Mensura runs into the ordinal scale: ‘very good’ need not follow mathematically from the pluses and minuses
The Council of State rejected Mensura’s extreme-urgency application against Houthalen-Helchteren’s award of its external prevention service to IDEWE: an ordinal rating scale (very good = 100 %, good = 75 %) requires no mathematical link with the number of plus and minus points, and assessment elements the specifications do not formulate as weighted sub-criteria need not be reviewed systematically and exhaustively.
What happened?
In 2022 the municipality of Houthalen-Helchteren, acting as a central purchasing body, sought an external service for prevention and protection at work through a negotiated procedure without publication, plus replacement cover for the internal prevention advisor and confidential counsellor. The specifications used five award criteria — price (15 points, split into 1A to 1D), quality of service (35), cooperation, availability, communication, reporting and software (40), and twice team composition (5 each) — and an ordinal rating scale: ‘very good’ earned 100 % of the points, ‘good’ 75 %, ‘satisfactory’ 50 %. Four tenderers submitted offers; three were found regular. The first award report of 6 July 2022 put IDEWE first with 94.02 points, ahead of Mensura (78.75). The municipality awarded to IDEWE on 28 September, withdrew that decision on 13 October after a letter from Mensura’s counsel, and after a reworked award report of 19 October (IDEWE 86.52, Mensura 78.75) awarded to IDEWE again on 24 October. Mensura went to the Council of State: the scores allegedly rested on assessment elements without basis in the specifications, a genuine comparison of offers was lacking, and it was disproportionate that its note under criterion 3 — one paragraph over the requested 15 pages — would cost it 10 points. The Council followed none of it. The award report did compare the offers, section by section, with verbal reasons (‘more extensive than the other tenderers’, ‘less extensive’). The ordinal scale requires no mathematical link between the number of plus and minus points and the score, as long as the verbal reasoning can support the overall rating — and Mensura had not challenged the choice of that scale itself. Moreover, the assessment elements were not formulated as sub-criteria with their own weighting, so they did not have to be reviewed systematically and exhaustively, and related elements could also count. The page overrun proved not to be the cause of the points gap; that was mainly IDEWE’s clearly more extensive software, and Mensura’s claim that MyMensura could do the same remained unsubstantiated. The criticism of criterion 1A (1.25 points) could not bridge the 7.77-point gap anyway and thus lacked interest. The application was rejected; Mensura bears the roll fee of 200 euros, the contribution of 24 euros and a procedural indemnity of 770 euros.
Why does this matter?
This judgment is the mirror image of the case law on hidden sub-criteria — and the date makes it piquant: one day later the same chamber, in judgment no. 255.286, suspended an award by the MIDOW police zone because a criterion had been silently carved into weighted sub-criteria. The difference lies in how the specifications are drafted. Whoever lists assessment elements as non-exhaustive points of attention within one criterion retains broad discretion: not every element must be expressly discussed and related strengths may count. Whoever, by contrast, applies a weighting per element after the fact creates sub-criteria that should have been announced in advance. The judgment also confirms the Council’s limited role in qualitative assessments: it reviews marginally, and an ordinal scale that sets the gap between ‘very good’ and ‘good’ at 25 % lawfully magnifies small quality differences as long as nobody challenges the specification clause itself. For tenderers there is a practical lesson too: whoever claims their software matches the winner’s must substantiate that with concrete elements — a bare assertion does not suffice in urgency proceedings.
The lesson
For tenderers: read how the specifications structure their criteria. Where they list non-exhaustive assessment elements (‘such as, without being exhaustive’), expect the authority to assess broadly and make sure you substantiate every section — and document your strengths concretely, because in urgency proceedings only what you make plausible counts. If you find the effects of an ordinal scale (25 % per step) disproportionate, challenge the specification clause itself in time rather than the score gap afterwards. For contracting authorities: this judgment rewards careful drafting. Formulate assessment elements expressly as non-exhaustive and without their own weighting, compare the offers verbally section by section, and you stand strong — even after an earlier withdrawal and re-award in the same file.
Ask yourself
Do you know whether the list in the specifications contains weighted sub-criteria or non-exhaustive assessment elements — and what that difference means for the authority’s discretion? Can you substantiate every strength you claim (for instance your software’s capabilities) with concrete evidence? Do you realise that an ordinal scale can produce a 25 % gap per step and that you must challenge the scale itself if you consider it disproportionate? And do you calculate whether your criticism of one sub-criterion can even bridge the total points gap — otherwise it lacks interest?
About this database
The Council of State (Raad van State / Conseil d'État) is Belgium's supreme administrative court. In disputes over public procurement — from contract awards to tenderer exclusions — the Council of State is the final arbiter. The rulings in this database are summarised by TenderWolf in plain language, with practical lessons for tenderers and contracting authorities. View all rulings →