First-Line Supervisors of Retail Sales Workers
41-1011.00Median wage $48,520/yr1,121,800 employed (US)Rank #136 of 923 scored · top 15% by substitution
Directly supervise and coordinate activities of retail sales workers in an establishment or department. Duties may include management functions, such as purchasing, budgeting, accounting, and personnel work, in addition to supervisory duties.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution42
Exposure37
Augmentation65
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
21 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
19%
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%36
panel mean rating 2.5/5 → substitution pressure 36/100
Technical feasibility todayw 20%39
panel mean rating 2.6/5 → substitution pressure 39/100
Cost vs. human wagew 15%42
panel mean rating 2.7/5 → substitution pressure 42/100
Adoption barriersw 20%inverted — strong barriers lower the score57
panel mean rating 2.7/5 (barrier strength) → substitution pressure 57/100
Sector adoption velocityw 10%41
panel mean rating 2.6/5 → substitution pressure 41/100
Task breakdown (21 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.
Keep records of purchases, sales, and requisitions.
89CI 79–100 · exposure 87 · augmentation 75 · importance 4.2/5 · click for rater detail
Keep records of purchases, sales, and requisitions.
89| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Retail is highly digitized and has been automating transactional record-keeping for decades via POS systems, inventory software, and e-commerce platforms. Adoption is near-universal in organized retail. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Retail is a highly digitized sector with widespread, mature adoption of automated inventory and sales tracking systems, though smaller independent retailers may lag. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems assist supervisors by automatically logging records, generating reports, and flagging anomalies, allowing supervisors to focus on exception handling and strategic inventory decisions rather than manual entry. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered dashboards and reporting tools significantly reduce the manual burden of record-keeping, letting supervisors focus on exceptions and analysis rather than data entry. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Record-keeping of purchases, sales, and requisitions is primarily data entry and transaction logging—highly structured, repetitive work. Current AI systems (RPA, ERP integrations, document processing) can extract, classify, and log this information end-to-end with >50% time savings and equal accuracy to human entry. |
| Task automatability | claude-sonnet-5 | 4/5 | Recording purchases, sales, and requisitions is largely structured data entry and tracking, which POS systems, ERP software, and AI-enhanced inventory tools already automate to a high degree with minimal human input. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few hard barriers exist; most retailers have already adopted automation for this function. Some organizational inertia and system integration friction remain, but no licensing or legal requirement mandates human record-keeping. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates human record-keeping for retail transactions; this is a purely administrative function with no legal barrier to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated record-keeping via existing software infrastructure costs negligibly per transaction—orders of magnitude cheaper than manual human data entry at retail supervisor wages. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated record-keeping software costs a small fraction of a supervisor's time spent on manual logging, and per-transaction costs are negligible compared to loaded wages. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed retail management software, inventory systems, and point-of-sale integrations already perform this task reliably at scale in thousands of organizations. Modern ERP and WMS platforms automate record-keeping with minimal human intervention. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Retail management systems (POS, inventory management, ERP platforms like Square, Shopify, SAP) reliably automate transaction and requisition record-keeping in production today across most retail environments. |
Review inventory and sales records to prepare reports for management and budget departments.
85CI 72–97 · exposure 87 · augmentation 88 · importance 4.2/5 · click for rater detail
Review inventory and sales records to prepare reports for management and budget departments.
85| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Retail chains and large organizations have adopted automated reporting pipelines and BI systems extensively; most mid-to-large retail enterprises now use dashboard automation rather than manual supervisor reporting. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Retail has moderate digitization with POS and inventory systems widely adopted, but many smaller retail operations still rely on manual or semi-manual reporting processes. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-driven BI dashboards and automated report generation strongly augment supervisor productivity by providing real-time, interactive data views and exception alerts that enable faster, data-driven decision-making than manual spreadsheet compilation. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI-powered analytics tools significantly speed up data review, trend identification, and report drafting, letting supervisors focus on interpretation and decision-making. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Current AI systems can autonomously extract data from inventory and sales databases, aggregate metrics, and generate formatted reports that meet typical management and budget department requirements with minimal human intervention—easily achieving >50% time savings. |
| Task automatability | claude-sonnet-5 | 4/5 | Compiling inventory and sales data into management/budget reports is largely structured data aggregation and summarization, which current AI and BI tools can do end-to-end with significant time savings, though setup and data integration reduce full automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | No licensing, legal, or regulatory requirement mandates human review of inventory reports; automation purely supports internal management visibility with no liability asymmetry or mandatory human sign-off. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates human preparation of these reports, though managers may want human review/sign-off for accuracy and context before submission. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Cloud-based analytics and BI platforms cost cents per query; AI report generation via LLMs or RPA pipelines runs at <$1 per report, orders of magnitude cheaper than a supervisor's fully-loaded wage for hours of manual compilation. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated reporting software costs a small fraction of a supervisor's time spent manually compiling reports, especially at scale across many stores. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | BI tools, reporting platforms (Tableau, Power BI), and AI-driven analytics systems are deployed at scale across retail organizations today, reliably pulling records and generating standardized reports in production environments. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Retail BI platforms and AI-based reporting tools (e.g., automated dashboards, POS-integrated analytics with AI summarization) are already deployed at scale in retail chains to generate these reports. |
Plan and prepare work schedules and keep records of employees' work schedules and time cards.
82CI 75–89 · exposure 80 · augmentation 88 · importance 4.2/5 · click for rater detail
Plan and prepare work schedules and keep records of employees' work schedules and time cards.
82| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Retail and food service—primary venues for this task—show rapid adoption of scheduling and workforce-management software, particularly post-2020. Major chains and franchises now routinely deploy AI-assisted or automated scheduling tools in production. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Retail has broadly adopted digital scheduling and time-tracking software over the past decade, though smaller independent stores lag behind chains. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | Current scheduling tools assist supervisors by generating candidate schedules, flagging compliance violations, managing time-off requests, and centralizing records, significantly raising their productivity and freeing time for higher-value supervisory work while supervisors remain the decision-maker. |
| Augmentation potential | claude-sonnet-5 | 5/5 | Even where full automation isn't used, scheduling software dramatically speeds up supervisors' planning process, optimizing shifts and flagging conflicts while the human retains final approval. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Scheduling and time-card record-keeping are highly structured, rule-based tasks with clear inputs (employee availability, labor laws, demand forecasts) and outputs (schedules, records). Current AI systems can automate 70–80% of this work via constraint-solving algorithms and automated record management, though human oversight of fairness and edge cases remains typical. |
| Task automatability | claude-sonnet-5 | 4/5 | Workforce scheduling software already automates shift creation, time-card tracking, and record-keeping based on demand forecasts and labor rules, meeting the time-saving threshold for most of this task.imated as employee constraints still need occasional human review. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few legal or regulatory barriers exist; scheduling is not a licensed activity and no liability asymmetry prevents automation. Main friction is organizational (preference for human discretion in fairness) and labor-relations concerns, but these are surmountable. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or legal requirement mandates human-only scheduling; retailers freely adopt automated tools with minimal regulatory friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Scheduling software (SaaS or integrated HR systems) costs far less per task than the loaded wage of a first-line supervisor spending hours weekly on manual scheduling and record-keeping; automation software typically runs $10–100/employee/year versus supervisor labor. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | SaaS scheduling tools cost a small monthly fee per employee versus the substantial supervisor time saved, making AI-assisted scheduling far cheaper than manual scheduling. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Multiple production systems (e.g., Zip Schedules, When I Work, 7shifts) and enterprise workforce-management platforms reliably perform scheduling and time tracking at scale across retail organizations. These tools are mature and widely deployed, though typically require human validation of complex requests. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature scheduling and time-and-attendance products (e.g., Deputy, When I Work, Kronos) are widely deployed in retail today and reliably handle scheduling and record-keeping at scale. |
Inventory stock and reorder when inventory drops to a specified level.
82CI 72–91 · exposure 80 · augmentation 88 · importance 4.2/5 · click for rater detail
Inventory stock and reorder when inventory drops to a specified level.
82| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Retail and food-service sectors show very high adoption of automated inventory and reorder systems, with most chains above a certain size operating these systems in production; small independent retailers lag but the sector trend is rapid. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Large retail chains have widely adopted automated inventory systems, but many small and mid-sized retail operations still rely on manual or semi-manual processes, giving mixed overall adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-driven inventory dashboards and predictive reorder alerts significantly augment supervisors' productivity by providing real-time visibility, demand forecasts, and automated recommendations, allowing them to focus on exception handling and supplier negotiation rather than manual counts. |
| Augmentation potential | claude-sonnet-5 | 5/5 | Even where full automation isn't in place, AI-driven inventory analytics and demand forecasting significantly boost a supervisor's ability to manage stock levels efficiently. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems with RFID integration, barcode scanning, and inventory management software can fully automate stock counting and reorder triggering when thresholds are reached, delivering substantial time savings and equal quality. The task involves routine data collection and rule-based decision-making, both well within deployed AI capabilities. |
| Task automatability | claude-sonnet-5 | 4/5 | Inventory tracking and reorder-point triggering is a well-structured, rule-based task that off-the-shelf inventory management and POS systems already automate with minimal human input beyond exception handling. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While some organizations prefer human oversight for exception handling and vendor relationships, there are no legal or regulatory barriers mandating human involvement in inventory counting or reordering. Integration and organizational habits create modest friction but not hard blockers. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing, legal, or safety requirement mandating human performance of inventory counting or reordering; it's a purely administrative/operational task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Fully automated inventory systems cost a fraction of the human supervisory labor required; annual SaaS licensing for inventory software is typically far cheaper than the salary of a first-line supervisor managing this task. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated inventory systems have low marginal software/API cost compared to paying a supervisor's time to manually count and reorder stock, though initial integration costs exist. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Mature inventory management systems (SAP, Oracle, Shopify, Toast) and automated reordering platforms are deployed at scale across retail organizations today, handling stock tracking and threshold-triggered reorders reliably in production environments. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Retail inventory management software (e.g., automated reorder systems integrated with POS and supply chain platforms) is widely deployed in production across large and mid-sized retailers today. |
Formulate pricing policies for merchandise, according to profitability requirements.
46CI 34–57 · exposure 38 · augmentation 75 · importance 4.4/5 · click for rater detail
Formulate pricing policies for merchandise, according to profitability requirements.
46| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Large retail chains and e-commerce firms have widely deployed algorithmic pricing and optimization tools, demonstrating fast adoption in digitized, data-rich retail environments. Smaller and non-chain retailers lag, but the sector overall shows strong momentum. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail supervisory roles are only moderately digitized; sophisticated pricing AI is common in large retail chains but this occupation broadly (small-to-mid retail supervisors) shows slower, shallower adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | Pricing analytics platforms significantly assist supervisors by automating margin calculations, competitor benchmarking, and scenario modeling, allowing faster policy iteration and more informed decision-making. The human typically remains accountable but wields AI-generated intelligence to make better recommendations. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can meaningfully assist by providing competitor price tracking, margin analysis, and demand forecasting, substantially speeding up the analytical component of pricing policy decisions while the supervisor retains final judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can assist with pricing analysis and recommendations based on cost and margin data, but formulating coherent pricing policies requires integrating competitive intelligence, brand positioning, inventory dynamics, and organizational strategy—judgments that demand human oversight and accountability. Most of the task remains manual. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can analyze cost data, competitor pricing, and demand elasticity to generate pricing recommendations, but final policy formulation requires judgment about brand positioning, local market context, and business goals that current AI cannot fully own end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Pricing is a sensitive business function tied to profitability and competition, so organizations typically maintain human supervisory sign-off on policy decisions. No hard licensing barrier exists, but organizational governance, liability concerns over systematic under/over-pricing, and brand-risk oversight create meaningful friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but organizational trust, accountability for profitability outcomes, and manager judgment on strategic/local factors create moderate friction against full delegation to AI. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | Enterprise pricing software is moderately expensive and requires ongoing licensing, integration, and data infrastructure, while a retail supervisor's loaded cost is substantial. The AI tools reduce labor per policy iteration, but integration and oversight costs keep the ratio roughly comparable. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Pricing software subscriptions plus data integration costs are moderate; for a small-scale retail supervisor context, cost savings versus manual judgment are present but not dramatic given the human oversight still needed. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Dynamic pricing and promotional optimization tools exist and are used in retail (e.g., Revionics, Blue Yonder), but they typically generate recommendations that supervisors review and adjust rather than autonomously formulating complete policies. These systems work within predefined guardrails and require human validation. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Pricing analytics and dynamic pricing tools exist in retail (especially large chains), but for a first-line supervisor role formulating store-level pricing policy, deployed turnkey products are narrow and mostly used by larger enterprises with dedicated pricing teams, not this role directly. |
Examine merchandise to ensure that it is correctly priced and displayed and that it functions as advertised.
45CI 35–55 · exposure 38 · augmentation 63 · importance 4.4/5 · click for rater detail
Examine merchandise to ensure that it is correctly priced and displayed and that it functions as advertised.
45| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Large retailers (Walmart, Target, Amazon) are piloting computer-vision shelf monitoring, but adoption remains uneven; most small and mid-sized retail still relies on manual inspection. Pilots are common in corporate chains, but production-scale displacement is limited to select operations. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail floor operations are physical and only moderately digitized; while some computer-vision shelf-monitoring pilots exist, widespread production adoption for this specific task is still rare. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI vision systems can dramatically assist supervisors by flagging pricing errors, empty shelves, and misplaced merchandise in real time, allowing humans to focus on verification and functional testing. This substantially raises supervisor productivity while keeping them in the loop for judgment calls. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered shelf-scanning apps and price-checking tools can help supervisors spot pricing discrepancies faster, though human judgment and physical checks remain central. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI vision systems can inspect merchandise for display compliance and pricing labels with reasonable accuracy, but verifying that items 'function as advertised' requires hands-on testing that current systems cannot reliably perform end-to-end. Setup and integration would be substantial, achieving roughly 50% time savings on visual inspection alone. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires physical inspection of merchandise, price tags, and displays on a sales floor plus functional testing of products, which current AI cannot perform end-to-end without robotics and physical presence. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No legal licensing requirement or regulatory mandate forces a human supervisor to perform this task; most retail operates under commercial liability rather than professional licensing. Some organizational preference for human judgment and customer-facing authority creates modest friction, but nothing prevents automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but physical retail environments and lack of existing automation infrastructure in most stores create moderate organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI-powered shelf inspection cameras and software are relatively expensive to deploy and maintain, with ongoing human oversight still required for disputed or complex cases. Costs are roughly comparable to employing a supervisor for this task, depending on store size and frequency. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Deploying cameras, robots, or vision systems for shelf auditing requires significant capital and integration costs that often exceed the marginal cost of a supervisor doing visual checks during normal floor duties. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Computer vision products exist for shelf monitoring and price verification in retail, deployed by some large retailers, but they struggle with occlusion, lighting variation, and confirming actual functionality claims. Narrow scope and material error rates prevent a 5 rating. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some computer vision systems for shelf/price auditing exist in pilot or limited retail deployments, but no mature product reliably performs the full physical inspection and functional testing task at scale. |
Estimate consumer demand and determine the types and amounts of goods to be sold.
45CI 41–49 · exposure 34 · augmentation 75 · importance 3.9/5 · click for rater detail
Estimate consumer demand and determine the types and amounts of goods to be sold.
45| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Retail is highly digitized and large chains deploy demand forecasting AI; however, small retailers and independent stores lag. Aggregate adoption in the sector is rising but unevenly distributed, with major retailers in production and smaller operators still on trial. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Retail has moderate digitization with growing use of demand-forecasting and inventory-management AI, but adoption is uneven across small vs. large retailers, placing it in the middle of the adoption spectrum. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI dashboards showing demand signals, trend analysis, and inventory alerts meaningfully assist supervisors in making better-informed assortment decisions. The human supervisor remains accountable and in the loop, but AI tools substantially raise the speed and quality of their demand estimation. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-driven analytics tools meaningfully assist supervisors by highlighting trends, seasonal patterns, and past sales data, improving decision quality even though final judgment remains human. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can analyze sales data and trends to forecast demand, determining product assortment requires judgment about local market preferences, seasonality, and strategic positioning. Current systems can support analysis but cannot fully automate the decision-making without significant human oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | AI forecasting tools can analyze historical sales data and suggest demand estimates, but the full task includes local judgment, supplier negotiation context, and store-specific factors that require human integration and decision-making. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Retail organizations face moderate friction: demand estimation is embedded in inventory systems and operational workflows, and supervisors remain accountable for stock decisions. No legal licensing barrier exists, but organizational inertia and human judgment precedent create adoption friction. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement forces a human to do this, but organizational trust, accountability for stock decisions, and integration with broader store operations create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI forecasting tools have recurring costs (licensing, data infrastructure) comparable to the supervisory labor they partially displace. Integration and oversight costs roughly offset any efficiency gains, making the cost ratio near-parity. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Forecasting software has licensing and integration costs, but for a single first-line supervisor's task the marginal cost of AI-assisted forecasting is often comparable to or somewhat cheaper than manual estimation given existing tool adoption. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Demand forecasting tools exist in enterprise systems (e.g., inventory management software with ML), but they typically require human curation of inputs and outputs. Production systems exist but with notable limitations in handling novel market conditions or complex multi-factor decisions. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Retail demand forecasting and inventory planning software (e.g., from major POS/ERP vendors) is deployed widely in production, though accuracy varies and human review is standard, especially at the individual store/supervisor level. |
Provide customer service by greeting and assisting customers and responding to customer inquiries and complaints.
44CI 35–52 · exposure 30 · augmentation 75 · importance 4.7/5 · click for rater detail
Provide customer service by greeting and assisting customers and responding to customer inquiries and complaints.
44| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Retail and e-commerce have rapidly deployed chatbots and automated customer service systems; major retailers and online platforms use AI to handle volume of inquiries. Adoption is accelerating in the sector, with human supervisors increasingly managing exceptions rather than frontline volume. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a moderately digitized but physically-anchored sector; AI adoption for customer-facing frontline supervisory roles remains in pilot stages compared to faster-adopting sectors like finance or software. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI systems meaningfully assist supervisors by pre-screening inquiries, summarizing complaints, suggesting responses, and handling routine issues before escalation. This transforms supervisory productivity by freeing them to focus on high-value judgment calls and relationship repair. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools (chat assist, sentiment analysis, scripted response suggestions, CRM summarization) meaningfully help supervisors handle inquiries and complaints faster and more consistently while they remain in control of customer interactions. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI chatbots can handle simple inquiries and greetings, but the task requires judgment on complex complaints, emotional de-escalation, and real-time problem-solving where supervisory authority matters. Even the best systems struggle with novel complaint scenarios and lack the accountability a supervisor carries. |
| Task automatability | claude-sonnet-5 | 2/5 | AI chatbots can handle simple inquiries, but the task as framed for a supervisor role includes in-person greeting, escalated complaint handling, and judgment calls that require physical presence and interpersonal authority not replicable end-to-end today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few hard regulatory barriers exist for AI-assisted or AI-handling basic customer service, though many retailers prefer human supervisors for brand and liability reasons and some customers demand human contact. Organizational friction and reputational preference for human interaction provide modest friction but are not legal or licensing blocks. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but customer preference for human interaction in physical retail settings and the supervisory nature of complaint escalation create moderate organizational friction against full substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Modern chatbot infrastructure and hosting are substantially cheaper than paying a supervisor's fully-loaded wage, especially for high-volume simple interactions. Oversight and integration costs are moderate, making the cost ratio heavily favor AI for this class of task. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI chat support is cheap per interaction, this task bundles physical presence and supervisory escalation handling, so a full AI substitute would still require human backup, keeping blended costs closer to human wages. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed conversational AI (chatbots, phone IVR systems) handles routine customer service in many retail contexts, but error rates on nuanced complaints remain material and most systems still escalate complex issues to humans. Production coverage is broad but with significant limitations on scope. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Deployed chatbots and IVR systems handle routine online/phone inquiries reliably, but in-store greeting and supervisor-level complaint resolution are not performed by production AI systems today. |
Plan and coordinate advertising campaigns and sales promotions and prepare merchandise displays and advertising copy.
42CI 30–55 · exposure 38 · augmentation 75 · importance 4.2/5 · click for rater detail
Plan and coordinate advertising campaigns and sales promotions and prepare merchandise displays and advertising copy.
42| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail is moderately digitized but adoption of AI-driven campaign planning and display coordination remains in the pilot phase in most organizations; large chains may experiment, but widespread production deployment is limited compared to information/finance sectors. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Retail and marketing functions have moderate AI adoption for content generation and campaign planning tools, though pilots are more common than fully autonomous deployment at scale in this supervisory role. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI tools can meaningfully assist supervisors by drafting marketing copy, suggesting design layouts, and generating promotional ideas, thus raising productivity in those components while the supervisor retains oversight and final decision-making authority. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI tools significantly speed up drafting ad copy, generating promotional ideas, and designing visual concepts, giving supervisors strong productivity gains while they retain oversight and final execution. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with generating advertising copy and basic campaign ideation, the task requires strategic judgment about target audiences, competitive positioning, and local market knowledge that current systems handle inconsistently. End-to-end automation with 50% time savings at equal quality is not yet demonstrated at scale in production. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can generate ad copy, campaign concepts, and promotional calendars efficiently, but coordinating displays physically, aligning with local store context, and final campaign decisions still require human judgment and execution. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Retail supervisors' authority to approve promotions and campaigns is often tied to business acumen and sign-off responsibilities; while not legally mandated like licensed professions, organizational policies and the need for human accountability create moderate friction to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates human authorship of ads or displays, though brand consistency and physical store execution create moderate organizational friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI systems for copywriting and design assistance have moderate inference costs, but integration, human review, and iteration overhead remain significant relative to the salary of an entry-level or mid-level supervisor performing these tasks. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI content generation is cheap for copy and creative concepts, but the physical display work and campaign coordination still require paid human labor, keeping overall cost roughly comparable when the full task is considered. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI tools exist for copywriting and basic design suggestions, but reliable production systems for full campaign planning and merchandise display coordination that meet retail standards are limited. Most deployed solutions require substantial human oversight and refinement. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Generative AI tools (copywriting, image generation, marketing platforms) are widely deployed for drafting ads and promotional content, but full campaign planning and physical merchandise display setup are not reliably automated end-to-end in production. |
Examine products purchased for resale or received for storage to assess the condition of each product or item.
33CI 30–35 · exposure 25 · augmentation 50 · importance 4.0/5 · click for rater detail
Examine products purchased for resale or received for storage to assess the condition of each product or item.
33| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail is adopting digital tools broadly, but inspection automation specifically is nascent; most retailers still rely on manual receiving and condition checks. Adoption outside large enterprise retailers is minimal, and ROI barriers limit diffusion. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail sales supervision is a moderately digitized sector but this specific physical inspection task lags behind more office-based AI adoption trends; automated visual inspection is mostly seen in large-scale distribution centers, not general retail stores. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-assisted image flagging or scoring of product condition could help supervisors prioritize inspection and reduce false negatives, moderately improving inspection throughput and consistency while the supervisor retains final decision authority. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered scanning apps and computer vision tools can help flag potential defects or verify counts/conditions, assisting the supervisor's inspection process, though final judgment often remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Visual inspection of product condition can be partially automated using computer vision, but the task requires nuanced judgment about wear, damage, and acceptability thresholds that vary by product type and retailer standards. Current AI struggles with the contextual decision-making needed for end-to-end automation at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Physical inspection of received merchandise for damage, defects, or quality issues requires manual handling and visual/tactile judgment that current AI cannot fully replicate without extensive hardware integration.dd Some visual inspection via cameras exists but is not a general solution for diverse retail goods. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Retail operations are digitizing but remain moderately fragmented; many small and mid-size retailers lack the infrastructure or standardization needed for automated inspection. Liability concerns (missed damage, returns) and lack of regulatory mandate for automation create moderate friction to adoption. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No strict licensing requirement, but liability for missed damaged goods, physical handling needs, and integration with existing receiving workflows create moderate organizational friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Camera systems, AI model inference, and integration infrastructure add non-trivial capital and operational costs. For typical retail supervisors earning $15–20/hour, the all-in cost of automated vision inspection often exceeds the savings from reducing human inspection time, especially for lower-volume facilities. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Setting up camera-based inspection systems for diverse retail inventory requires significant capital investment, likely exceeding the marginal cost of a supervisor performing visual checks during normal receiving duties. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Computer vision systems exist for defect detection in manufacturing, but deployment in retail receiving/inspection workflows remains limited and typically requires human verification. Products available are mostly in pilot or narrow-scope applications rather than production-scale retail inventory assessment. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Computer vision quality inspection systems exist in manufacturing and some warehouse settings, but deployed products for general retail receiving/inspection across varied product types are narrow and not widely adopted by retail supervisors. |
Direct and supervise employees engaged in sales, inventory-taking, reconciling cash receipts, or in performing services for customers.
31CI 30–32 · exposure 25 · augmentation 63 · importance 4.6/5 · click for rater detail
Direct and supervise employees engaged in sales, inventory-taking, reconciling cash receipts, or in performing services for customers.
31| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Retail has moderate digitization and some adoption of scheduling and inventory AI tools, but deployment of AI for direct workforce supervision remains limited; pilots exist but production-scale autonomous supervision is rare. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a relatively low-digitization, high physical-presence sector where AI adoption for direct supervisory roles remains slow, though back-office tools are spreading. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist with reporting dashboards, anomaly detection in sales/inventory, scheduling optimization, and data-driven insights, improving a supervisor's visibility and decision-making, though the human remains essential for interpersonal management and judgment calls. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered scheduling, inventory analytics, and cash reconciliation tools significantly boost a supervisor's efficiency in managing these functions even though the human remains in charge. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI could assist with monitoring inventory, cash reconciliation, and scheduling, the core supervision task—directing employees, coaching, conflict resolution, and real-time personnel decisions—requires human judgment and presence that AI cannot fully replace end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Direct supervision of people requires real-time interpersonal management, motivation, and situational judgment on the floor that current AI cannot perform end-to-end; only sub-components like scheduling or reporting can be automated. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | While no hard legal requirement mandates a human supervisor in most retail contexts, organizational liability, customer expectations, and labor law (duty to maintain safe conditions, handle disputes) create friction; however, these are not absolute prohibitions on automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement, but organizational norms, liability for personnel decisions, and customer/employee expectations of human oversight create meaningful friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools for inventory and reconciliation have modest costs, but the loaded wage of a first-line supervisor is relatively low and their role spans many tasks; automation of only the administrative portions does not approach an order of magnitude cost advantage. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI tools can cut costs for scheduling/reporting subtasks, but a human supervisor is still needed for the core people-management function, so overall cost savings versus a full-time supervisor are modest. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some deployed products exist for inventory tracking and cash reconciliation, but no production system reliably handles the full supervisory role including employee direction, performance management, and customer service oversight at the reliability needed for autonomous deployment. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Workforce management and scheduling software exists and is deployed, but no product actually 'supervises' employees in the interpersonal, in-person sense described here. |
Instruct staff on how to handle difficult and complicated sales.
30CI 25–35 · exposure 25 · augmentation 63 · importance 4.3/5 · click for rater detail
Instruct staff on how to handle difficult and complicated sales.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail remains largely analog in supervision practices; while some chains use e-learning platforms, live supervisor-led coaching on difficult sales remains the norm, with slow sector-wide AI adoption in coaching contexts. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a moderately digitized sector with growing use of AI for training content, but supervisory coaching remains largely human-led with slow AI penetration. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist supervisors by generating sample scripts, flagging common objection patterns, or summarizing training needs from sales data, meaningfully reducing prep time while the supervisor retains instructional delivery and judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can generate training materials, role-play scenarios, and provide guidance scripts that meaningfully support supervisors in coaching staff on complex sales situations. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI could draft instructional materials or case examples, but the task requires real-time responsiveness to nuanced interpersonal dynamics, employee learning styles, and context-specific sales scenarios that AI cannot reliably orchestrate end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Training staff on handling complex, situational sales scenarios requires real-time judgment, relationship context, and adaptive coaching that current AI cannot fully replicate end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Organizational expectations and employee morale depend on trusted, human supervisors delivering real-time coaching and judgment calls; there is also implicit professional liability for poor training outcomes that favors human accountability. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but organizational preference for hands-on managerial coaching and customer-facing judgment creates moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-generated training modules are cheap at scale, but integrating them into a supervisor's workflow, verifying employee comprehension, and handling edge cases still require supervisory oversight, making the all-in cost comparable to partial human instruction. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI training modules are cheap to deploy, the need for human oversight and situational coaching keeps blended costs closer to comparable rather than order-of-magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Chatbots and learning systems can deliver generic sales coaching content, but no deployed product reliably handles the adaptive, one-on-one instruction and behavioral feedback that characterizes effective supervision of difficult sales interactions. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some AI-driven coaching tools and chatbot scripts exist for sales training, but they are narrow and don't reliably handle the nuanced, in-person mentoring required for difficult sales situations. |
Assign employees to specific duties.
30CI 25–35 · exposure 25 · augmentation 50 · importance 4.3/5 · click for rater detail
Assign employees to specific duties.
30| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail remains relatively low in digital sophistication and AI adoption; while large chains use basic scheduling software, the substitution of AI for supervisor judgment on duty assignment remains rare and limited to pilots in advanced retail operations. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a moderately digitized sector with scheduling software adoption, but dynamic task assignment automation is still nascent and not widely deployed at scale. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Scheduling software and AI-driven task-suggestion systems can assist supervisors by flagging conflicts, proposing efficient assignments, and surfacing availability—genuinely useful productivity aids—but the supervisor remains the decision-maker. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-based scheduling and workforce management tools can meaningfully assist supervisors by suggesting optimal assignments based on skills, availability, and demand patterns, improving efficiency while humans retain final control. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Assigning employees to duties requires understanding individual skills, availability, customer needs, and real-time operational context. While AI could suggest assignments based on scheduling data, the task demands judgment about fit and motivation that current systems cannot reliably provide end-to-end without substantial human oversight. |
| Task automatability | claude-sonnet-5 | 2/5 | Scheduling and duty assignment tools can suggest allocations, but the actual decision involves situational judgment, employee relations, and real-time floor conditions that current AI cannot fully handle end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Labor law and union agreements often require that a human supervisor make duty assignments and be accountable for fairness and compliance; scheduling also carries legal liability for wage and hour violations if misconfigured, creating liability barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement, but organizational norms and the need for interpersonal judgment (motivation, conflict resolution, adapting to circumstances) create moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Implementing and maintaining an AI scheduling system, plus the oversight needed to validate and correct assignments, likely costs as much or more than the salary of a supervisor performing this task directly, particularly in smaller retail operations. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Software licensing costs are modest but a human supervisor must still monitor and override assignments in real time, so total cost savings versus a supervisor's judgment are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some workforce management software offers scheduling and task-assignment suggestions, but these are narrow tools requiring significant human refinement and decision-making. No deployed product reliably handles the full socio-technical complexity of retail duty assignment at scale without material human intervention. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Workforce management software (e.g., scheduling systems) exists and is deployed, but they mostly optimize shift scheduling rather than dynamic task/duty assignment on the sales floor, which still requires human oversight. |
Monitor sales activities to ensure that customers receive satisfactory service and quality goods.
28CI 25–30 · exposure 25 · augmentation 63 · importance 4.4/5 · click for rater detail
Monitor sales activities to ensure that customers receive satisfactory service and quality goods.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail adoption of AI monitoring is slow outside large chains, and even then it augments rather than replaces first-line supervisors. Most retailers still rely on human floor presence for service quality assurance, with analytics remaining secondary to management presence. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a moderately digitized sector with growing use of analytics tools, but supervisory roles in physical retail environments adopt AI slowly compared to information/professional services. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist supervisors by flagging underperforming registers, inventory mismatches, or customer wait-time patterns, raising visibility without removing the supervisor's judgment role. This is a meaningful but limited augmentation since supervisors still drive intervention decisions. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-powered dashboards, sales analytics, and customer feedback analysis can meaningfully help supervisors monitor performance and service quality more efficiently, even though the human remains central to enforcement and coaching. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Monitoring sales activities for service quality requires nuanced judgment about customer satisfaction, staff performance, and context-dependent interventions. AI can track transaction data and flag anomalies, but cannot reliably assess 'satisfactory service' or evaluate the human dynamics of customer interaction that current systems struggle with end-to-end. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires physical presence, real-time human judgment, and interpersonal management of staff and customers on a sales floor, which current AI cannot fully replicate end-to-end.assistive tools exist but don't replace the core supervisory function. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant barriers exist: employee privacy and labor law constraints on surveillance, customer expectations that a human supervisor maintains service standards, liability for missed quality issues, and organizational inertia around removing supervisory headcount. Retailers face legal and reputational risk automating away human judgment on customer experience. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement, but organizational reliance on human judgment, customer relations, and staff management creates meaningful friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Surveillance and analytics infrastructure (cameras, software, integration) combined with ongoing false-positive handling and human oversight remains costly relative to paying a supervisor for this work. The all-in cost per monitored shift is not yet significantly lower than human supervision. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While software dashboards are cheap, replicating full supervisory judgment, coaching, and in-person quality control still requires human labor, keeping all-in AI cost comparable or higher when factoring integration and oversight. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While computer vision and transaction analytics exist in retail, no deployed system reliably monitors *customer satisfaction* and service quality holistically across a store or team without human oversight. Most systems require supervisors to validate alerts and interpret context. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some AI-based tools (camera analytics, customer sentiment tracking, sales dashboards) exist to support monitoring, but no deployed product autonomously performs the full supervisory oversight of service quality and goods in production at scale. |
Establish and implement policies, goals, objectives, and procedures for the department.
28CI 25–30 · exposure 25 · augmentation 63 · importance 4.1/5 · click for rater detail
Establish and implement policies, goals, objectives, and procedures for the department.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail sectors show slower AI adoption overall; policy establishment is a high-touch governance function where organizations remain cautious and prefer human supervisors to retain formal accountability. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail management is a moderately digitized but still largely people-driven sector, with AI adoption for managerial policy-setting still nascent and mostly limited to pilot tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by generating policy drafts, analyzing competitive benchmarks, and summarizing best practices, helping supervisors work faster while they retain final authority and implementation responsibility. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully assist supervisors by drafting policy language, summarizing best practices, and suggesting procedures, significantly speeding up the planning portion of the task. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft policies and procedures using templates and organizational data, establishing them requires strategic judgment, stakeholder input, and accountability for outcomes—tasks that demand human decision-making and organizational authority that current AI cannot fully assume. |
| Task automatability | claude-sonnet-5 | 2/5 | Policy drafting text can be AI-assisted, but establishing and implementing actual departmental policy requires contextual judgment, stakeholder buy-in, and accountability that current AI cannot autonomously execute end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | A licensed or formally authorized manager must legally establish and take responsibility for departmental policies; liability, regulatory compliance, and organizational hierarchy create strong barriers to AI-driven automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No formal licensing requirement, but organizational accountability, managerial authority, and employment law responsibilities create meaningful friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI-assisted drafting may reduce initial time cost, but the human supervisor remains essential for approval, accountability, and implementation oversight, making total cost savings marginal compared to supervisory wages. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cheaply generate draft text, but the full task involves implementation, communication, and enforcement that still requires paid human management time, keeping overall cost comparable to human labor. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed products reliably perform end-to-end policy establishment and implementation for retail departments; AI tools can assist with drafting but lack the organizational standing and accountability to implement policies autonomously. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | No deployed product independently sets and implements retail department policy; at best AI tools help draft policy documents that a human reviews and enacts. |
Plan budgets and authorize payments and merchandise returns.
28CI 25–30 · exposure 25 · augmentation 50 · importance 3.9/5 · click for rater detail
Plan budgets and authorize payments and merchandise returns.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail remains a laggard sector in AI adoption, with limited production deployment of autonomous budget and payment systems. Most automation in this domain is piloting or process-support (dashboards, alerts) rather than full decision authority transfer. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a moderately digitized sector with POS/ERP automation but AI-driven budget planning and return authorization agents are still in early pilot stages, not widespread production use. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist supervisors by analyzing sales trends, flagging return anomalies, or suggesting budget allocations, improving decision quality. However, the human supervisor typically remains in the loop for final authorization, so augmentation is meaningful but not transformative. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can meaningfully assist with sales forecasting, budget modeling, and flagging anomalous return patterns, improving supervisor efficiency without replacing final judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Budget planning involves complex judgment about inventory, sales forecasts, and staffing that require contextual reasoning. Payment authorization and return processing have rule-based components AI could handle, but the judgment-intensive planning aspects (resource allocation, variance analysis, strategic trade-offs) remain difficult for current systems to execute end-to-end with equal quality. |
| Task automatability | claude-sonnet-5 | 2/5 | Budget planning requires contextual judgment and authorizing returns/payments involves policy exceptions and interpersonal decisions that current AI cannot fully replace, though it can assist with data aggregation and rule-based approvals.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Financial authorization and budget approval typically involve explicit liability and fiduciary responsibility that companies require a named human supervisor to sign off on, backed by compliance and audit trails. Regulatory expectations and internal controls around payment authorization create meaningful barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement, but organizational policy typically requires managerial sign-off for financial authorizations and budget decisions, creating moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | The integration overhead, oversight requirements, and error-handling costs for AI systems to handle financial authorization are substantial relative to a first-line supervisor's wage. Current systems still require significant human review and exception handling, offsetting labor savings. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Rule-based systems for authorizing standard returns are cheap, but full budget planning still requires human oversight and integration costs that keep AI costs comparable to or above supervisor time savings. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | While transaction processing for returns can be partially automated, AI systems today lack reliable deployment for full budget planning authorization in retail supervisory roles. Existing tools assist with data compilation but do not independently make or authorize budget and payment decisions at production scale with acceptable error rates. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some POS and ERP systems automate routine return authorization under set thresholds, but comprehensive budget planning and exception-handling by AI in production retail settings is not yet demonstrated broadly. |
Confer with company officials to develop methods and procedures to increase sales, expand markets, and promote business.
26CI 20–32 · exposure 20 · augmentation 63 · importance 3.7/5 · click for rater detail
Confer with company officials to develop methods and procedures to increase sales, expand markets, and promote business.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While retail is digitizing, strategic conference-level decision-making remains heavily human-centered. Adoption of AI for this specific task is minimal; companies rely on human supervisors and managers for business development strategy. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Retail and business strategy functions are adopting AI analytics and forecasting tools at a moderate pace, though the specific collaborative conferring activity is not yet automated in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by generating market analysis, competitor insights, and procedure drafts that supervisors then refine and present. This support can improve the quality and speed of strategy development, though the core conference and decision-making remain human-led. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can meaningfully augment this task by providing sales trend analysis, market research summaries, and scenario modeling that supervisors bring into discussions with officials. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task requires strategic judgment, stakeholder negotiation, and organizational knowledge. While AI can assist with market analysis and procedure documentation, the collaborative conference with officials and contextual decision-making about sales strategy remain fundamentally human activities that cannot be fully automated to the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 2/5 | This is a collaborative, judgment-heavy strategic task involving negotiation, relationship dynamics, and organizational context that current AI cannot conduct end-to-end, though it can support analysis feeding into these conversations. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strategic business decisions typically require human accountability, organizational authority, and legal sign-off from management. The human supervisor's role in representing company interests and making binding decisions creates meaningful barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing barrier, but organizational trust, accountability for strategic decisions, and the inherently interpersonal nature of executive conferring create real friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI systems can provide supporting analysis and drafting, but the core activity—conferring with officials and synthesizing strategic direction—requires human judgment that is difficult to cost-competitively replace. Oversight and integration costs would approach or exceed the value of partial automation. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Replacing the human conferring process would still require significant human time for stakeholder alignment and trust-building, so AI mainly supplements rather than substitutes at meaningfully lower cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably conducts strategic business conferences with company officials or independently develops organizational sales procedures at production scale. This requires autonomous negotiation and organizational decision-making beyond current AI capabilities. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | No deployed product autonomously confers with executives to co-develop sales strategy; AI tools exist for generating data-driven recommendations but the interpersonal conferring and decision-making remains human-led. |
Hire, train, and evaluate personnel in sales or marketing establishments, promoting or firing workers when appropriate.
25CI 25–25 · exposure 25 · augmentation 50 · importance 4.2/5 · click for rater detail
Hire, train, and evaluate personnel in sales or marketing establishments, promoting or firing workers when appropriate.
25| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While retail and marketing organizations use AI for candidate sourcing, actual adoption of AI for making hiring, promotion, and firing decisions remains limited and cautious. Legal risk and employee relations concerns slow deployment; most adoption remains in the pilot or advisory phase rather than autonomous decision-making. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a moderately digitized sector with some AI adoption in recruiting tools, but frontline supervisory HR functions like firing remain largely untouched by AI in practice. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist supervisors by surfacing candidate screening summaries, flagging performance trends from sales data, and generating training recommendations, improving decision quality and speed. However, the human supervisor remains firmly in the loop for all consequential personnel decisions. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can meaningfully assist with drafting job postings, screening applicants, structuring training content, and summarizing performance data, improving supervisor efficiency without replacing their judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with resume screening, job posting, and preliminary candidate evaluation, hiring decisions remain heavily dependent on human judgment regarding cultural fit, interview assessment, and personnel management. The full end-to-end task—especially evaluation, promotion, and firing decisions with legal and interpersonal complexity—cannot be reliably automated to meet the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can assist with resume screening, drafting training materials, and performance metrics, but the core managerial acts of hiring decisions, firing, and evaluation require human judgment, interpersonal assessment, and accountability that current AI cannot fully replace. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant legal barriers exist: employment law, discrimination law, and liability for wrongful termination create regulatory friction. Many jurisdictions require human accountability and documentation for hiring and firing. Organizations also face reputational and morale risks from perceived algorithmic bias in personnel decisions, strongly discouraging full automation. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Hiring and firing carry significant legal, HR compliance, and liability requirements (discrimination law, labor law) that generally mandate human decision-making and accountability. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI applicant tracking and screening tools are relatively cheap, but the cost of deploying AI for sensitive personnel decisions (evaluation, firing) with required human oversight often approaches or exceeds the marginal cost of a supervisor spending time on these tasks. Integration and governance overhead remain high. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI screening tools are cheap, the overall task still requires substantial human oversight, interviews, and legal review, so total cost savings versus a human supervisor are modest. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI tools exist for candidate sourcing and resume filtering, but no deployed product reliably performs the full hiring, training, evaluation, and termination cycle. These human-centric decisions require judgment that current systems struggle with at scale, and legal/liability concerns limit production deployment for final hiring and firing decisions. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Products like applicant tracking systems and AI-assisted performance review tools exist and are used in production, but full end-to-end hiring/firing decision-making by AI is not deployed due to legal and practical constraints. |
Establish credit policies and operating procedures.
25CI 25–25 · exposure 25 · augmentation 63 · importance 4.1/5 · click for rater detail
Establish credit policies and operating procedures.
25| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail and small-business sectors show moderate AI adoption, but policy-establishment tasks are strategic and infrequent; most organizations rely on established human supervisors or external consultants rather than automated systems, limiting measured adoption velocity. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail management is a moderately digitized sector but strategic policy-setting tasks like this see limited AI deployment compared to more transactional retail functions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist supervisors by generating policy drafts, analyzing compliance requirements, and summarizing best practices, improving the productivity of the human drafter. However, the human supervisor must review, customize, and authorize the final policy, limiting augmentation to the front-end generation phase. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can research industry benchmarks, draft policy language, and model risk scenarios, meaningfully speeding up a supervisor's policy development process while they retain final decision authority. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft policy templates and procedures based on data, establishing credit policies requires strategic business judgment, legal compliance expertise, and organizational context that cannot be fully automated end-to-end with equal quality today. Current AI lacks the domain-specific accountability and authority to set binding credit policies. |
| Task automatability | claude-sonnet-5 | 2/5 | Setting credit policy involves judgment calls about risk tolerance, legal compliance, and business strategy specific to a store, which AI can inform but not autonomously decide with full accountability. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Credit policies are subject to regulatory oversight (Fair Credit Practices, state lending laws) and must be signed off by authorized management; liability and legal compliance requirements mean organizations cannot delegate policy establishment to unsupervised AI systems. Human supervisors remain legally and fiducially responsible. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Credit policies carry legal and financial liability (fair lending, consumer protection laws) so a responsible manager must approve and be accountable, creating a strong barrier to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Policy establishment is infrequent and high-stakes, making end-to-end AI automation economically unfeasible. AI assistance (drafting, analysis) may reduce labor cost marginally, but cannot substitute for supervisor time and expertise, keeping all-in costs comparable to or higher than human effort. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Because this is an infrequent, high-stakes strategic task requiring managerial sign-off, AI assistance saves some drafting time but the human review and liability oversight keep costs comparable to manual work. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed AI product reliably establishes complete credit policies in production; tools exist for drafting policies or analyzing compliance, but organizations require human supervisors to review, customize, and officially adopt policies. This remains a human-led process with AI as a support tool. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | No deployed product autonomously establishes a retail store's credit policies end-to-end; existing tools only assist with drafting or benchmarking against industry norms. |
Perform work activities of subordinates, such as cleaning and organizing shelves and displays and selling merchandise.
24CI 13–35 · exposure 13 · augmentation 38 · importance 4.2/5 · click for rater detail
Perform work activities of subordinates, such as cleaning and organizing shelves and displays and selling merchandise.
24| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail adoption of AI agents for supervisory tasks remains nascent; most stores still rely on human supervision, though some larger chains pilot inventory automation and chatbots for basic customer queries. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a lower-digitization, physically-oriented sector where AI adoption for manual merchandising and floor sales tasks remains minimal and pilot-stage at best (e.g., limited robotic shelf-scanning trials). |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist supervisors by automating shelf inventory audits, suggesting product placement, and handling routine customer inquiries, thereby raising supervisor productivity without fully replacing their authority and interpersonal oversight. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with inventory tracking, restocking alerts, or sales scripts that inform supervisors, but it offers little direct help with the physical act of organizing shelves or hands-on selling. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While some components like inventory monitoring and sales can be partially automated, physical shelf cleaning and organizing requires embodied robotics not reliably deployed in retail today, and selling merchandise involves human judgment and customer rapport that current AI struggles with at equal quality. |
| Task automatability | claude-sonnet-5 | 1/5 | This task requires physical manipulation of shelves, displays, and merchandise plus in-person customer interaction, none of which current AI systems can perform end-to-end without robotics far beyond commercial deployment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Retail supervisors often hold explicit organizational authority over staff (quasi-managerial role), customer preference for human sales interaction, and physical store operations require authorized personnel; however, these are organizational rather than legal barriers. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates a human for these tasks, but the physical dexterity and customer-facing nature create practical friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Physical automation (robotics for shelf work) remains capital-intensive and expensive per task-unit compared to paying a supervisor minimum wage; software for sales assistance is cheaper but covers only partial workflow. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for physical shelf-stocking and in-person sales, so any hypothetical robotic solution would be far costlier than human retail labor today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some retail automation exists for inventory tracking and basic customer service, but no mature product reliably performs end-to-end shelf management (physical manipulation, organizational decision-making) or authentic sales engagement at scale in real retail environments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs physical shelf organization or in-person retail selling; this remains firmly in the physical/manual labor domain unaddressed by current AI products. |
Enforce safety, health, and security rules.
18CI 11–25 · exposure 13 · augmentation 50 · importance 4.0/5 · click for rater detail
Enforce safety, health, and security rules.
18| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Retail is moving toward monitoring cameras and mobile tools, but actual automated enforcement of rules (rather than flagging for human review) remains rare in production. Most adoption remains in logging and alerting, not autonomous correction. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Retail is a moderately digitized sector with some AI-based surveillance/loss-prevention tools, but active rule enforcement remains largely manual and adoption for this specific function is slow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered monitoring dashboards and real-time alerts on safety violations can help supervisors prioritize and respond faster, but the core task of coaching staff and applying judgment on discipline remains human-driven. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered surveillance, anomaly detection, and compliance checklists can help supervisors identify issues faster, though the human must still act on and enforce the rules. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Enforcement of safety, health, and security rules requires contextual judgment, discretion in discipline, and handling of non-compliant individuals. While monitoring systems can flag violations, the interpersonal negotiation and corrective action decisions are difficult for AI to execute reliably without human oversight. |
| Task automatability | claude-sonnet-5 | 1/5 | Enforcing rules requires physical presence, real-time observation of employee and customer behavior, and situational judgment/authority that current AI cannot execute end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Enforcement of workplace rules carries legal liability risk; many jurisdictions require a human supervisor to witness, document, and sign off on disciplinary actions. Customer-facing contexts also create expectation that a human representative enforce policy. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Enforcement of safety/health/security rules often involves legal responsibility, liability, and managerial authority that organizations assign to accountable human supervisors. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Monitoring infrastructure (cameras, sensors, integrations) plus AI inference and oversight overhead can be substantial; the savings relative to a supervisor's wage are modest and offset by implementation and false-positive triage costs. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Monitoring tools (cameras, sensors) can be cheap, but actual enforcement still requires a human supervisor's presence and authority, so overall cost savings are limited. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Current AI can detect some safety violations via video or sensor data, but deployed products lack robust real-time integration with enforcement protocols and cannot reliably conduct the interpersonal conversations needed to correct behavior or determine proportionate responses. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously enforces workplace safety/security rules on a sales floor; at most AI provides camera-based alerts, not enforcement itself. |
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.