The capability transfer test: will this firm make you self-sufficient or park billable staff

Consulting firms make money two ways. Some earn fees by solving a problem and leaving behind a team that can handle the next one. Others earn fees by becoming load-bearing, so that removing them breaks something. Both kinds will tell you they "build client capability" during the sales process, because by now everyone knows that is what buyers want to hear. The difference shows up in documents, staffing plans, and contract language, and you can detect it before you sign.

Capability transfer carries a suggested weight of 15% on the seven-criterion rubric described on the main guide, and it deserves every point for a credit union. You are probably not going to staff a ten-person AI team. You need three or four people who can operate what gets built, evaluate the next vendor pitch without a chaperone, and answer an examiner's questions about a system without phoning a consultant at $400 an hour. A firm that leaves you dependent has not reduced your risk; it has become your risk.

The economics of dependence, stated plainly

Understand the incentive before judging the behavior. A firm billing two full-time consultants at blended rates earns roughly $60,000 to $90,000 per month from your account. If those consultants become your de facto AI operations team, that revenue continues indefinitely with near-zero sales cost. Teaching your staff to replace them destroys it. So capability transfer is not a feature a firm adds; it is a business model a firm chooses, and firms that chose the other model cannot fake it for long once you know where to look.

This is not an accusation against large firms specifically. Some global firms run genuine enablement practices, and some boutiques quietly staff themselves into permanence. The test is structural, not brand-based. Run it on everyone, including any firm this site's publisher would benefit from you hiring.

Five things to demand during the sales process

1. Named training deliverables in the statement of work. Not "knowledge sharing throughout the engagement," which is a vibe, but line items: "two half-day workshops for lending operations staff on prompt evaluation, delivered weeks 3 and 6, with materials your team retains and may reuse." If training appears in the sales deck but not in the deliverables table, it does not exist. Ask the firm to point to the deliverable language, which is question eight on the interview questions page.

2. Documentation your staff can actually use. Ask for a redacted runbook or admin guide from a past engagement, then read one page. Material built for capability transfer explains why decisions were made and how to change them safely. Material built for billing continuity is a screenshot inventory that only makes sense with a consultant in the room. The difference takes about ninety seconds to spot.

3. A role map for your side. Strong firms name the internal roles you will need after they leave: who owns the model inventory, who reviews outputs, who triages exceptions, who approves changes. Better firms help you write those role definitions and identify which current employees can grow into them. A firm that has never thought about your org chart after their exit was never planning for their exit.

4. A planned disengagement curve. Ask the firm to sketch their hours by month across the engagement. Capability-transfer firms show a ramp down: heavy in months one and two, tapering as your team takes over operations, ending in an optional light-touch advisory arrangement. Dependence-model firms show a flat or rising line, often justified as "ongoing optimization." Flat lines are the parking of billable staff, rendered as a chart.

5. A shadow-then-own transition plan. The credible mechanics of transfer are boring and specific: your people shadow theirs, then your people drive while theirs observe, then your people own it with a support window. If the firm cannot describe this sequence for your specific engagement, with dates, they are improvising, and improvised transfers default to no transfer.

Red flags

• Every meaningful skill sits with the firm's staff and the proposal contains no mechanism for that to change.
• The proposal includes a multi-year managed service before the first project has shipped. Managed services have legitimate uses, but selling one before proving anything is selling dependence at the door.
• "Your team isn't ready to own this" gets used to defer every transfer conversation, without a plan to make the team ready. That line, unaccompanied by a training plan, is a revenue strategy.
• Tooling is proprietary to the firm and unlicensed to you at exit, so even fully trained staff cannot operate what was built.
• References describe the firm as "still with us" three years later in the same operational role, at the same intensity. Long relationships are fine; unchanged ones are the tell.

Green flags

• The firm answers question twelve, how their role shrinks if you build an internal team, with visible enthusiasm and a real example of a client they graduated.
• Fractional or interim arrangements come with a stated end state, such as a fractional AI lead who explicitly works themselves out of the seat by helping you hire and train a permanent one.
• Training materials, evaluation sets, prompts, and documentation are contractually yours to keep, reuse, and modify.
• They propose a smaller engagement than you expected because your team can do part of the work with coaching. A firm that leaves money on the table to build your capability is showing you its actual model.
• Their references can name specific employees who got promoted into roles the engagement created.

A worked example

A $900 million credit union compared two finalists for a document automation program. Firm A proposed a fourteen-month engagement, flat staffing of three consultants, and a managed service for "model operations" starting month six, total first-year cost around $850,000. Firm B proposed a five-month build at $240,000, two workshops written into the deliverables table, a role map naming four internal positions, and a taper to eight advisory hours per month by month seven.

The buying team asked both firms the same question: "Walk us through the last client you no longer work for, and why." Firm A's answer was a contract non-renewal it clearly regarded as a loss. Firm B named a client that had hired two analysts, taken over operations after nine months, and now called twice a year for governance reviews. That story was told as a success, because in Firm B's model it was one.

On the rubric, Firm A scored 1 on capability transfer (no transfer mechanism, dependence priced in from month six). Firm B scored 5 (named deliverables, role map, taper, a referenceable graduation). The credit union hired Firm B, and eighteen months later runs the system with internal staff, at a total cost still below Firm A's first-year proposal.

The one-question version

If you have time for a single test, use this: "Describe what our team will be able to do without you twelve months after this engagement ends, and show me where in the contract you commit to making that true." Firms built for transfer answer with specifics and point to deliverable language. Firms built for dependence answer with adjectives.

Score what you hear, in writing, alongside the other six criteria. And if you want to see how a firm structures a bounded first engagement when self-sufficiency is the stated goal, review the Advisor Labs credit union practice, then hold it to every test in this article. The rubric only works if nobody is exempt from it.