The 5 Discovery Questions for AI in Any Business

Every AI initiative starts the same way: someone asks, “Where should we use AI?”

The answer isn’t in a vendor pitch deck. It’s in your operations — the places where work slows down, information gets stuck, and humans spend time doing things that machines should handle.

These five questions surface the hidden friction in any operation. Every point of friction is an AI opportunity.

Question 1: Where Does Information Stop Moving?

Information wants to flow. When it stops, something’s wrong.

Look for places where information sits in one system waiting for a human to move it to another:

• The email that needs to become a CRM entry
• The invoice that needs to become an ERP record
• The customer complaint that needs to become a quality ticket
• The sales order that needs to become a production schedule

Every time information stops moving, you have a person — usually an expensive one — acting as a human integration layer between systems.

Example: Professional Services Firm

A consulting firm receives project change requests via email. A coordinator reads each email, determines scope impact, updates the project plan in Microsoft Project, adjusts the forecast in the financial system, and notifies team members via Slack.

• Average time per change request: 45 minutes
• Average change requests per week: 22
• Total: 16 hours/week — 40% of an FTE — moving information between systems

The AI opportunity: An agent monitors the inbox, extracts change request details, assesses scope impact, updates project management and financial systems, and routes notifications. Human involvement drops to approval-only for material changes. The 45-minute process becomes 2 minutes of review.

Example: Healthcare Clinic

A specialty clinic receives referrals via fax. A staff member reads each fax, enters demographics into the PMS, checks insurance eligibility, schedules the appointment, and sends confirmation to the referring physician.

• Average time per referral: 25 minutes
• Average referrals per day: 35
• Total: nearly 15 hours daily — two FTEs doing data entry that creates no clinical value

The AI opportunity: Document extraction from faxed referrals, automated eligibility verification, intelligent scheduling, and automated confirmation. The two FTEs get redeployed to patient care coordination where human judgment matters.

Example: Manufacturing

A contract manufacturer receives EDI orders. 23% trigger exceptions — wrong part numbers, missing instructions, quantities not matching contract terms.

• Average time per exception: 52 minutes
• At 150 orders/week: 18 exceptions costing 15.6 hours of labor — plus production delays

The AI opportunity: An agent validates incoming orders against contract terms, resolves unambiguous issues automatically (like updating obsolete part numbers), and drafts communications for ambiguous ones. Human involvement shifts from data correction to relationship management.

What this question reveals: Every information stoppage is missing integration. AI agents don’t just move data faster — they understand context, handle exceptions, and make decisions that traditional integration can’t.

Question 2: Where Does Someone Translate Between Two Systems?

This is different from information stopping. This is information moving, but requiring a human interpreter.

Look for people who read output from one system and create input for another — not just copying, but translating:

• The analyst who reads sales data and produces a demand forecast
• The planner who reads customer orders and creates a production schedule
• The accountant who reads bank statements and categorizes transactions

Translation work is expensive because it requires understanding both systems. It’s also error-prone because it depends on human judgment applied repetitively.

Example: Wholesale Distribution

A distributor’s buyers receive vendor price increases via PDF. They read each PDF, identify affected SKUs, calculate margin impact, determine which prices need adjustment, update the pricing system, and notify sales.

• A single increase affecting 200 SKUs: 4-6 hours to process
• 15-20 increases per month: 60-120 hours monthly on a purely mechanical task

The AI opportunity: Document extraction, automated SKU matching, margin analysis, price adjustment recommendations based on competitive positioning, and system updates pending approval. The buyer’s role shifts from data processing to strategic pricing decisions.

Example: Commercial Insurance

An underwriter receives submissions with loss runs, financial statements, and property schedules. Analysis of a standard commercial property submission takes 2-3 hours. Complex accounts take 6-8 hours.

A team of 10 underwriters processes 400 submissions/month, spending 60% of their time on document analysis rather than risk evaluation.

The AI opportunity: Automated data extraction, comparison to underwriting guidelines and historical loss patterns, preliminary rate calculation, and exception flagging. Underwriters spend time on risk evaluation and broker relationships — work that actually requires their expertise.

Example: Recruiting

Initial sourcing for a technical role takes 8-12 hours. Specialized positions can take 20+ hours before the first qualified candidate reaches the hiring manager.

The AI opportunity: An agent reads the requisition, generates search strategies across platforms, evaluates candidates against requirements, and produces a ranked shortlist with reasoning. The recruiter’s value shifts from searching to evaluating cultural fit and managing the candidate relationship.

What this question reveals: Translation work is knowledge work trapped in a processing task. AI agents free the knowledge worker to apply expertise to decisions rather than data manipulation.

Question 3: Where Does Someone Make the Same Decision Repeatedly?

Decisions that repeat are decisions that can be learned.

Not all decisions are candidates for automation. Strategic decisions, novel situations, and ethical dimensions should stay with humans. But look for decisions that:

• Happen frequently
• Have clear inputs
• Produce consistent outputs most of the time
• Occasionally require escalation

Example: Credit and Collections

A credit manager reviews new customer applications — check business credit, verify references, review financials, apply credit policy, assign limits.

• 50-80 applications per month, 30-60 minutes each
• 80% are straightforward approvals or denials based on clear criteria

The AI opportunity: An agent auto-approves the 60% that clearly qualify, auto-declines the 15% that clearly don’t, and routes the 25% requiring judgment with analysis and a recommendation. The credit manager’s volume drops 75%, but the decisions they make actually need human judgment.

Example: Maintenance Scheduling

A supervisor reviews work orders each morning and assigns technicians based on skill, location, priority, and availability. Daily assignment takes 45-60 minutes — the same decision repeated with changing constraints.

The AI opportunity: An agent monitors work orders, technician status, parts inventory, and customer SLAs, then produces an optimized daily schedule. The supervisor reviews and adjusts rather than building from scratch. Emergencies trigger automatic rescheduling with impact analysis.

Example: Inventory Replenishment

A buyer reviews inventory levels weekly for 10,000 SKUs. Most decisions are formulaic — when stock drops below X, order Y. But the buyer makes 10,000 micro-decisions because the system doesn’t integrate all relevant inputs. This takes 20-30 hours weekly.

The AI opportunity: An agent monitors inventory continuously, predicts demand, generates replenishment orders within approved parameters, and flags only the exceptions — new products, stockout risks, vendor issues, unusual demand patterns.

What this question reveals: Repeated decisions are training data. Every time a human makes a decision that works, that’s a labeled example the AI can learn from.

Question 4: Where Is the Exception Path Longer Than the Happy Path?

Every process has two versions: what happens when everything goes right, and what happens when it doesn’t.

The happy path is documented. The exception path is improvised — it lives in the heads of your experienced employees. In most operations, the exception path consumes more resources because exceptions are handled by your most expensive people using your least efficient processes.

Example: E-Commerce Returns

• Happy path: Customer initiates return, prints label, ships item, receives refund. Time: 30 seconds of labor.
• Exception path: Item arrives damaged, wrong item returned, customer disputes refund, or 91 days on a 90-day policy. Each exception: 15-45 minutes of labor.

At 15% exception rate on 10,000 monthly returns, that’s 1,500 exceptions consuming 375-1,125 labor hours — more labor than all happy-path returns combined.

The AI opportunity: An agent triages exceptions, gathers context, applies policy rules, resolves unambiguous cases, and presents ambiguous ones with full context and a recommendation. Exception handling time drops 60-70%.

Example: Accounts Payable

• Happy path: Invoice matches PO and receiving record, gets paid. Time: 2 minutes.
• Exception path: Quantity differs, price differs, PO doesn’t exist, item never received. Each exception: 20-45 minutes.

At 20-30% exception rate on 3,000 monthly invoices: 600-900 exceptions consuming 200-675 hours monthly.

The AI opportunity: An agent investigates — checking receiving records, POs, email, contract terms, and historical patterns. It resolves the 70% with clear answers. AP handles only genuinely ambiguous situations.

Example: Project Delivery

• Happy path: Project proceeds on plan, deliverables accepted, invoice sent.
• Exception path: Scope changes, dependencies slip, rework needed, resources reassigned. The PM spends their time managing exceptions to the project.

The AI opportunity: An agent monitors project data, identifies variances before they become crises, models impact of changes, recommends corrective actions, and drafts client communications. PMs spend time on relationships and leadership rather than spreadsheet wrestling.

What this question reveals: Exception handling is where your best people spend their worst hours. AI agents handle the investigation and resolution work that buries your team.

Question 5: What Do New Employees Get Wrong in the First 90 Days?

This question surfaces tribal knowledge — institutional understanding that lives in experienced employees’ heads.

New employees make predictable mistakes. They don’t know the customer who always needs rush handling. They don’t know the vendor who ships early. They don’t know the ERP workaround. This knowledge takes months to acquire and walks out the door when people leave.

Example: Customer Service

A new rep doesn’t know that Customer ABC’s orders go through a custom workflow from a pricing agreement negotiated three years ago. They process the order normally. It invoices wrong. A senior rep has to fix it.

The AI opportunity: An agent recognizes customers with special handling and surfaces context before processing. Tribal knowledge becomes system knowledge, available to every rep on every interaction.

Example: Sales

A new rep doesn’t know that Prospect XYZ has been quoted three times, always objects on delivery lead time, and only closes when the VP of Operations demonstrates manufacturing capacity.

The AI opportunity: An agent provides deal intelligence — prior quotes, historical objections, win/loss analysis, similar deals that closed, and recommended approach based on pattern matching across the entire sales history.

Example: Operations

A new production planner doesn’t know that Machine 4 runs 8% slow, Thursday second shift has attendance issues, and three customers get priority regardless of what the system says.

The AI opportunity: An agent incorporates all known constraints — documented and undocumented — into scheduling. When an experienced planner overrides the schedule, the agent learns the constraint. Tribal knowledge accumulates in the system rather than in individual heads.

What this question reveals: Tribal knowledge is AI training data you already have. Every override, every exception, every “actually, we do it this way” is a labeled example of how your business actually operates versus how your systems think it operates.

From Questions to Action

These five questions surface 80% of the AI opportunities in any operation. But finding opportunities isn’t the same as capturing them.

The next step is prioritization:

• Which opportunities have the highest volume?
• Which have the clearest data?
• Which produce the most measurable outcomes?
• Which can you start with minimal infrastructure?
• Which build toward strategic capability?

These 5 questions are the first chapter of The Operator’s AI Playbook — a 45,000-word implementation manual with 15 templates. It walks through discovery-to-deployment: how to quantify each opportunity, design agent architectures, build the knowledge base, and sequence implementation so each phase funds the next.

Get it at joshuaschultz.com/ai-playbook for $24.99.

Your AI opportunities are already there. These questions make them visible.