Score a partner before you sign them

When to use this

Use this before signing anything longer than a pilot agreement, and before committing meaningful budget or staff time to a partner launch. It is built for the moment when a partnership looks promising on relationship terms — a good meeting, a recognisable institution, mutual enthusiasm — and the team needs a second, colder pass before that goodwill turns into a commitment. A logo is not a channel. A channel is a repeatable path from a defined buyer need to a delivered service or product, with clear economics and evidence behind it, and this scorecard exists to test for that repeatability specifically.

It is the wrong tool once a pilot is already running and generating its own data — at that point, re-score using the Pilot Design One-Pager's actual measurement results, not fresh guesses at the same ten criteria.

How to score

Score each criterion 1 (weak evidence) to 5 (strong evidence), based on what you can actually verify today — not on what the partner has promised. Multiply by weight. A score below 3 on any gate criterion pauses the decision regardless of the total, because a gate failure is not something a strong total can average away.

Buyer access, delivery capacity, and commercial clarity carry the most weight deliberately — a partner can be trustworthy, well-connected, and easy to reach and still fail as a channel if it cannot actually fulfil at the proposed volume or if nobody has agreed who pays whom and when.

Gate questions

Answer yes or no before approving a pilot. A single "no" here is a stop condition, independent of the weighted total above.

• Is there a named buyer or user, not only an institutional logo?
• Can the partner fulfil the promised service at the proposed sites and volume?
• Is the payment event defined, including who pays and when?
• Can activity be evidenced without collecting more patient information than necessary?
• Is there a named operational owner on both sides?
• Is there a stop condition if quality, safety, or economics fail?

Write down the operating design

Once the scorecard clears, use this table to record the actual operating design before drafting a pilot agreement.

Questions for the partner interview

Ask these directly, in the room, before the scorecard is filled in — the answers usually tell you more than the partner's pitch deck does.

1. Walk me through the last time this customer problem was handled. Where did the process actually slow down?
2. Which person on your team owns the hand-off, and what else is on their scorecard?
3. What can you evidence weekly without creating a new reporting burden on your side?
4. What would make this unattractive for your team after the first month?
5. Which promises should we refuse to make until the operation is proven?

The fourth and fifth questions are the ones partners are least prepared for, and the ones most worth pushing on. A partner who cannot answer question four has not thought hard about their own incentives, which means the alignment you're counting on is assumed rather than tested.

Decision rule

Record the score, the evidence behind it, the decision, and the reviewer's name. Re-score after the pilot using actual operating evidence, not the same estimates. A partner can score well here and still turn out to be the wrong channel once real volume moves through it — the scorecard reduces the odds of a bad launch, it does not eliminate them.

How to read the result, and where the scorecard breaks

The scorecard's main failure mode is optimism bias on the "evidence / gap" column — scoring a criterion a 4 because the partner said something plausible in the meeting rather than because you verified it. Treat every score above 3 as a claim that needs a source, the same evidence-grading discipline as the 48-Hour Revenue Leakage Audit. If the evidence column is empty, the score in that row is a guess, and a channel decision should not rest on ten guesses that happen to sum to a comfortable total.

The second failure mode is running the scorecard once, at the enthusiasm stage of the relationship, and never again. Scores decay — a delivery-capacity score of 4 given in January can be a 2 by June if the partner's staff turned over. Treat this as a living document for any partnership doing meaningful volume, not a one-time gate.

Worked walkthrough

A generic scenario: a diagnostics company evaluating a proposed partnership with a twelve-branch pharmacy chain in Ikeja to run sample collection at the counter, against an existing informal relationship with three of those branches that has worked reasonably well for six months.

Buyer access scores a 4 — the chain's footfall genuinely matches the target customer profile, verified against foot-traffic estimates the pharmacy itself shares openly. Delivery capacity scores a 2: only three of the twelve branches have a trained staff member and cold-chain storage for samples today, and the chain's plan to train the other nine "within the quarter" is a promise, not evidence, so it is logged honestly as a gap rather than rounded up. Commercial clarity scores a 3 — a price and margin split exists, but the payment event ("who pays whom, when") is still being worked out for the walk-in-without-referral case, which is a meaningful share of expected volume.

With delivery capacity below 3, this is a gate failure regardless of the weighted total elsewhere. The correct action, per the decision rule, is not "launch across twelve branches" but "narrow": pilot at the three branches that already have trained staff and storage, use the operating-design table above to lock the payment event before go-live, and re-score the full chain only once training data on the other nine branches exists as evidence rather than intention.

If you're weighing a partnership that looks good in the room and need a second pass before signing, contact Dr Tolu Ajidahun to run the scorecard together.