When a funding programme runs for the second or third time, the same thing happens almost every time: the scoring rubric grows. After each jury meeting one more criterion is added, because some aspect was missing. After three rounds twelve jurors are scoring eighteen criteria on a scale of one to ten — and nobody can say any more why one project came out on top.
This is not a minor detail. The rubric is the point where a programme translates its funding strategy into a decision. If the rubric is vague, the decision will be vague too.
Fewer criteria, clearer weighting
In the programmes we support, five to seven main criteria work best. Each one gets an explicit weighting, and together they add up to one hundred per cent. That weighting is the real strategic statement: if «team» counts for thirty per cent and «product» for ten, then the programme has decided that it invests in people rather than prototypes. That deserves to be communicated as such.
Underneath each criterion sit two or three guiding questions — not further sub-criteria with their own points, but wording that gives everyone on the panel the same understanding:
- Can this team build the product and sell it?
- Is the problem described urgent, specific and recurring?
- Is there a credible plan for the first twelve months?
Short scales, clear anchor points
Ten-point scales suggest a precision nobody has. There is no difference between a 6 and a 7 that two people would judge the same way independently. A short scale is enough, provided every step is described: what does the second-lowest step mean, what does the second-highest? Without those anchor points, strict jurors score systematically lower than lenient ones, and the result ends up measuring how the dossiers were assigned rather than how good they are.
A rubric should prepare the discussion, not replace it. The interesting cases are the ones where the scores diverge.
Divergence is the signal
The most common mistake in the analysis: averaging the scores and reading the ranking from the top down. That throws away the most valuable information. A dossier where three assessors are far apart is not average — it is contested. Those are exactly the cases that belong on the meeting agenda, while the unanimously scored dossiers usually need little discussion time.
That is why Reviu can display the spread alongside the weighted average and sort by it. Programmes that work this way regularly tell us their meetings get shorter — and their reasoning towards rejected applicants more robust.
Comments are mandatory fields
A number without a justification is worthless six months later. Anyone who has to explain a rejection, justify follow-up funding or defend a programme to a supervisory body needs text. A short mandatory field per main criterion — two sentences will do — adds a few minutes per dossier and makes the entire procedure traceable.
In practice: Before the next call, a test run with five dossiers from the previous round is worth the effort. If two people assess them independently and end up far apart, the cause is almost never the people — it is a criterion that measures two things at once.