Discovery field guide / Industry
How to find AI opportunities in Financial Services & Banking
Where should discovery start?
Follow one customer or transaction journey across relationship management, operations, and risk. Look for repeated evidence gathering, inconsistent handoffs, and avoidable rework before proposing automated financial decisions. Start with an operational problem a business owner can verify. Separate assistance with preparing a file from decisions about credit, fraud, access, or customer treatment, which require a different level of governance and scrutiny.
The shared framework
Fourteen stages, from objective to portfolio.
Every guide uses this sequence to connect a business problem to an accountable investment. Revisit earlier stages when new evidence changes the assessment.
Stage 1: Business Objective
Name the outcome the organization is trying to move, so every later idea has something to be measured against.
Stage 2: Department
Locate the team that owns the objective and the work that feeds it.
Stage 3: Process
Describe how the work actually runs today: steps, handoffs, systems, and exceptions.
Stage 4: People
Capture who performs and depends on the work, and what they know that no document records.
Stage 5: Problem
State the friction in plain terms: delay, rework, error, cost, or risk, with evidence.
Stage 6: Opportunity
Frame a specific improvement that addresses the problem, not yet tied to any technology.
Stage 7: AI/Automation Fit
Compare AI, RPA, IDP, workflow automation, process redesign, and no technology change.
Stage 8: Value
Estimate benefit in the objective's own terms, with assumptions written down.
Stage 9: Feasibility
Check data, systems, skills, and change capacity needed to deliver.
Stage 10: Risk
Surface compliance, quality, security, and adoption risks and the controls they need.
Stage 11: Scoring
Rate each opportunity on the same dimensions so comparisons are consistent and explainable.
Stage 12: Prioritization
Sequence opportunities by score, dependencies, and capacity, with reasons people can challenge.
Stage 13: Investment
Match approved effort to the sequence; people with budget authority decide what gets funded.
Stage 14: Portfolio
Track opportunities over time as a living set: delivered, in flight, waiting, and retired.
Start with one observable journey
Choose a bounded service journey within one product and jurisdiction, such as resolving an incomplete onboarding file. Compare an ordinary case, an escalated case, and a customer complaint. Map who owns the outcome and which records justify each decision.
Where to look across departments
Onboarding and servicing operations
Trace missing documents, repeated customer requests, and handoffs between the relationship team and operational reviewers. Determine whether staff are interpreting variable evidence or compensating for inconsistent identifiers and disconnected systems. A data-quality or workflow repair may remove the cause without adding model risk.
Risk, compliance, and reporting preparation
Look at how teams assemble evidence for a review, reconstruct a transaction history, and explain exceptions in internal reports. Keep preparation distinct from an approval or regulatory conclusion. Record the provenance and effective date of each source rather than assuming all available customer data may be combined.
Who to interview and what each person knows
Frontline staff and case reviewers
Ask a relationship manager or service employee to reconstruct a delayed case, then interview the operations reviewer who accepted or rejected it. Include a complaint handler to find harms not visible in throughput measures. Establish whether the delay came from investigation, customer waiting, or a required control.
Accountable control and technology owners
Include compliance, privacy, information security, model-risk or equivalent governance, and the product-system owner. Ask who determines the applicable review path for the proposed use. A technology label does not decide whether a bank's model, outsourcing, or operational-resilience requirements apply.
What to ask: follow the evidence
What evidence made the case decision-ready?
Ask which documents were missing, which facts conflicted, and who resolved the conflict. What information was available at the time? Which review steps must remain independent? Use an actual exception to distinguish administrative completeness from a judgment about the customer's eligibility or risk.
How would a customer challenge an error?
Ask how an incorrect summary, classification, or recommendation would affect service, account access, or a decision. Who notices and corrects it? Capture recordkeeping, explanation, escalation, and complaint-handling needs before estimating any benefit from faster handling.
Signals of opportunity, and reasons to look elsewhere
Recurring preparation with preserved evidence
A repeatable file-preparation task with accessible records and a qualified reviewer is a stronger starting point than an ambition to automate underwriting. Check whether disagreement is about missing facts or contested judgment. Historical decisions alone are not a trustworthy ground truth if the process itself is under question.
Queue friction with an identifiable cause
Repeated case returns for the same missing evidence may justify a completeness check. Long queues caused by required independent reviews or staffing shortages need a different response. Track rework and correction across teams so a faster first step does not create a larger downstream review burden.
Industry-specific risks and constraints
Consequential customer treatment
Credit, account access, financial crime, and similar decisions can have serious effects and jurisdiction-specific duties. Involve qualified owners to assess fairness, explanation, oversight, and contestability. A high expected benefit cannot waive those constraints, and a successful internal summarization trial does not approve decision automation.
Sensitive records and third parties
Check permitted data use, confidentiality, retention, provider arrangements, and resilience obligations with your institution's owners. Validate access at the customer and product level. Test misleading document content and missing records without exposing real customer data unnecessarily.
How to evaluate what you find
Replay cases without hindsight
Use permissioned examples with the evidence available before the original review. Include conflicting documents, unusual customer circumstances, and missing information. Compare source fidelity, omissions, reviewer effort, and escalation behavior; do not grade a polished answer against the final decision alone.
Evaluate the complete controlled service
Include retained independent review, audit evidence, corrections, monitoring, and vendor dependency in feasibility and cost. Establish a business acceptance owner and governance checkpoints. Test whether results remain useful across the customer populations and products actually included in the proposed scope.
How to compare, prioritize, and invest
Sequence preparation before decision authority
Consider a bounded, reviewer-controlled evidence task where failure is detectable and reversible. Fund identity, data-quality, or integration prerequisites first. Keep higher-consequence customer decisions in a separately governed workstream rather than extending a low-risk pilot by convenience.
Compare controlled outcomes
Rank candidates by better decision-ready evidence, less rework, service quality, and delivery capacity after control costs. Define stop conditions for unsupported assertions or adverse customer treatment. Business and risk owners should approve expansion together and record why competing opportunities were deferred.
Turn observations into a comparable opportunity record
Describe the current process before naming a solution. Record the trigger, owner, input, steps, handoffs, systems, output, and exception path. Separate what you observed from what someone recalled and what you still need to verify. A vivid interview story is a useful lead, not proof that every case behaves the same way. Validate it against a sample of ordinary work and difficult exceptions, with the limits of that sample made explicit.
Measure active work separately from waiting. Note the observation period, case mix, frequency, and source of each estimate. Include the receiving team's effort, checking, correction, and escalation so a local improvement does not merely transfer work. Describe the outcome in operational terms before attaching a monetary value. Capacity released is not automatically a budget saving; explain what can actually change and which manager has authority to make that change.
Compare interventions before selecting a tool
If a rule is explicit and the input is structured, test ordinary workflow automation first. If the task repeats a stable interface action, examine RPA and an integration alternative. If the difficulty is reading documents, consider IDP with source checks. AI may help where language or variable evidence requires interpretation, but its output still needs a task-specific acceptance standard. Process redesign or no technology change may be the right answer when policy, ownership, or an unnecessary step causes the problem.
Write down why each alternative would or would not address the observed cause. Do not compare a carefully controlled AI proposal with an artificially weak version of the current process. Include data repair, access approval, integration, training, review, and upkeep in each feasible option. An unavailable prerequisite is a reason to sequence work or pause, not a detail to hide inside a feasibility score.
Agree on a decision, not just a demonstration
Before a trial, name the business owner, evaluator, permitted data, review boundary, and person who can stop it. Define acceptable outputs and failures that must trigger escalation. Use examples held out from development, including missing information and uncommon cases. Compare the complete task with the baseline, including human verification and downstream consequences. Preserve the evidence behind the result so another owner can challenge the conclusion.
Keep value, feasibility, risk, and evidence confidence visible separately. A candidate with an attractive benefit estimate is not ready if its owner cannot obtain the data or operate the controls. Record the decision to investigate, pilot, wait, reject, or retire, with a reason and a condition for revisiting it. After delivery, compare observed results with assumptions rather than leaving the original estimate in the portfolio as fact.
Use the evaluation guide for the evidence standard and the prioritization guide to compare qualified candidates. Your leaders make funding decisions; a score or a discovery tool supplies structured evidence, not investment authority.
Continue through the discovery framework
Browse all departments and industriesRelated practice notes
For focused interview techniques, examples, and scoring detail alongside this field guide:
Qualify a banking workflow in IntakeOS
Bring a sanitized journey, the accountable owner, and the control boundary. Explore how IntakeOS can capture the evidence and constraints for an opportunity review; your institution retains governance, approval, and investment decisions.
IntakeOS supports structured interviews, qualification, and comparison; it does not make funding decisions or replace professional review. Learn about VARA, review the trust page, or explore the platform before sharing sensitive information.
Explore discovery in an IntakeOS demo