On the Friday afternoon a municipal digitalisation project wraps up, the business case gets filed away and no evaluation report ever appears that revisits the benefits it promised — that's what happens with the vast majority of projects. Dutchmind is building six connected AI tools that assess that same documentation across a project's entire lifecycle, from first idea to completed evaluation.
The six steps of the lifecycle
CaseCheck assesses the business case: the justification for why a project should be started at all. Is there a seriously worked-out zero option, are the benefits realistically estimated, was there an actual trade-off between options rather than documentation of an already-chosen solution?
PlanScore assesses the plan of approach: the translation of the business case into an executable plan. Is the planning realistic, are roles and responsibilities clearly assigned, does the chosen governance fit the size and complexity of the project?
RisicoRadar analyses project risks across five categories, with an overview of the top risks and concrete recommendations. Risks underestimated at the start of a project are among the most common causes of later overruns.
StakeholderScan looks at the project's environment: who are the stakeholders, what are their interests, and is there a demonstrable approach to involving them throughout the project rather than discovering only afterwards that an important party felt excluded?
RapportageScore assesses the progress reports produced during execution itself: the periodic reporting to the steering group or sponsor. A good progress report signals problems early and honestly, rather than staying optimistic until it's too late to course-correct.
EvaluatieScore, the capstone of the suite, assesses the evaluation report after the project closes: were the promised benefits realised, was there methodologically independent testing, and are the lessons concrete enough to actually be applied to the next project.
Why this order is more than marketing
It's tempting to read this as a sales pitch: six products that happened to be built one after another. The underlying logic is stronger than that. The categories these six tools test genuinely interlock. The benefits promised in the business case (CaseCheck) are exactly the benefits EvaluatieScore later tests to see if they were realised. The risks identified at the start (RisicoRadar) are the risks an evaluation should later state whether they materialised, and if so, how they were responded to. The planning from the plan of approach (PlanScore) is the reference point against which an evaluation measures actual duration and cost (category 5 of the EvaluatieScore rubric: performance on time, cost, quality, scope and risk).
In other words: each of the six tools tests a document at the moment it is drawn up, but the content of those documents simultaneously forms the reference point for the next step in the cycle. An organisation that has all six moments tested builds a continuous chain of clear reference points, instead of six separate documents, each assessed on its own without any mutual coherence.
The gap this suite addresses
As we describe in another article, there is an instrumental gap in the Netherlands between mandatory, institutionalised testing beforehand (think of the Adviescollege ICT-toetsing) and voluntary, often unstructured evaluation afterwards. These six tools don't close that gap by creating a new obligation (that is not Dutchmind's place to do), but by making a low-threshold, fast instrument available for every phase that does apply the frameworks of authoritative sources (OECD, PRINCE2, CIPP, the British Gate methodology, the Five Case Model) without requiring a heavyweight, external testing process.
Where to start
Not every organisation needs to start with CaseCheck. Anyone who has just completed a project and wants, for the first time, to seriously check whether the evaluation has substance, logically starts with EvaluatieScore. Anyone considering a new project starts with CaseCheck or PlanScore. The strength of the suite lies not in a mandatory order, but in the ability to get the same kind of sharp, source-grounded judgement, regardless of where a project currently stands.
An organisation that has just wrapped up a project can have the evaluation report tested at dutchmind.com/producten/evaluatiescore — for €15, a concrete verdict instead of a filed-away report.
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.
- Stufflebeam, D.L. (2015), CIPP Model checklist (2e ed.) — rszarf.ips.uw.edu.pl.
- PRINCE2-praktijk — End Project Report en Lessons Report Template (prince2.wiki, secundaire bron, geen officiële AXELOS-manual).
- Tijdelijke commissie ICT-projecten (commissie-Elias), Eindrapport "Grip op ICT" + kabinetsreactie (2014/2015) — pianoo.nl.