Stakeholder evidence matrix
4 min read
Lesson objectives
- Produce a evidence matrix and follow-up question list from the case facts and state decision limits.
Please sign in to use all the tools
Once you sign in, the reading tools are yours: larger or smaller text, typeface, line spacing, focus mode, highlighting and bold, and notes saved to your account and gathered in My study.
When accounts of a place conflict, the specialist should neither select the loudest voice nor average statements that measure different things. Begin an evidence matrix with columns for the claim, speaker and role, time and setting of the experience, direct observation, proposed explanation, and a question that could challenge it. A trader saying that seating reduces sales reports experience or expectation; it is not a demonstrated economic effect without a suitable comparison. A caregiver saying that the square feels uncomfortable is important to understanding that experience, but does not alone locate the cause or its prevalence. Record interests and decision authority without automatically dismissing someone because they have an interest.
Sort material into observations, testimony, hypotheses, and binding constraints. In the observation column put a count, location, or time with its collection method. In testimony preserve the meaning without turning one person's view into a whole group's view. In hypotheses state a testable explanation, such as shade scarcity reducing afternoon staying. In constraints identify matters needing formal review: shop loading, accessible travel, or site ownership. Put the one-day limit beside a one-day count. Repeated statements from the same participants do not create a representative sample, and people unable to access the square may be absent from the evidence altogether.
Select follow-up evidence that distinguishes rival explanations. If traders need a short customer loading stop while caregivers worry about vehicles beside seating, the issue might be when and where loading occurs or a poorly separated route, rather than one group rejecting the place. Map the loading location and vehicle path, observe comparable periods with and without loading, and ask departing visitors a focused reason question. This is not permission to conduct a risky traffic experiment; movement changes need an authorized owner. Both accounts may be true at different hours, which makes a time map more useful than a generic question about liking the square.
When participation is uneven, design outreach that includes those absent from the usual meeting. State the purpose and decision boundary of participation, list the user groups whose experience matters, and record channels, times, responses, and missing voices. Do not derive a representation percentage from a convenience sample. If fourteen traders and three caregivers attend one meeting, the numbers describe attendance, not the makeup of local users. Carry that uncertainty into the decision brief and avoid saying the intervention has support from the whole community. Seek input from people directly affected by the trial location and check whether language or timing blocks participation.
The lesson ends with a bounded recommendation, not indefinite deferral. If a concrete access conflict appears, contain or refer it immediately before a survey is complete. When choosing between two otherwise feasible seating locations, show each option's effects, a specific missing fact, and a short trial decision date. Keep a response log identifying which claim changed the design and which did not, with reasons. Participation does not mean implementing every suggestion. A clear rationale lets others review the decision as evidence changes.
Lesson summary
Accept if the response separates speech from counts and explanation, calculates 120 and 108 passages per hour, notes weather and missing sales, asks a testable access question, and conditions continuation on an obstruction check.