Bulletproof vests for the police: read the selection guide carelessly and you lose the shortlisting — and the Council of State will not rescue you
The Israeli manufacturer Source Vagabond Systems scored lowest (43.30/100) in the third selection phase for the federal police’s framework agreement for bulletproof vests and was the only one of the five suitable candidates not invited to tender; the Council of State found the shortlisting to four candidates, the rule-of-three formula not explained in advance and the zero score for ‘study and research’ for lack of an in-house ballistic laboratory prima facie lawful, and dismissed the application for suspension under extreme urgency.
What happened?
In spring 2022 the Procurement service of the federal police launched a negotiated procedure with publication for a multi-year framework agreement for individual bulletproof vests and body armour for the integrated police. As security equipment was involved, the Act of 13 August 2011 and the Royal Decree of 23 January 2012 on defence and security procurement applied. The selection guide organised qualitative selection in three phases: first the exclusion grounds, then minimum requirements (an average annual turnover in bulletproof vests of at least 500,000 euros and a list of the main deliveries of at least 100,000 euros each, totalling at least 200,000 euros), and finally — only if more than three candidates passed those thresholds — an evaluative phase with seven weighted criteria, after which ‘only the best candidates’ would be retained: at least three and at most five, provided their candidacy was found ‘sufficiently qualitative’. Seven undertakings applied. Five met the minimum requirements, including Source Vagabond Systems. In the third phase things went wrong for the Israeli manufacturer. For the technical-capacity criterion (20 %) the police added up the amounts of all listed deliveries above 50,000 euros — 32.8 million euros for Source Vagabond — and compared them by a rule of three against the highest total; that gave 8.66/20, against 20/20 for Seyntex. For ‘study and research’ (20 %) Source Vagabond received 0/10: it was the only candidate without an in-house ballistic laboratory and it described no innovative projects, only developments carried out for specific orders. Sioen scored 10/10 here, Ambassador Arms 7, Mehler Vario 6, Seyntex 4. For ‘technicians or technical bodies’ (15 %) the police saw in the file one employee for the ballistic part, which despite cooperation with several laboratories yielded only 4/10. The final ranking: Sioen 83.90, Ambassador Arms 80.80, Seyntex 58.62, Mehler Vario 54.22 and Source Vagabond 43.30 out of 100. On 3 October 2022 the Minister of the Interior decided to invite the four best-ranked candidates to tender and not Source Vagabond. On 18 October 2022 it applied for suspension under extreme urgency. Its single plea had five limbs. First: the notice mentioned a maximum of 99 selected candidates and the guide a maximum of five; with exactly five suitable candidates, the third phase should not have been applied. Second: the assessment method — in particular the formula with the parameters Pmax and Pmin — was not in the guide, was unusual and amounted to a repetition of the turnover criterion; moreover it was not clear that all references above 50,000 euros had to be included in annex F, so it had listed only its main deliveries. Third and fourth: the zero score for study and research and the 4/10 for technicians were unsustainable — an external laboratory is worth no less than an in-house one, and it had listed seven technicians. Fifth: the total scores were therefore unreliable. Councillor of State Patricia De Somere, acting president, rejected everything. The guide did justify the shortlisting: the third phase applied as soon as more than three candidates met the minimum requirements, and a contracting authority may, in choosing three to five best-ranked candidates, limit itself to those scoring more than half the points, even if the guide states no minimum score. The ‘maximum 99’ in the notice was a clerical error without legal consequences. A rule of three for quantitative criteria and a descriptive assessment with an overall score for the others are customary methods that fit within the guide; candidates did not need to be informed of them in more detail beforehand, and with Pmin (50,000 euros) and Pmax (the highest total) the calculation could be checked. There was no repetition of the turnover criterion: that looked at average turnover over three financial years, this criterion at deliveries above 50,000 euros over five years, and article 73, 3°, of the Royal Decree expressly treats experience with previous deliveries as proof of technical capacity, so more deliveries may also mean more capacity. Above all: Source Vagabond had submitted an identical list for the minimum requirement (annex E, ‘main’ deliveries of at least 100,000 euros) and for the capacity criterion (annex F, ‘the deliveries’ above 50,000 euros), whereas the other candidates had understood the distinction. A candidate is itself responsible for drawing up its application correctly and cannot correct its misreading with a supplemented annex attached to the application to the Council. The zero score also stood: Source Vagabond did not dispute that it alone lacked an in-house ballistic laboratory, and the expert documents on innovative research it produced only in its pleading note should have been attached to its candidacy. The reasoning did differ from that for Seyntex, which did have its own laboratory. The criticism of the technicians criterion lacked interest: even three extra points out of fifteen would not close the gap with the fourth-ranked candidate. The application was dismissed, with a roll fee of 200 euros, a contribution of 24 euros and a procedural indemnity of 770 euros borne by Source Vagabond.
Why does this matter?
Shortlisting — limiting the number of candidates allowed to submit a tender — is a phase that bidders often take less seriously than the award itself, yet they can fall just as hard at that hurdle. This judgment sharpens three points. One: a contracting authority that announces in its guide an evaluative third phase with a range of three to five ‘sufficiently qualitative’ candidates may fill in that range as it sees fit — here by retaining only those scoring more than half the points — without having to fix a minimum score in advance. Two: the Council of State does not require every scoring formula to appear in the guide. A rule of three for quantitative criteria and a descriptive assessment for qualitative criteria are so customary that candidates must expect them, as long as the criteria and their weighting are known. Whoever complains that a method surprised him must moreover show concretely what he would have done differently. Three, and this is the sharpest lesson: the court reads the guide literally and expects the candidate to do the same. The difference between ‘the main deliveries of at least 100,000 euros’ and ‘the deliveries above 50,000 euros’ was not a detail but the hinge on which 11 points out of 20 turned. Whoever withholds information in the selection phase or produces evidence only before the Council of State gets no second chance. The judgment also illustrates how the interest requirement works in extreme-urgency proceedings: criticism that, even if well founded, cannot reverse the ranking is no longer examined.
The lesson
Treat a selection guide with an evaluative shortlisting as a full competition, not a formality. Read each annex separately and watch the exact wording: if the minimum requirement asks for ‘the main deliveries’ and the capacity criterion for ‘the deliveries above 50,000 euros’, submit two different lists and put everything that qualifies in the second — with a rule of three, more relevant turnover means a higher score. Put everything that substantiates your capacity — laboratory equipment, research projects, expert reports, the full list of technicians with their tasks — in the candidacy itself; what you bring later in a pleading note does not count. If in doubt about the assessment method, ask for clarification before submission, because complaining afterwards that a formula was ‘unusual’ only succeeds if you can show you would have prepared your file differently. As a contracting authority: describe the phases, the criteria, their weighting and the range of the number of selected candidates clearly in the guide and make sure the notice is consistent with it — a ‘maximum 99’ was forgiven here as a clerical error, but that is no guarantee. Give specific reasons per candidate so that different scores for seemingly comparable files remain defensible.
Ask yourself
Have you laid the selection guide and all its annexes side by side and checked whether the minimum requirements and the capacity criteria ask for different (broader) reference lists? Do you know that in a shortlisting of three to five candidates a contracting authority may retain only those scoring more than half, even without a predetermined minimum score? Is all your evidence on laboratories, research and technicians in the candidacy itself, or are you counting on supplementing it later? If you challenge an assessment method, can you say concretely what you would have done differently had you known it in advance? And as a contracting authority: does the maximum number of selected candidates in your notice match that in your guide, and is your reasoning per candidate sufficiently distinctive?
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 →