Impact Measurement

One metric, defended, beats five claimed

SDG impact appears in every lock’s rubric and peaks at Lock 3, where "measurable impact" is a required field. The trap is breadth: teams claim lives improved, time saved, costs reduced and awareness raised, and can evidence none. Choose the single metric closest to your core loop and build your measurement around it.

The metric-quality test

• Observable: your product or pilot can actually record it — not a survey hope
• Attributable: the change is plausibly caused by your tool, not the weather
• Timely: it moves within weeks, not years — you have a build window, not a decade
• Honest: it measures user value, not your engagement (screen time is not impact)

Baselines make numbers mean something

"Eighty per cent follow-up completion" is noise without the before: was it sixty or eighty-five? Your Lock 1 field research is your baseline instrument — measure the current state while you are there. Where a true baseline is impossible in the window, use a comparison group, published benchmarks (cited), or worst case a clearly-labelled assumption.

A pilot small enough to finish

Ten real users measured for two weeks beats a thousand hypothetical users. Recruit from the community you researched in Lock 1, get consent, instrument the metric in the product (see the prototyping guide), and report the actual number with its actual context — including drop-offs. Evaluators trust small honest numbers and discount large vague ones.

Reporting: the sentence that scores

Aim for one sentence of this shape in your Lock 3 submission: "Across N users over T weeks, [metric] moved from [baseline] to [result], measured by [method]; the main limitation is [limitation]." Every clause is defensible. That sentence, with its evidence, is what measurable impact means.