurisdictions require disclosures only a licensed agent can provide, creating meaningful regulatory and professional barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI systems require significant human oversight, verification, and liability management, making the all-in cost per advisory interaction comparable to or higher than direct human consultation. The integration overhead offsets any inference savings. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI-generated market data and basic Q&A are cheap, but the full advisory service still requires human oversight and liability coverage, keeping overall cost comparable rather than dramatically cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Deployed products (chatbots, market analysis tools) exist but lack the reliability, personalization, and legal accountability required for genuine client advisory. No production system today independently advises clients on complex real estate decisions without human oversight and final sign-off. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Chatbots and AI tools (e.g., Zillow estimates, mortgage calculators) exist and are used for informational support, but no product reliably substitutes for an agent's holistic advisory role in production at scale. |
Appraise properties to determine loan values.
26CI 20–31 · exposure 25 · augmentation 63 · importance 3.2/5 · click for rater detail
Appraise properties to determine loan values.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate and lending sectors show slow, cautious adoption of AI for appraisals due to regulatory and liability constraints; pilots exist for data preprocessing, but production displacement of the appraisal function itself remains minimal and legally restricted. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Real estate and mortgage lending have adopted AVMs and AI-assisted valuation tools moderately, with pilots and partial integration common, but full displacement of human appraisal judgment remains limited. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist appraisers by automating data gathering, comparable sales analysis, and preliminary valuation estimates, raising appraiser productivity in research and documentation phases while the licensed professional retains final determination authority. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered valuation tools and comparable sales analytics significantly speed up an agent's ability to estimate value and prepare supporting data, meaningfully boosting productivity even though a human retains final judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Property appraisal requires site inspection, assessment of structural condition, local market analysis, and professional judgment on comparable sales. While AI can assist with data aggregation and valuation models, the in-person inspection and final appraisal determination legally require a licensed appraiser, leaving only preparatory data work automatable today. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can assist with comparable sales analysis and preliminary valuation estimates, but final appraisal for loan purposes requires site inspection, judgment on condition, and often licensed sign-off that current AI cannot fully replace end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Strong regulatory barriers: licensed, independent appraisers are legally required by lending standards (Dodd-Frank, FIRREA) and loan securitization rules; lenders cannot rely on unlicensed AI appraisals for mortgage underwriting, creating a hard ceiling on automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Loan appraisals are often subject to regulatory requirements (e.g., USPAP standards, licensed appraiser requirements) and lender risk policies that require human sign-off, creating strong legal and liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Licensed appraisers charge $300–$600+ per appraisal in many markets. Current AI tools for valuation models and market analysis cost significantly less, but cannot replace the licensed appraiser requirement, so meaningful cost savings remain limited to partial tasks. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AVMs are cheap to run compared to a human appraiser's fee, but the need for human verification, physical inspection, and liability coverage keeps blended costs closer to comparable rather than an order of magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI product reliably performs full property appraisals independently; regulatory requirements mandate licensed human appraisers. AI tools exist for comparative market analysis and preliminary valuation estimates, but banks and lenders cannot rely on autonomous AI for loan-supporting appraisals at scale. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Automated valuation models (AVMs) exist and are used by lenders as a supplementary tool, but they have material error rates and are not treated as reliable stand-alone replacements for formal appraisals in most loan transactions. |
Act as an intermediary in negotiations between buyers and sellers, generally representing one or the other.
25CI 25–25 · exposure 25 · augmentation 75 · importance 4.8/5 · click for rater detail
Act as an intermediary in negotiations between buyers and sellers, generally representing one or the other.
25| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains heavily relational and human-driven despite digitization of listings and tools. Adoption of AI negotiation remains at pilot/tool level, not production automation of the intermediary role itself. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a moderately digitized but relationship-driven, fragmented industry with slow AI adoption for core negotiation tasks, though back-office and marketing uses are growing. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can meaningfully assist agents by drafting communications, analyzing market data, summarizing counteroffers, and flagging negotiation points, thereby accelerating the human negotiator's workflow and decision-making without replacing the need for human judgment and legal authority. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist agents by analyzing comparable sales, drafting counteroffer language, and simulating negotiation scenarios, improving efficiency while the agent remains the primary negotiator. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft communications and summarize positions, negotiation requires real-time judgment, emotional intelligence, and legal accountability that current systems cannot handle end-to-end. An agent would need constant human intervention in the core negotiation loop, making <50% time savings unlikely. |
| Task automatability | claude-sonnet-5 | 2/5 | Negotiation involves real-time relationship management, reading emotional cues, trust-building, and representing a client's interests, which current AI cannot fully replicate end-to-end despite being able to draft offers or suggest counteroffers.atable savings are partial at best.rationale |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Real estate agents must be licensed in all jurisdictions, and fiduciary duty to represent a buyer or seller creates legal liability that cannot be delegated to an AI system. Clients expect and often require human contact and accountability. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Agency law, fiduciary duty, and licensing requirements typically mandate a licensed human agent represent a client's interests in negotiations, creating strong legal and liability barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for real estate (CRM, email drafting) are useful supplements but do not reduce the per-transaction cost below that of a human agent, since the human remains essential for closing deals and managing client relationships. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Human agents are compensated via commission tied to deal closure and trust, and AI cannot yet substitute for the relationship-based negotiation, so cost comparison favors humans in most cases. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably conducts real estate negotiations independently. Existing systems can support document preparation and communication drafting, but they lack the judgment and accountability required to represent a party's interests in a binding transaction. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some AI tools assist with pricing analysis or draft communications, but no deployed product autonomously conducts real estate negotiations on a client's behalf at scale. |
Coordinate property closings, overseeing signing of documents and disbursement of funds.
25CI 25–25 · exposure 25 · augmentation 63 · importance 4.6/5 · click for rater detail
Coordinate property closings, overseeing signing of documents and disbursement of funds.
25| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption remains slow; most closings still rely on traditional title companies and escrow services with human coordinators. While document platforms are digitizing workflow, the end-to-end closing coordination and oversight remain stubbornly human-dependent across the sector. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a moderately digitized but still relationship- and paper-heavy sector; adoption of AI/automation in the closing process itself has been slow and mostly limited to e-signature and document management tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by automating document assembly, flagging inconsistencies, and organizing fund flows, but the human agent retains essential coordinative and supervisory duties. Moderate productivity lift for agents managing multiple closings simultaneously. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI and software tools (e-signature platforms, transaction management systems, automated checklists and reminders) meaningfully streamline document tracking and communication, letting agents manage closings more efficiently while remaining in control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While document preparation and fund tracking are partially automatable, the core requirement of overseeing and coordinating the closing—ensuring all parties are present, understanding executed agreements, and managing real-time issues—demands human judgment and legal accountability. Current AI cannot reliably replace the coordination and oversight role. |
| Task automatability | claude-sonnet-5 | 2/5 | Closing coordination involves scheduling, verifying documents, resolving last-minute issues, and interfacing with multiple parties (title, lender, attorney) which requires judgment and real-time coordination that current AI cannot fully replace end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant regulatory and legal barriers exist: escrow and closing coordination often require licensed professionals (title agents, escrow officers) or attorney sign-off depending on state law. Liability and fiduciary duty create hard constraints on automation without human authorization. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Fund disbursement and document execution in real estate closings often require licensed professionals (attorneys, title agents, notaries) and are subject to state regulations and fiduciary/liability requirements, creating strong barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | The cost of AI tooling plus required human oversight (title company, escrow officer, legal review) does not yet undercut the loaded wage of a real estate agent performing this task, especially when liability and error correction are factored in. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Software tools reduce administrative overhead somewhat cheaply, but the human coordination, liability management, and problem-solving during closing still require paid agent/attorney time, keeping costs comparable. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI product reliably orchestrates end-to-end property closings with signing oversight and fund disbursement. Tools exist for document preparation and management, but the coordinating, supervisory, and fiduciary aspects remain human-centric in production. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some transaction management platforms automate document tracking and reminders, but no deployed product independently oversees signing and fund disbursement without human oversight in production at scale. |
Present purchase offers to sellers for consideration.
19CI 13–25 · exposure 17 · augmentation 63 · importance 4.9/5 · click for rater detail
Present purchase offers to sellers for consideration.
19| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains a highly relationship-driven, human-centric profession with slow digital transformation in core negotiation tasks. While some firms use AI for lead generation and document preparation, actual offer presentation continues to rely on agents, and adoption of AI in this specific task is minimal. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate remains a relationship-driven, moderately digitized industry with slow AI adoption for direct client-facing negotiation tasks, despite growth in AI-assisted tools for listings and paperwork. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist agents by preparing offer summaries, analyzing comparable sales data, and flagging negotiation risks before the presentation. These tools improve agent productivity and confidence in the interaction, though the human agent remains essential for the persuasive, relationship-building component. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can help agents draft offer summaries, comparative market analyses, and communication templates, improving efficiency while the agent still delivers and negotiates the offer personally. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Presenting offers to sellers requires interpersonal negotiation, reading nuanced reactions, understanding local market context, and building trust—human skills that current AI cannot perform end-to-end. AI can draft offer language but cannot substitute for the agent's presence and judgment in the critical negotiation moment. |
| Task automatability | claude-sonnet-5 | 2/5 | Presenting an offer involves communicating terms, negotiating, and reading seller reactions/relationships, which requires interpersonal judgment beyond current AI capabilities to fully replace, though drafting and summarizing offers can be automated.on |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Sellers typically expect a licensed real estate agent to present offers in person, and fiduciary duties require human judgment about offer presentation timing and strategy. Many jurisdictions legally require a licensed agent to handle material parts of the transaction, creating regulatory and contractual barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Real estate transactions often require licensed agents for fiduciary duties, disclosures, and legal representation, and sellers typically expect a licensed professional to present offers and advise on negotiation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of AI infrastructure, legal review, and human oversight for offer presentation would exceed the cost of a real estate agent performing the task directly. The task's high-stakes nature (large financial transactions) demands human accountability that increases, not decreases, total cost. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI can help draft or summarize offer terms cheaply, the human agent still must conduct the actual presentation and negotiation, so cost savings on the full task are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably performs the full task of presenting and negotiating offers with sellers. While AI can assist in drafting documents and analyzing comparables, the actual presentation—which involves persuasion, rapport, and real-time response to objections—remains dependent on human agents in practice. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | No deployed product autonomously presents purchase offers to sellers in real transactions; agents still handle this interpersonal, negotiation-laden interaction directly. |
Visit properties to assess them before showing them to clients.
16CI 5–26 · exposure 8 · augmentation 38 · importance 3.2/5 · click for rater detail
Visit properties to assess them before showing them to clients.
16| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains relatively non-automated in core functions despite digitization efforts. While virtual tours and 3D imaging are adopted, actual property assessment remains agent-driven, reflecting organizational and market resistance to full automation. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Real estate remains a physically grounded, relationship-driven sector with low digitization of in-person property assessment tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted tools (automated photo analysis, condition flagging from images, comparative market data overlays) can assist agents in prioritizing what to inspect and documenting findings, but the human agent remains central to the assessment process. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with pre-visit research (comps, area data, photos analysis) but offers little help during the actual physical walkthrough and assessment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Physical property visits require on-site presence and embodied assessment of condition, layout, and environment—tasks current AI cannot perform. While AI could process photos/videos of properties already captured, autonomously visiting and inspecting properties remains infeasible with today's technology. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physically traveling to and walking through a property to assess its condition, layout, and features—no current AI system can perform this physical inspection task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | This task requires physical presence and liability exposure; agents' professional judgment about property condition directly affects client trust and legal exposure. The embodied nature of assessment and fiduciary responsibility create substantial barriers to automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human specifically visit the property, but practical necessity of physical presence and liability for accurate representation create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | AI cannot perform this task end-to-end, making cost comparison moot. Any partial automation (photo analysis) would still require human site visits, leaving the baseline cost largely unchanged. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute for the physical visit itself, so AI cost is not comparable—the human must still perform this in-person task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can independently visit and assess physical properties. Computer vision can analyze pre-recorded imagery, but autonomous property inspection requiring physical navigation and real-time judgment is not a solved production capability. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs autonomous physical property visits and assessments; this remains a purely human, in-person activity. |
Accompany buyers during visits to and inspections of property, advising them on the suitability and value of the homes they are visiting.
13CI 5–21 · exposure 8 · augmentation 50 · importance 4.0/5 · click for rater detail
Accompany buyers during visits to and inspections of property, advising them on the suitability and value of the homes they are visiting.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Real estate remains heavily dependent on local, relationship-driven sales by licensed agents. Adoption of autonomous AI for client accompaniment during inspections is negligible; the sector has slow digitization in this specific role and strong regulatory/human preference requirements. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate remains a relationship-driven, physically-grounded sector with modest AI adoption for tasks like descriptions or lead-scoring, but not for in-person property visits. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could usefully assist by providing property data, comparable sales information, and inspection checklists that the agent references during the visit, raising their productivity and confidence. However, the human agent remains the primary decision-maker and client-facing communicator. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can help agents prepare talking points, comparables, and valuation data beforehand, and provide virtual pre-visit tours, meaningfully aiding the human agent's in-person work without replacing the visit itself. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could assist in preparing property information and initial suitability assessments, the core task requires real-time human presence, judgment about buyer preferences, negotiation, and relationship-building during property visits. AI cannot meaningfully substitute for the in-person advisory role that requires contextual understanding and persuasion. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical presence, in-person navigation of properties, and real-time in-person rapport with buyers, none of which current AI systems can perform end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: real estate transactions involve fiduciary duties, liability for misrepresentation, and regulatory licensing requirements that typically mandate a licensed human agent. Buyers also expect human presence and trust-building, creating preference for human contact. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing barrier prevents an agent from being replaced generally, but buyers strongly prefer human presence and trust for high-stakes purchases, and physical accompaniment inherently requires a human or robot to be there. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying an AI agent to accompany clients physically and provide reliable real-time advice would far exceed the cost of a human real estate agent's time for this task. The infrastructure, liability, and oversight required would be prohibitively expensive. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot perform the physical accompaniment component at all, so there is no viable cost comparison for full task substitution. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs the full task of accompanying buyers on property inspections and providing real-time advisory judgment. Virtual tours exist, but they do not replace the human agent's physical presence and interpersonal guidance during actual property visits. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product physically accompanies buyers to properties and provides live in-person advisory; virtual tours and chatbots exist but do not replace this physical task. |
Develop networks of attorneys, mortgage lenders, and contractors to whom clients may be referred.
10CI 7–13 · exposure 0 · augmentation 38 · importance 3.8/5 · click for rater detail
Develop networks of attorneys, mortgage lenders, and contractors to whom clients may be referred.
10| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Real estate remains relatively low in AI adoption for client-facing and relationship work; the industry is traditional, geographically fragmented, and heavily dependent on personal networks and reputation. Automation of relationship-building remains peripheral to industry practice. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Real estate is a moderately digitized but still relationship-driven, physically-oriented sector where AI adoption for networking and referral-building specifically remains minimal. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can provide minor assistance by identifying potential referral partners, organizing contact information, or drafting outreach messages, but these are peripheral support tasks. The core work of establishing trust and mutual obligation with attorneys, lenders, and contractors requires the agent's own credibility and ongoing personal engagement. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI tools can help agents track contacts, draft outreach messages, and identify potential referral partners via CRM and data analysis, offering moderate productivity support around the edges of the task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Developing professional networks requires genuine relationship-building, trust-establishment, and ongoing personal interaction that AI cannot perform autonomously. Current AI systems cannot conduct the complex social negotiation, credibility assessment, and reciprocal obligation-building necessary to create functional referral networks. |
| Task automatability | claude-sonnet-5 | 1/5 | This is fundamentally a relationship-building and networking task requiring in-person trust development, reputation management, and reciprocal referral arrangements that AI cannot perform autonomously. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong barriers exist: real estate transactions involve legally-licensed agents who bear fiduciary responsibility, and referral network quality directly affects client outcomes and agent liability. Trust-based referral relationships inherently require human judgment and accountability that cannot be delegated to automated systems. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no license is strictly required to network, professional referral relationships depend on personal trust, reputation, and reciprocity that create strong organic barriers to AI substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The task fundamentally requires human social capital and relationship management; AI cannot substitute for the personal credibility and judgment an agent brings to network relationships. Any cost savings from AI assistance would be marginal compared to the human-dependent core work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no AI substitute performing this networking function, so cost comparison favors the human entirely since AI cannot replace the trust-based relationship building involved. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI product reliably performs relationship development and network cultivation with external professionals. While AI can assist in identifying contacts or drafting outreach, it cannot execute the repeated personal interactions, trust-building, and mutual benefit arrangements that constitute actual network development. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product builds or maintains professional referral networks; this remains a human relationship-management activity with no production automation. |
Related occupations — Sales & Related
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.