Start with the decision rather than the dashboard. An evaluation is easier to use when the reader can identify the intervention, the population, the comparison and the time period. A useful field account also explains what did not work and which conditions may prevent another team from reproducing the result.
Separate activity from outcome
Training delivered, visits completed and software installed describe activity. Changes in continuity of care, service use or patient experience address different questions. State which outcome the project actually measured.
Make the comparison visible
Explain what happened before the intervention or in the comparison group. Note changes in staffing, eligibility, funding and measurement. A before-and-after difference can have explanations other than the intervention.
Report uncertainty and limits
Give the size and selection of the sample and describe missing information. Do not present a short pilot as evidence of a lasting effect. Tell a potential adopter which resources and circumstances were essential.