Use the same test for every candidate
First describe the current bottleneck in operational terms: the founder owns every discovery call, research produces weak-fit accounts, or qualified evaluations stall without a commercial lead. Give the same problem, segment and product constraints to each agency. This makes their proposals comparable.
Then inspect a complete workflow. Ask for an anonymized account rationale, accepted opportunity, rejected handoff and handover artifact. Explain that you are evaluating the reasoning rather than asking for confidential customer data. A provider can demonstrate the process with a clearly labeled hypothetical example if real evidence cannot be shared; that example is a capability demonstration, not outcome proof.
| Decision | Evidence to use | What changes next |
|---|---|---|
| Scope | Named owner, acceptance evidence and client dependencies for each stage | Identify uncovered work before comparing fees |
| Evidence | Attribution, timeframe, denominator and relevant examples | Distinguish credible outcomes from broad experience claims |
| Continuity | Client-owned systems and an independent operating test | Estimate the cost of operating after the engagement |
Work through the decision
Illustrative selection exercise: give two candidates an account with a plausible problem, a positive reply and an unknown budget owner. Ask them to write the next action, discovery questions and acceptance conditions. One immediately counts a $40,000 opportunity; the other records an interested conversation and identifies what discovery must establish.
The second response is more inspectable, even if its reported pipeline is initially smaller. Now ask what changes after the buyer confirms an active project. A useful evaluation reveals whether the team can move between uncertainty and evidence rather than simply fill a CRM field.
A scorecard becomes false precision
A weighted score can hide a critical failure. Treat client ownership, truthful evidence and responsible buyer handling as minimum conditions. A strong average should not compensate for a provider that cannot explain who owns discovery or refuses to clarify which claimed outcomes it produced.
A concrete next step
Make a comparison sheet with scope, proof, people, costs and handover. Record a source or observed example beside each assessment, and keep unanswered questions separate from confirmed weaknesses.
Sources and research notes
- GitLab commercial opportunity stagesCompany operating handbook
- Attio: Back up workspace dataProduct documentation
Primary sources reviewed October 6, 2026. The operating recommendations and worked scenarios are Daavid’s analysis. Illustrative numbers are assumptions, not measured client results. Company marks identify sources and prior experience; they do not imply a customer relationship or endorsement.
Schematic