Public procurement: evaluation stage
Public procurement: evaluation stage
Looking at what happens when evaluation goes wrong in public procurement with five lessons learned from case law, Jenny Beresford-Jones, Shailee Howard and Claire Gamage of Mills & Reeve presented this webinar as part of our Five in Twenty Five procurement series in September 2022.
Key questions and answers
Please note that although the information in our our 5 in 25 webinars and FAQs was correct at the date of recording, this is an area of law subject to development and change.
Do check if in doubt as to the latest position - you can email [email protected].
Please note that the responses provided represent the general views of the public procurement team at Mills & Reeve, however they should not be relied on or treated as a substitute for specific advice relevant to a particular scenario/matter. If you require specific legal advice, our procurement team would be happy to discuss this further.
FAQs
- When doing a restricted above threshold tender, when and how should we give feedback to suppliers that have not been shortlisted at selection stage, when they ask? Can we meet with this supplier when they request feedback?
There is no requirement to give feedback to deselected candidates at this point. However, it can be helpful to do so.
If deselected candidates have already received feedback at this point, then there is no need for them to be sent a standstill letter later in the process (which can reduce the administrative burden of drafting standstill letters).
Also, there is a limitation period for bringing procurement challenges, which is 30 days from the day that the candidate knew or ought to have known that it had grounds to bring a challenge. If you send feedback straight after deselection, it is likely that “time will start to run” at this earlier point (and deselected candidates may well be out of time by the time you get to contract award).
Feedback is usually given to deselected candidates via a letter that sets out their scores and explains areas of weakness on the SQ. There is no prohibition on meeting with deselected candidates, but authorities need to be aware that, to the extent that the meeting gives them “new” information, it has the potential to restart the clock on the 30-day limitation period.
- What does the acronym "RWIND" stand for?
Reasonably Well Informed and Normally Diligent Tenderer. Please see the recording for a full explanation of this concept. - Sometimes we ask for different pricing options as part of tenders or requests for quotations, eg, purchase, lease, fixed price or variable. How should we reflect this in evaluation criteria – do we need to state which pricing model we are going to base our scoring on? Concern is that scoring may vary between suppliers based on each option which could influence final award decision.
Yes, you do need to make it clear to the RWIND tender how you will evaluate the various pricing models – and if the intention is to choose one pricing model from amongst those offered, how you will go about making that decision. Otherwise, you run the risk of a challenge that you have used undisclosed criteria to make your decision. - Do you have any advice on inherent de facto advantage – in relation to incumbent providers – and the expectation on contracting authorities to manage that?
It is appreciated that it is simply not possible to completely level the playing field – the courts recognise that the incumbent is in a different situation to a “new” supplier – and that it enjoys a (fair) advantage because of this. The authority must however guard against giving any bidder (incumbent or otherwise) an unfair advantage. Be aware of discriminating unfairly against the incumbent – who cannot help the fact that it has some advantage thanks to its position. - Can the wording around exclusion on pass/fail be framed using the words "could result or may result" in exclusion? Or should we be firmer in our wording?
You may wish to consider using wording such as “the Authority reserves the right to exclude …” which then preserves the right to go back to the bidder and clarify. However, this depends on what the question is. If the authority is confident that the question concerned is asking the bidder to confirm/demonstrate something which is "non-negotiable", then in the circumstances our advice is to use the wording "will" exclude, as this avoids a later debate about whether the authority should have excluded the bidder concerned or not. If the authority wishes to retain some discretion as to whether to exclude, it may be pertinent to go back to the bidder concerned to provide them with a further opportunity to clarify their answer (noting that you will need to check the ITT in terms of what was stated in respect of clarifications as per the Manova case). In asking such a clarification, the Authority should be clear in its request that, if the bidder does not respond by the stated deadline or provides an unsatisfactory response, exclusion will then be the result. - We use the word "may" exclude with detail. Is this too ambiguous and should we state "will" exclude?
Please see our response to the question above. - If a fail doesn’t necessarily mean exclusion, doesn’t that go against RWIND?
If the ITT/procurement documents are very clear about the consequences of a fail (whatever these consequences may be), the RWIND tenderer test is likely to be met.
- Is clarification related to after tender submission?
Yes, in the context we are discussing it here in Lesson 2, we are thinking of a situation where a bid has been submitted but is unclear or incomplete – to what extent may or must the authority seek clarification? It is also usual for the terms and conditions of the procurement to include a timetabled pre-tender “clarification phase" – during which bidders can ask questions to clarify their understanding and authorities issue a spreadsheet of responses to all bidders (to ensure equal treatment). - Duty to clarify - If the ITT states a timetable that includes for clarifications are you then bound to follow that process?
Yes, the principles of transparency and equal treatment require that you do stick to whatever process you have stated. If you need to change the process, then you would at least need to give all bidders notice and analyse whether the change could be unfair to any bidder. - How do the judgments in the cases you mention align with Regulation 56(4) (the regulation which says you can ask for information if missing, obvious, etc.)?
In our view, Regulation 56(4) essentially codifies the case law that we discussed in Lesson 2 – and confirms that, whatever approach an authority chooses to take, it must always respect the principles of transparency and equal treatment. - Can bidders resubmit the commercial element of tender if a significant amount of time has passed since original submission?
This would need to be assessed on a case-by-case basis – looking at the reason for the delay and following the principles in the Manova case (discussed in Lesson 2). - If a supplier has referenced an appendix in their submission but not attached it to their submission, can the appendix be requested so it can be considered in the evaluation?
Again, this would need to be assessed on a case-by-case basis, and the approach is likely to vary depending on what information the appendix contains (and what the authority has said in the procurement documents about clarifications). For example, is this information which obviously pre-dated the tender deadline? (eg, the balance sheet example we talk through in the webinar in respect of the Manova case – Lesson 2). - What counts as a clarification? For example, if tenderer has answered "no" to a statement but the supporting documents/evidence they have provided support "yes" and therefore it is assumed they should have said "yes" – would this count as a clarification?
This example sounds rather like the situation in the Tideland case that we discussed in Lesson 2. Where there is likely to be an easy explanation for and resolution to the confusion, then the authority may have an active duty to clarify. In this case, it looks like the bidder simply ticked “no” when it meant “yes”. However, each case is individual and it would be relevant here to understand more about the statement being evidenced in order to give a definitive view on whether the authority would need to clarify here.
- To what extent is it permissible to evaluate bids comparatively? I have had challenges alleging that this is not permissible and that each tender should be evaluated on its own merits.
In general, bids should not be evaluated comparatively. There is a balance to be struck between ensuring consistency in the approach to scoring and evaluating each bid in isolation. This means that each bid should be evaluated against the published scoring criteria, but ensuring that one bidder is not "scored down" for something where another bidder has not (as this could expose the authority to allegations of unequal treatment and manifest error). - Do we know what proportion of challenges are linked to evaluation?
We have not analysed the numbers, but it is certainly the case that almost all procurement challenges include allegations that the evaluation has been carried out unlawfully or that it is simply wrong. Challengers want to show that they would have won the contract had the evaluation been carried out differently. - Any advice when it is clear you are running out of evaluators? Most councils are light on resources and "bums on seats" might be the only thing to do.
We appreciate that resourcing is always difficult. However, ultimately, if an authority does not identify a sufficient number of suitably qualified evaluators, this increases legal risk further down the line that the procurement will be challenged. You may also wish to consider appointing an odd rather than even number of evaluators, as this can be helpful where the evaluation team reach an impasse in terms of agreeing a score, it can then be recorded that the majority of evaluators were in agreement. - How is past performance accounted for in evaluations? An ITT response may sound perfect but, if we know their previous performance is poor, can/how do we reflect this in scoring a procurement process?
A bidder’s past performance should be evaluated at SQ (selection) stage, not at ITT (award) stage. At ITT stage, you must evaluate the bid received against the published award criteria. To bring in a past performance assessment at ITT stage would be a breach of the regulations. It may be helpful to note that, during the SQ stage, poor prior performance is a discretionary ground for exclusion. However, such performance must have led to early termination, damages or other comparable sanctions (see Regulation 57(8)(g) of the Public Contracts Regulations 2015). - On evaluation, is it not possible for a group evaluation from the outset, sometimes separate scoring and moderation isn’t necessary?
It is not necessarily wrong to evaluate as a group, but the Aquatron Marine case does show how a consensus evaluation of this kind has the potential to undermine the integrity of the procurement. It is arguable that, where evaluators evaluate as a group, there is less careful individual scrutiny of the detail of the bid as against the published award criteria, which could expose the authority to allegations of bias or manifest error. - What would you say is a reasonable number of evaluators?
This really depends on the timeline, the nature and the complexity of the procurement. You would of course need enough appropriately qualified evaluators to meet your deadlines (building in sufficient time for them to do their job properly). Having an odd number of evaluators is better than even numbers as this can avoid an impasse and allow a majority decision to be reached. - Although the bid should be assessed against the specification and published criteria, a sense check of early scoring is often needed eg, if the first bid evaluated appeared to be acceptable, but later submissions were better, is it acceptable to adjust the earlier scores?
We would advise against this approach. You must always score each bid (1) individually and (2) against the published criteria. You should never score a bid comparatively against another bid. That is not to say you might not revisit a score later in the day if you feel there has been an inconsistency in how you have applied the published criteria to each of the bids, but this is different to running a comparative evaluation (which is not permitted). - We have had a situation where the client has insisted the tender documents state all tenderers must provide all the required documents by the tender deadline. This resulted in the highest scoring tenderer being eliminated as they did not provide a copy of their insurance certificates. Is this right?
This would depend on what you have stated in your ITT/procurement documents. If you have stated you will exclude in these circumstances, then you would need to exclude. If you have a reserved a discretion, then you would apply the principles in Tideland (duty to clarify) and Manova (flexibility to clarify) to see if clarification was required and/or possible. - If the winner of a tender via M.E.A.T. has come in well over budget, and this leads to the abandonment of the tender in order for the authority to re-review the specification, is this likely to be challenged?
A full answer would depend on the specific facts of the case and how it came about that the procurement needed to be abandoned. We have certainly seen decisions to abandon being challenged in the courts, usually where an authority has not been sufficiently transparent or the reasons behind the abandonment are argued to be as a result of an "irrational" decision (see Amey v West Sussex for example). Our advice would be to ensure that you have included clear wording within the ITT which states that the authority reserves the right to abandon the procurement in certain stated scenarios and indeed for any reason. This may assist later down the line if a bidder challenges the authority if the procurement is abandoned, however the reasoning relates to a scenario which was already envisaged in the ITT. If you have a maximum budget, our suggestion would be to make this very clear within the tender documentation to enable you to exclude a bidder which proposes a price above this. While there may be perceived disadvantages to this (eg, bidders may all propose a price which meets the budget, and not below), the authority can still score pricing which falls below the maximum threshold stated, which does help to retain an element of competitive tension between the bidders. - I have previously been told that calculating an average of the scores isn’t permitted as the score given needs to align with the scoring table (ie, 1-5). I don’t believe this is true and, provided you state in ITT docs that you will be using an average (and to a set no. of decimal points!) it is okay to use. Can you please confirm?
Each evaluator needs to award his or her own score in alignment with the scoring methodology/table. You may then go on to take the approach of averaging scores so as to achieve one outcome – but this must be stated in the ITT documents, as you say. Our view is that, even though you may have stated this, this approach is more susceptible to the risk of challenge. This is because, if the scoring was challenged and disclosure of the evaluation record was requested (which is usually the case), the rationale for the final scores is likely to consist only of the individual scoring sheets used by the evaluators. A moderation process, which aims to agree a final consensus score is usually a more considered exercise and flushes out potential misunderstandings by individual evaluators. If an average is taken based on individual evaluators’ scores, the final score could be largely based on an individual evaluator scoring too high/too low against the published scoring criteria, and therefore it is very possible that a bidder could allege a manifest error has occurred - Does the ITT have to state that you are conducting a moderation process?
This issue is not addressed in the Public Contracts Regulations 2015 or case law. However, our view is that authorities should expressly state that a moderation process will be conducted. Otherwise, there is a risk that a challenger could argue that the authority is not following the process it said it would, and advance an argument that this lacks transparency. - Regarding excluding the highest bidder. I can understand you need to tell them there may be a ceiling (for example the authority may have a limited budget) but do you have to tell them the amount of that budget? I’d have thought doing the latter may affect the opportunity to get value for money if all the bidders bid near the ceiling.
It is not necessary to tell them the amount of the budget. However, if you wish to mark down bids that are, for example, priced at over 90% of the budget, then it is hard to see that this criterion can be said to have much meaning for bidders (nor pass the RWIND test) unless the authority discloses what the budget is. A better approach might be to weight lower price offers appropriately. Please also see our response to Question 23 which addresses this issue. - I do keep a lot of internal emails which some colleagues think is too much. What would be the adequate amount?
Our advice is to keep all internal emails, notes, records and communications relating to the procurement. If there is a procurement challenge, the court will usually expect to see the “workings out” that sit behind the results in the evaluation/moderation report. Indeed, the authority was criticised in the EnergySolutions v NDA case for shredding records. It seems that deleting/withholding emails could therefore be criticised by the Courts. - I struggle with getting the evaluation panel to provide meaningful reasons/explanations for their scores. Is there any good guidance available?
There is no official guidance as far as we are aware (although we understand that the Government is drafting a wide range of guidance and templates to accompany the new Procurement Bill when it becomes law (expected to be later in 2023) so this may change). However, we appreciate the issue and often find that evaluator training in advance of the scoring exercise, which points to examples from real life case law, can help focus evaluators’ minds. A good rule of thumb is to advise evaluators to stick to each bullet point on the question and to forensically apply the scoring methodology to each part of the question, making sure the evaluators address each one. Ensure the descriptors used match the score (eg, don’t describe a score as “good” but then only award it 2/10!). - What potential is there to re-open a moderation before issuing standstill notice to remedy a manifest error in scoring/defect in reasoning?
This may sometimes be required if there are clear errors identified in the final scores/rationale for scores to avoid allegations of manifest error (eg, in scenarios which are similar to what happened in the Braceurself case - where the evaluators had simply not appreciated the difference between a stair lift and stair climber). If this requires an extension to the timetable, it may be prudent to write to bidders to explain that the evaluation process is taking longer than originally anticipated and, therefore, it is hoped that the award decision will be announced on [insert date]. It may only be necessary to re-moderate the criteria where the authority believes errors have been made. However, we would stress the importance in retaining all documentation/records, including those from the original moderation process. In the event of a challenge, the Court will expect such records to exist and for the authority to explain why a further moderation exercise was conducted. In our view, a Court is unlikely to criticise an authority for re-performing the moderation process/certain aspects of it, if this was in the interests of ensuring that the final scores were correct. - Is averaging scores an acceptable method if explained in the tender document and, if so, do individual scores and comments need to be disclosed within feedback?
See our response to question 24 above. If you have stated that scores will be finalised for each question by way of taking an average of each evaluator’s initially awarded score, then this final score is “the score” and it is this that you would need to disclose in the standstill feedback.
- If the lowest priced tender is worryingly low (in terms of concern they have not included everything or have cut corners which will cause further costs later in the contract) does their tender still have to win as it is the lowest (and the price was the criteria)?
Regulation 69 PCR 2015 is relevant here. If the authority considers that the tender may be abnormally low, then it has to duty to investigate this and to give the bidder an opportunity to explain why this may be the case. The bidder may only be excluded if the authority is not sufficiently satisfied with this explanation. The process needs to be fully documented in the Regulation 84 report to evidence that the authority properly followed procedure and fully considered the bid concerned. - Is there an objective test that can be applied to identify a potentially abnormally low tender?
Unfortunately, there is not. The question of what amounts to an abnormally low tender depends very much on the particular procurement. The case law shows that the courts have preferred to retain the flexibility to make this assessment on a case-by-case basis.