Four worlds that don't know each other
The OECD's Development Assistance Committee evaluates projects in development cooperation — a framework that has since become widely applied as a transferable framework, well beyond that sector. The UK government reviews infrastructure projects through a Gateway review. In PRINCE2 practice, an evaluation belongs at the closure of a project. Stufflebeam's CIPP model comes from educational evaluation. Four frameworks, four different origins, four audiences that have almost nothing to do with each other.
And yet, independently of one another, they arrive at the same four building blocks.
The four building blocks
Put OECD DAC (2019), the UK's Gate 5 review (2021), PRINCE2 practice, and Stufflebeam's CIPP model (2015) side by side, and the same elements keep reappearing:
- Testing benefits against the original promise — not against an expectation that was adjusted along the way.
- Independent validation — not the project manager assessing their own project.
- Criteria set in advance — not criteria invented after the fact to make the result fit.
- Action-oriented lessons learned — a lesson nobody picks up is not a lesson, just a note.
That's striking. A framework for development cooperation, a government review framework, a common project management practice, and an educational evaluation model could easily have arrived at completely different structures. They didn't.
What convergence does and doesn't prove
Convergence between four independent frameworks says something about the robustness of the structure itself: if four parties arrive at the same building blocks without consulting each other, it's unlikely to be coincidence. It says something about what makes an evaluation complete.
It does not say that a project which ticks these four boxes will subsequently be executed better. Whether a good evaluation actually leads to better subsequent projects has barely been studied; the scarce evidence that exists is old, indirect, and based on a single case. An evaluation is a tool for learning, not a guarantee that the learning happens.
What this means in practice
Most organizations don't use any of these four frameworks literally. They have their own template, or no template, or a template that never gets filled in after the project deadline has passed. That's where the convergence becomes practical: you don't have to choose between OECD DAC, Gate 5, PRINCE2 practice, or CIPP. You can use the four building blocks as a benchmark for what you already do.
At your next evaluation, ask: are we testing benefits against what was originally promised, or against what turned out to be feasible along the way? Does someone who wasn't on the project assess it, or only the project team itself? Were the criteria set in advance, or filled in after the result was already known? And is there someone who owns the lessons, or does it end with a document nobody opens again?
Test your own latest evaluation against these four building blocks with the EvaluatieScore, for €15 at dutchmind.com/producten/evaluatiescore.
Sources
- OECD DAC Network on Development Evaluation (EvalNet) (2019), Evaluation Criteria (herzien, DCD/DAC(2019)58 FINAL) — one.oecd.org.
- HM Government / Cabinet Office–IPA (nu NISTA) (2021), Gate 5: Operations Review and Benefits Realisation (Assurance Portfolio Standard, V1.0) — assets.publishing.service.gov.uk.
- PRINCE2-praktijk — End Project Report en Lessons Report Template (prince2.wiki, secundaire bron, geen officiële AXELOS-manual).
- Stufflebeam, D.L. (2015), CIPP Model checklist (2e ed.) — rszarf.ips.uw.edu.pl.