The person completing your assessment is lending you their attention. They may already suspect that client handovers are messy; they do not need a dramatic label telling them their business is broken. They need a clearer view of what to inspect next.
For your Leeds consultancy, start with one decision: does this firm need a simple handover checklist, a small process review or more evidence before deciding? Build the assessment backwards from those useful outcomes. Keep your commercial invitation visible without making every answer lead to the same urgent sales call.
Gbeya’s existing Intelligence assessments demonstrate a question-to-score-to-result pattern. They are self-report tools with configured scoring, not independently validated audits. Use that distinction as a design constraint when creating your own assessment brief; a bespoke version needs to be built and tested, not simply described in new copy.
Would this still help someone who never buys?
Use these questions to review the assessment brief.
- What specific decision is the person trying to make?
- Does each question change an action or reduce uncertainty?
- Can the person answer honestly without choosing a forced or flattering response?
- Can you explain the scoring rule and what it does not prove?
- Does the result offer a useful next step and a separate, appropriate commercial invitation?
Rewrite any question or result that exists mainly to manufacture urgency. Preserve a useful route for people who do not need your service.
This is a design review, not independent validation of a measurement instrument.
Step 1
Ask about an observable handover
For example: who confirms that the receiving colleague has accepted the client brief?
Step 2
Identify the evidence
Distinguish a named owner and visible confirmation from an informal assumption.
Step 3
Allow an unknown answer
If the person cannot tell, suggest checking one recent handover rather than inventing a score.
Step 4
Offer a useful action
Define the owner and confirmation step; make a further review optional.
This is a proposed question-to-action method, not a live customer result or a claim that a general custom-assessment builder is available.
1. Choose the decision before writing the questions
Name the problem boundary. This assessment concerns the handover from a sold project to the person delivering it, not the health of the entire business. Define the possible next steps and what evidence would justify each one.
Write a question-purpose table privately: what decision does this answer change, what will you do with it and could you obtain the same useful information with less effort? Remove questions that only make the form look sophisticated. Questionnaire guidance treats respondent burden and relevance as design concerns, not decoration.
- Why this matters
- A decision-first brief prevents collecting information with no use. It also makes the result easier to explain because each question has a role in the recommendation.
- Work through the example
- In this proposed assessment, ask whether the last completed handover named a delivery owner. The answer changes whether the next step is clarifying ownership or investigating a different boundary.
- When the approach changes
- An exploratory consulting interview can legitimately be wider. A short self-service assessment should not pretend to replace that open-ended discovery.
- Your decision
- Keep a question when it changes an action or clarifies uncertainty. Remove it when its only purpose is to make a score possible.
2. Ask about one observable thing at a time
Prefer a concrete reference period and behaviour to a flattering self-description. “Was a delivery owner named in the last handover?” is easier to interpret than “How mature is your client experience?” Avoid combining ownership, budget and timing into one answer.
Give uncertainty a legitimate route. The respondent may not know what happened in another team or may not have a recent project. Do not force that person to choose an invented answer. Explain what an unknown answer means for the next step instead of treating it as a moral failure.
- Why this matters
- Specific questions reduce ambiguity between respondents. An honest unknown identifies missing evidence; it should not be silently converted into evidence of poor performance.
- Work through the example
- For this handover example, separate “Was the scope recorded?” from “Did the delivery owner confirm it?” A firm can have a document without an agreed handover.
- When the approach changes
- Some questions concern perceptions, such as confidence about the process. Those can be useful if labelled as perceptions rather than objective facts.
- Your decision
- Ask one concept, define the context and make the answer options cover the real situation without steering the person.
3. Explain the score without giving it borrowed authority
If a score helps the reader, show what it summarises. Explain the direction, the weighting and the limits in ordinary language. A 0–100 scale is a presentation choice; it does not by itself make the instrument validated, predictive or comparable with other firms.
In Gbeya’s Business OS example, eight equally weighted questions each offer values from zero to three, and the total is normalised into a score band. That mechanism is inspectable, but the answers remain self-reported and the framing reflects an explicit product viewpoint. A consulting assessment should be equally honest about its own assumptions.
- Why this matters
- A transparent calculation lets the reader understand the result and challenges the author to justify weights and thresholds. It does not create evidence that a score predicts revenue, risk or future performance.
- Work through the example
- If the eight assumed answers sum to fifteen out of twenty-four possible points, the rounded score is sixty-three. That number describes the scoring rule; it is not “63% healthy” or a percentile among businesses.
- When the approach changes
- A short checklist may produce better guidance without any total score. Use a score only when it helps prioritisation rather than disguising uncertainty.
- Your decision
- Keep the calculation reproducible and the interpretation modest. Validate a stronger claim before publishing it.
4. Give one useful action before the commercial invitation
A result should connect the observed answer pattern to a proportionate action. State what to inspect, what a useful outcome would look like and what you cannot conclude from the answers. Avoid sending every respondent to the most expensive service regardless of their situation.
For the handover example, a low-confidence ownership answer could lead to reviewing the last three handovers and naming a responsible person for each. If the evidence shows that ownership is already clear, another action should be available. A paid review can remain an option without becoming the only way to obtain value.
- Why this matters
- A useful action turns the result into a decision aid. A commercial invitation is more credible when the person can see why it follows from their situation and when it may not be needed.
- Work through the example
- The proposed result says: “Check three recent handovers. If none records an accepting delivery owner, agree that step before changing software.” A secondary invitation offers a focused process review if the team cannot resolve the boundary.
- When the approach changes
- Some cases require a specialist conversation because the available answers are insufficient. Say that directly rather than manufacturing a confident diagnosis.
- Your decision
- Provide the smallest useful next step supported by the answers, and label the paid invitation as an option rather than an inevitable conclusion.
5. Test understanding, saved results and the handover
Ask a small set of intended users to explain each question in their own words and describe what they would do after the result. This is a proposed test, not evidence already collected. Look for misunderstood terms, forced answers and recommendations that do not follow from the situation.
Then test the working journey separately: answer validation, result display, any consent choice, saved result and optional next action. A completed assessment, a clicked invitation and a qualified consulting enquiry are different outcomes. Do not report the first as proof of the last.
- Why this matters
- Question testing checks meaning; functional testing checks behaviour. Both are needed because a technically correct form can still ask a confusing question or produce an unhelpful recommendation.
- Work through the example
- For this handover example assessment, one tester may interpret “owner” as the business founder while you mean the delivery lead. Fix the wording and retest before analysing scores as if everyone answered the same question.
- When the approach changes
- A small pilot does not validate a diagnostic instrument for every industry or prove commercial lift. Record what it found and what remains unknown.
- Your decision
- Release only when the intended reader can understand the questions and next step, and the implemented journey preserves its stated consent and evidence boundaries.
Useful assessments earn the next conversation
Relevance, understandable questions and proportionate interpretation make the assessment more defensible. Government and practitioner guidance converge on those design concerns, while their contexts differ from a commercial consulting funnel.
The existing Gbeya pattern shows how configured answers can lead to a score and result. It does not establish that the score predicts business outcomes or that a custom assessment is already available to every workspace. Keep the product example, the proposed design and the evidence you still need distinct.
The practical standard is straightforward: the reader should understand what the result means and have a sensible action they can take. Whether that produces qualified enquiries is a separate question to measure after release.
Medium confidence in this proposed design method. Transparent scoring explains a calculation; it does not establish predictive accuracy or commercial value.
No original interviews, instrument-validation study or enquiry-lift experiment supports this example. The government questionnaire guidance is currently marked under review; its general design principles do not validate a consulting assessment.
Scoring calculation with stated assumptions
A reproducible score is not a validated diagnosis
Use eight equally weighted answers scored from zero to three, matching the structure of the existing Business OS example. Assume the chosen values sum to fifteen; no actual respondent is represented.
Maximum = 8 × 3 = 24. Normalised score = round(15 ÷ 24 × 100) = 63.
A different weighting or set of questions can produce a different result. The score is not a percentile, probability of success or audited business-health measure.
Show the interpretation rule and check whether it produces a useful next action. Do not attach stronger meaning merely because the result is numerical.
Assumptions worth checking
A 0–100 score is objective validation.
It can be a transparent calculation without being a validated measure.
Every low score should lead to a sales call.
The next action should follow the actual need, including a useful self-service route.
Unknown means poor performance.
It may mean missing evidence or an inapplicable question.
Experienced consultants already know how users interpret their questions.
Test the reader’s interpretation; domain familiarity can hide ambiguity.
A completed assessment is a qualified lead.
Completion, contact capture, qualification and paid work are separate outcomes.
A saved result proves consent to all follow-up.
Respect the specific choice and purpose; do not infer wider permission.
Your next 30 days
Days 1–7
Write the decision brief
Define one audience, situation, decision and set of useful next actions.
Ready for the next step: Every proposed question has an explicit purpose.
Days 8–14
Build the question and result map
Write observable questions, uncertainty routes and transparent scoring if needed.
Ready for the next step: Results follow the answer patterns without manufactured urgency.
Days 15–21
Test meaning and implementation
Pilot comprehension with intended readers, then test the implemented validation, result and consent journey safely.
Ready for the next step: Dated findings, corrected ambiguity and verified functional boundaries.
Days 22–30
Release and measure separately
Publish the accepted assessment and distinguish completions, chosen next actions, qualified enquiries and paid outcomes.
Ready for the next step: An evidence-led improvement list without invented attribution or lift.
Questions before you proceed
Does every assessment need a score?
No. A checklist or decision path can be more useful when a total would hide important differences.
Can I use Gbeya’s existing assessments as examples?
Yes, as examples of configured self-report scoring and result handovers. They are not independent audits or proof that your custom instrument is implemented.
How many questions should I ask?
Use the fewest that support the intended decision. There is no universal count that guarantees quality or completion.
What should I do with unknown answers?
Make their meaning explicit. They may call for evidence gathering rather than a low score; verify how the implementation represents them.
Should the result always ask for a call?
No. Offer a useful next step proportionate to the answers, with a call when investigation would help.
How do I measure whether it works?
Check comprehension and next-step usefulness, then measure the separate commercial outcomes. Do not equate a completion or click with a qualified customer.
Sources and further reading
Prepared by Gbeya · Sources reviewed 30 October 2026.