Choose a small set of measures tied to the partnership’s purpose and each organisation’s contribution. Define the population, period, source and owner for every measure. Review benefits alongside effort, exceptions and unintended effects so one attractive number does not conceal a weak customer or delivery experience.
Begin with the decisions the scorecard should support
Identify what the partners will decide during a review: whether to repeat an activity, fix a handoff, change a customer message or explore a larger commitment. Select measures that can inform those decisions. A scorecard containing every available metric can create a reporting burden without making the next action clearer. Explain how each chosen measure connects to the partnership’s purpose.
Include the perspectives that matter to delivery. Customer outcomes, each partner’s aims and the resources consumed may require different observations. Avoid treating the partnership as successful for everyone because one organisation sees more leads. A useful scorecard makes those differences visible without forcing every form of value into an artificial single rating.
Write definitions that both organisations can reproduce
For each measure, record its meaning, unit, population, reporting period, exclusions and source. If the scorecard counts accepted referrals, explain what accepted means and who records it. A referred contact, an agreed conversation and a completed purchase are different events. Define how repeated, withdrawn or unresolved cases are represented so the total does not change with the person preparing it.
Name a source owner and show the relevant cut-off or freshness limit. If the partners use different systems, agree how records are reconciled without sharing unnecessary personal information. Keep estimates and self-reported observations labelled. An exact-looking number is not automatically comparable across organisations simply because both dashboards use the same heading.
Balance benefits with effort and exceptions
Pair the intended benefit with an observation about the work required or a consequence that could undermine it. An increase in introductions may be accompanied by low relevance or a large support queue. A completed delivery may require more coordination than either partner can sustain. These observations should inform investigation rather than become automatic evidence that a particular team performed badly.
Use a small number of meaningful qualitative observations alongside counts when needed. Record recurring customer questions, unclear responsibilities or a reason a case was declined. Keep the evidence proportionate and appropriately handled. A balanced scorecard does not require inventing a precise numerical score for trust or relationship quality when the team cannot explain how it was measured.
Connect the review to an owned action
Set a review rhythm that fits the activity and the availability of useful evidence. Discuss changes, unusual cases and limitations before drawing conclusions from a total. Where measures disagree, identify the next question rather than averaging them into a reassuring composite. Explain any definition change so comparisons across periods remain understandable.
End the review with an action, owner and next check. The action may be to change an invitation, clarify a handoff or gather better evidence before deciding. Remove measures that repeatedly consume effort without informing a decision. A scorecard is useful when it helps the partners improve or stop an activity, rather than becoming a polished report that assumes the relationship must continue.
A fictional referral scorecard reveals a relevance problem
A fictional scenario to explain the method, not a customer result or a market benchmark.
A fictional consultancy and software reseller record twenty introductions during a month. The first report highlights that count alone. In a joint review, the partners define later stages separately: twelve recipients agreed to a conversation, seven conversations addressed the intended customer problem and three led to a separately scoped proposal. No sale is assumed from those stages.
The scorecard also records coordinator effort and the main reasons contacts declined. Several introductions were outside the agreed customer segment. The partners therefore change the referral brief before increasing volume. They retain the stage definitions and reporting cut-off so the next review can compare like events. These fictional counts illustrate a measurement structure; they are not a benchmark or evidence that the partnership generated commercial returns.
TAKE THIS INTO YOUR NEXT CONVERSATION
A practical next-step checklist.
- Tie each measure to a real review decision.
- Define population, event, unit, period, exclusions and source.
- Separate introductions, conversations, proposals and completed outcomes.
- Review effort, exceptions and customer experience alongside benefits.
- End with an owned action and explain definition changes.
Common questions.
Should a partnership have one overall score?
Only if the components and weighting have a defensible purpose that the partners understand. Separate measures are often more useful because they show trade-offs and exceptions. A composite can hide a customer or delivery problem behind a strong promotional result.
How many measures should the first scorecard include?
Use enough to assess the purpose, important contributions and practical delivery limits without creating unnecessary reporting work. Start with the next decisions and add a measure only when its definition, source and potential use are clear.
Further reading & source context.
These primary references provide background. They are not evidence for the fictional example above and do not endorse AccelerateLink.
- SBA: business growth resources (U.S. context)
- UK Government: project delivery standard (public-sector context)
- GOV.UK Service Manual: define success and measure performance
Found an error or a missing qualification? Send an editorial correction.
Find my next move.
Use a practical tool to turn this idea into a clearer starting point.
Find my next move