Instrument
A working framework for measuring the before-and-after impact of a European project, demonstration, or socially innovative action, set up at the start, run while the work is live, closed out with proof.
This template is built to be filled in. It is deliberately light: every field earns its place by feeding a decision or a final claim. Adapt the indicators to your project; keep the structure, because the structure is what makes one project's evidence comparable to another's.
How to use it: complete Part A at kick-off, before any activity begins. Run Part B at the agreed midpoints while the project is live. Complete Part C at close. Parts D and E are designed once and reused throughout.
| Field | Entry |
|---|---|
| Project / action name | |
| Funding programme & call | |
| Grant agreement / reference | |
| Lead partner | |
| Measurement partner | Transitions Lab |
| Reporting period | Start ▢▢/▢▢/▢▢▢▢, End ▢▢/▢▢/▢▢▢▢ |
| Demonstration / pilot sites |
State, in one sentence each, what this project claims it will change. Be specific enough that someone could disagree.
| # | We claim this project will… | For whom | By when |
|---|---|---|---|
| C1 | |||
| C2 | |||
| C3 |
A claim that cannot be wrong cannot be evidenced. "Raise awareness" is not a claim; "increase the share of participating households who switch supplier within six months" is.
For each claim, make the logic explicit, so everyone can see what is being asserted and what would count as evidence.
`` ACTIVITY → OUTPUT → OUTCOME → IMPACT (what we (what it (what (the change do) produces) changes) that lasts) ``
| Claim | Activity → Output | Outcome | Impact | Key assumption that must hold |
|---|---|---|---|---|
| C1 | ||||
| C2 | ||||
| C3 |
Choose a small number of indicators per claim. Mix the three dimensions, reach, depth, experience, so the project is not measured on headcount alone.
| ID | Indicator | Dimension | Unit / source | Baseline → Target |
|---|---|---|---|---|
| I1 | reach | |||
| I2 | depth | |||
| I3 | experience | |||
| I4 | ||||
| I5 |
Reach = who is affected and who is left out. Depth = how materially things change. Experience = what the change feels like to the people inside it.
The baseline is the single most important field in this document. Without it, there is no "before," and impact cannot be measured, only asserted.
| Indicator ID | Baseline value | Date captured | Method / sample | Captured by |
|---|---|---|---|---|
| I1 | ||||
| I2 | ||||
| I3 |
Run this short, repeatable cycle at each midpoint. The point is to catch what is working and what is not while there is still time to act.
| Indicator ID | Baseline | Midline 1 | Midline 2 | On track? (↑ / → / ↓) |
|---|---|---|---|---|
| I1 | ||||
| I2 | ||||
| I3 |
| What the data is showing | What it suggests | Action taken (and by whom) | Date |
|---|---|---|---|
A midline reading that changes nothing was not worth collecting. This table is the project's steering log.
A handful of structured questions put to the people the project affects, repeated each cycle so a single survey becomes a time series. (See the companion interview guide for depth.)
| Question | Baseline | Midline 1 | Midline 2 | Endline |
|---|---|---|---|---|
| Has [the change] happened for you? (yes / partly / no) | ||||
| How much has it changed things? (0-10) | ||||
| What is the single thing you would fix first? (open) | ||||
| Would you recommend / continue? (0-10) |
The headline output. The same instrument as baseline, run again.
| Indicator ID | Baseline | Endline | Change (abs. / %) | Target met? (+ confidence) |
|---|---|---|---|---|
| I1 | ||||
| I2 | ||||
| I3 |
For each claim, a plain-language verdict, including where the evidence is inconvenient.
| Claim | Verdict (supported / partly / not supported) | What the evidence actually shows | What we cannot conclude |
|---|---|---|---|
| C1 | |||
| C2 | |||
| C3 |
The willingness to write down these answers is the difference between an evaluation and a press release.
A one-line, fixed definition for every indicator, so "participant," "user," and "reached" mean the same thing across every site and partner. Incomparable definitions are the most common reason multi-site evidence fragments.
Target sample per site; how respondents are selected; how informed consent is obtained and recorded; how the data is stored. Contact only consented participants.
One row per site, same columns, so results can be read side by side rather than as a pile of separate reports.
| Activity | Owner | Cadence |
|---|---|---|
| Baseline capture | Measurement partner | Once, at kick-off |
| Midline readings | Measurement partner + site leads | Per agreed midpoints |
| Steering log review | Project coordinator | Each cycle |
| Lived-experience surveys | Measurement partner | Baseline → endline |
| Endline & final reading | Measurement partner (independent) | Once, at close |
Companion documents: the Field Research service page (methods behind this template), the Impact Measurement approach it feeds into, and the Interview Guide (the depth this evidence can reach). This template is designed to be operated by the Lab as the independent European Impact Tracking function, or handed over for a partner to run with the Lab's support.