Search published Practitioner Experience Notes (PENs). Members can view any PEN; only the author and admins can edit. Authors can choose to display their name or remain anonymous to Members, though public search results will always be made anonymous.
“Watermelon KPIs”: avoiding bad performance data
When I rejoined SCUK in 2019 and then SCI in 2020 as Director of Evidence and Learning, there was a fair amount of chuckling at what were termed “watermelon KPIs” and management performance data at the organisational level that people – often rightly – felt wasn’t reliable. The results were nominally green (on track), but if look below the surface and it was definitely red – hence watermelon!
The KPIs most prone to being watermeloned were related to programme quality. (At that time, the main concerns were about accountability to affected populations and gender sensitivity of projects.) This was troubling, as it meant most judgements about performance were based on more reliable spend and compliance data: important, but not the full story. We identified 3 criteria for KPIs that made them far more prone to watermelon behaviour:
(1) Indicators that are hard for others to check. Someone in HQ or the regional office can’t easily check if your systems for accountability to local populations are as good as you claim without actually visiting; whereas spend/ burn rate data was more verifiable.
(2) Indicators relating to topics that reporting staff didn’t feel mattered as much to more senior staff: watermelon KPIs tended to be seen as “nice to have” compared to KPIs on ‘harder’ issues like spend, donor reporting and fraud, for which there were more felt consequences for poor performance.
(3) Indicators that were ambiguously measured, where there was scope for judgement: if it was an unclear judgement call between a good and bad score on something, why not save yourself hassle and go with the better result?
I can’t say we eliminated the problem, but things that helped:
- Simplifying/ clarifying indicators and the guidance (based on input and feedback from country teams
- Regional teams doing spot checks: not huge in volume but enough to know you might get challenged and called out
- Providing more support, engagement and leadership interest in those indicators: demonstrating more that it mattered (as always, some countries/ regions were more active on this than others)
- Alongside the positive support, trying to change the tone away from ‘watermelon KPIs’ being seen as kind of humorous to framing it potentially as a form of fraud with real implications for impact if it was hiding poor quality programming. The carrot of support was essential alongside the stick of challenge and not letting it be acceptable.
So if you’re planning new KPIs, consider the potential for them to be “watermeloned”, and think about how to avoid that!