Innovation Advisory · Framework 10
5 questionsThe Feedback Loop
Five questions to test whether the loop between your programme and the people it serves closes, or whether it collects feedback into a dashboard nobody reads.
Who counts as an accountable stakeholder in this programme, and who does not?
A feedback loop with beneficiaries closes only for the people on your list of beneficiaries. Every other group (non-participants, people who refused to take part, partners' staff, the community around the household) is outside the loop by design. Name the excluded groups. Leaving them out of the loop is a policy decision, whether or not you meant it to be one.
What form of feedback suits the population, whatever platform you would prefer?
A WhatsApp bot works if respondents already use WhatsApp. A phone survey works if respondents have phones and time. A community meeting works if the community meets and trusts the space. A drop box in the ration shop works if people visit the ration shop. Choose the form according to where the respondents already are, even if your dashboard would prefer them somewhere else.
Who inside the organisation reads incoming feedback, and what authority do they have to change something?
Feedback that lands in a Google Form nobody reviews is worse than no feedback, because it creates a paper trail of consent without any of the responsibility. Name the person whose inbox receives it. Name the decision they can make on their own, and the decision they must escalate. If any of those is unclear, the loop is for show.
What has changed in the last cycle because of feedback received?
The unit to look for is a concrete change, such as a delivery mechanism revised, a schedule moved, a message rewritten or a partner replaced. If you cannot name one change from the last cycle, then the feedback was not useful, or it was not read, or it was read and refused. Whichever it was, say so plainly.
What feedback did you decline to act on, and why?
This is the hardest question and the most important one. Every programme refuses some feedback, sometimes for good reasons (safety, mandate, sequencing) and sometimes for bad ones (cost, discomfort, inconvenience). A closed loop names what it refused and why. An open loop pretends every voice was heard, and over time that pretence erodes the loop.
How to use the loop
Question one checks the boundary of the loop. Question two checks the channel, and question three checks the routing. Questions four and five are the loop itself, the two steps that turn feedback into either a change or a recorded refusal.
Most programmes fail on question five. The polite habit of pretending every voice was heard is what erodes trust in the loop over time. Naming what you refused, and why, keeps the loop honest from one cycle to the next. It also makes the next round of feedback more truthful, because respondents can see that their earlier feedback was processed and answered.