What is bid evaluation, and how do you run it fairly?
Short answer
Bid evaluation is the stage of a tender where the buyer checks each bid, scores it against criteria published in advance, and picks the bid that offers the best value.
Key takeaways
- Bid evaluation has more steps than scoring: open the bids, check they are complete, check the bidder qualifies, score quality, evaluate price, test low prices, agree the scores, then record and explain the decision.
- Score only against the criteria and weights published in the tender. US federal rules and UK and EU public procurement law all require this, and private buyers gain the same protection against a choice made first and justified later.
- The way price is turned into points can change the winner. With relative price scoring, a bid that loses can still decide which of two other bids wins; the worked example shows how.
- Use at least two evaluators who score alone first, then a moderator who leads them to one agreed score per answer. An average of disagreeing scores is not a consensus.
- Clarification fixes an unclear or missing point in a bid; it never lets a bidder improve its offer. A very low price is a question to ask the bidder, not an automatic win or an automatic rejection.
What is bid evaluation?
Bid evaluation is the process of checking and comparing the bids received in a tender against criteria published in advance, to select the bid that offers the best value. It covers compliance checks, quality and price scoring, the treatment of very low prices, the agreed final scores and the record of the decision.
Take a common case. A city council tenders a four-year contract for street lighting maintenance and receives six bids. One arrives ten minutes after the deadline. One leaves the price schedule blank. Two evaluators disagree on the quality of a third bid’s safety plan. The cheapest bid is 30% below the council’s own cost estimate. Each of these is a bid evaluation question, and each needs a rule decided before the bids were opened.
The UK Cabinet Office’s bid evaluation guidance note, written in 2021 before the Procurement Act 2023, puts the aim plainly: identify the most economically advantageous tender, which “may not be the lowest cost bid”, and run a process that is documented well enough to stand up to internal and external review. “Bid”, “tender” and “proposal” mean the same thing here. How evaluation fits into a request for proposal is summarised in the RFP process; how the same comparison works for a first purchase from a new vendor is in the vendor procurement process.
What are the stages of bid evaluation?
Most evaluations follow the same order, whether the buyer is a government agency or a company. Simple purchases merge some stages; large or risky ones take each in full.
- Receive and open the bids. Record every bid received and the time it arrived. Bids that arrive after the deadline are normally rejected unless the published rules say otherwise. On an e-sourcing platform, bids stay sealed until the deadline and the system keeps this record.
- Check compliance. Confirm each bid is complete and answers what the tender asked: every form signed, the price schedule filled in, no conditions added that change the requirement. EU law allows an award only after the buyer has verified that the tender complies with the requirements set in the procurement documents.
- Check the bidder. Apply the pass or fail conditions: legal grounds for exclusion, financial standing, insurance, certificates, experience. A bidder that fails a condition is out, however good its offer.
- Score quality. Each evaluator reads the answers alone and scores them against the published criteria and scoring guide, with a written reason for each score.
- Evaluate price. Check the arithmetic, add any extra costs the tender says will be counted (such as running or exit costs), and turn the result into points with the published method. Finance staff often do this apart from the quality evaluators.
- Test unusually low prices. Where a price looks too low to deliver the work, ask the bidder to explain it before deciding anything.
- Moderate. A moderator brings the evaluators together to agree one score per answer and records the reason for every change.
- Combine and recommend. Add the weighted quality and price results, rank the bids and write the award recommendation for approval by someone with the authority for that amount.
- Notify, debrief and record. Tell every bidder the outcome and its own scores, keep the evaluation report and, in public procurement, wait for the standstill period before signing.
The order matters. A buyer who scores quality before checking compliance spends days on a bid that must be rejected anyway. A buyer who looks at prices before scoring quality risks letting the price colour the quality scores, which is why many teams keep the two apart until the end.
Which award methods can a buyer use?
The award method is chosen and published before the bids arrive. Five are common. In US federal buying, sealed bidding awards the contract to the responsible bidder whose bid, conforming to the invitation, is most advantageous considering only price and price-related factors. Negotiated buying can use a trade-off or lowest price technically acceptable. EU law requires the most economically advantageous tender, which can be based on price or cost alone or on the best price-quality ratio. The UK Procurement Act 2023 now calls it the most advantageous tender; GOV.UK guidance says this “is not a change in policy” and that price alone is allowed but “unlikely to be suitable for most contracts”.
| Method | How the winner is chosen | Fits when | Watch for |
|---|---|---|---|
| Lowest price | Every compliant bid from a qualified bidder is ranked on price only | The requirement is fully specified and quality above the minimum adds no value (standard goods, simple works) | Bids that meet the letter of the specification but cost more to run |
| Lowest price technically acceptable | Each bid is rated acceptable or not on quality; the cheapest acceptable bid wins, with no trade-off | A clear minimum standard exists and extra quality is not worth paying for | A pass mark set too low lets weak bids through |
| Quality threshold, then price | Bids must reach a minimum quality score; the cheapest of those that pass wins | The buyer knows its minimum quality and is indifferent to quality above it | The threshold must be published and set high enough |
| Weighted price and quality | Quality and price are each scored and weighted, for example 60% quality and 40% price; the highest total wins | Quality varies between suppliers and matters to the result (services, IT, consulting) | The price scoring formula can change the ranking (see the next section) |
| Fixed price, compete on quality | The price or budget is set in the tender; bidders compete on quality only | The budget is fixed and the buyer wants the most for it | The fixed price must be realistic or nobody bids |
Pick the simplest method that reflects what matters: price only for fully specified needs, a quality threshold or weighting when suppliers differ in ways that change the result.
Private buyers are free to choose any of these; weighted scoring is common for services. The trade-off and lowest price technically acceptable approaches in US federal proposals are explained in how are RFP responses evaluated.
How should price be scored?
Price is usually scored in one of three ways. Relative price scoring gives full marks to the lowest bid and fewer to the others in proportion, for example lowest price divided by the bid’s price. Price per quality point divides each bid’s price by its quality score; the lowest cost per point wins. Benchmark scoring compares each price to a figure set by the buyer, such as its own cost estimate.
The UK guidance note says relative price scoring “should be treated with caution”, because it can reward bidders who aim to win on price alone. The worked example shows a second problem. Three fictional bids compete with 60% for quality and 40% for price, and the price score is the lowest price divided by the bid’s price, times 40. Then bid B withdraws.
| Bid | Quality score (out of 100) | Price (€) | Quality points (60%) | Price points, three bids (40%) | Total, three bids | Price points, B withdrawn | Total, B withdrawn | Price per quality point (€) |
|---|---|---|---|---|---|---|---|---|
| A | 80 | 100,000 | 48.00 | 32.00 | 80.00 | 40.00 | 88.00 | 1,250 |
| B | 60 | 80,000 | 36.00 | 40.00 | 76.00 | — | — | 1,333 |
| C | 90 | 120,000 | 54.00 | 26.67 | 80.67 | 33.33 | 87.33 | 1,333 |
With three bids, C wins; when B, which ranks last, withdraws, A wins. Nothing changed in A or C. Price per quality point ranks A first in both cases.
The reason is that relative scoring compares bids with each other, not with a fixed standard. Price per quality point and benchmark scoring judge each bid on its own, so the ranking of two bids does not depend on a third. Price per quality point has its own limit, which the UK note also gives: very different bids can reach the same cost per point. Whatever the formula, test it on sample bids before the tender is published, and publish the formula with the tender.
Score the price that will be paid, not only the headline price. If bidders may add assumptions (for example “price assumes the client provides site access at night”), the tender needs to say how each assumption will be priced in the evaluation. Otherwise the bid with the most assumptions looks cheapest and costs most.
Who should evaluate, and how do you keep scores consistent?
Use a small panel of people who know the subject, have signed a conflict of interest declaration and have been briefed on the scoring guide. The UK guidance note recommends at least two evaluators per bid, scoring independently, the same evaluator for all the answers to one question, a separate moderator who does not score, and an evaluation of the price model apart from the technical answers.
- A scoring guide with clear bands. Describe what each score means, for example 0 for no answer and 100 for an answer that fully meets the need with evidence. With too few bands, most bids get the same score and the evaluation cannot separate them; the guidance note calls this bid-bunching.
- Reasons for each score. A score without a written reason cannot be moderated or explained to a losing bidder.
- Only what is in the bid. An evaluator who scores a bidder lower because of a bad experience on another contract is scoring something that is not in the bid. The moderator removes such comments.
- Consensus, not an average. When two evaluators give 40 and 80 to the same answer, the moderator asks each to explain and leads them to one score they can both justify. The guidance note says the agreed score “should not simply be an average”.
In the street lighting example, one evaluator gave the safety plan 40 because it did not mention night work, the other gave 80 because it was detailed. At moderation, the first evaluator points to the requirement for night work; both agree on 60 and the reason is recorded. That record is what the council will show if the bidder challenges the score.
What can you ask bidders during evaluation?
You can ask a bidder to clarify, complete or correct information that is unclear, missing or obviously wrong, in writing and on equal terms for all bidders. EU law allows the buyer to ask bidders to “submit, supplement, clarify or complete” information, provided equal treatment and transparency are respected. You cannot let a bidder change its price or improve its solution, unless the published procedure includes a negotiation stage.
Two examples show the line. A bidder whose insurance certificate expired last week is asked to send the current one: that is clarification. A bidder whose answer on response times was weak is asked whether it could do better: that is a chance to improve the bid, and other bidders did not get it. The UK guidance note adds that evaluators send their questions through the procurement team rather than contacting bidders themselves, so that every question is checked first. Negotiation after bids, where the procedure allows it, is covered in the negotiation step of the RFP process.
What do you do with an abnormally low bid?
Ask the bidder to explain the price before you decide. A price that looks too low may come from a better method, cheaper sourcing or a mistake; it may also mean the bidder will cut corners or ask for more money later. The UK guidance note lists what follows a contract won at a price that cannot cover the work: cost cutting, change requests and extra funding to keep the service running.
Public rules set the steps. In the EU, Article 69 of Directive 2014/24/EU requires the buyer to ask the bidder to explain a tender that appears abnormally low, and allows rejection only where the evidence “does not satisfactorily account for the low level” of price or costs. Under the UK Procurement Act 2023, the buyer must first notify the supplier and give it a reasonable chance to show it can perform at that price before disregarding the tender. Private buyers are not bound by these rules, but the same check protects them: ask how the price was built, compare it with your own cost estimate, and record the answer.
What records and feedback does bid evaluation need?
Keep a record that lets someone who was not in the room follow the decision: the published criteria and method, each evaluator’s scores and reasons, the moderation notes, the price calculations, the clarification questions and answers, and the approved recommendation. US federal rules require strengths, deficiencies, significant weaknesses and risks found in proposals to be documented in the contract file. The UK guidance note asks for an evaluation report at the end of every evaluation.
Then tell each bidder the result. Good feedback gives the bidder its own scores, the winner’s scores and the main reasons, with examples from its bid, so it understands what to improve. Public buyers in the EU must wait at least 10 calendar days after sending the award decision electronically before signing, and in the UK eight working days after the contract award notice, so that a losing bidder can challenge the decision. Feedback that explains the scores clearly is also the best way to avoid a challenge.
Where does bid evaluation usually go wrong?
- Criteria changed after the bids arrive. A new criterion or a new weight added during evaluation makes the result unfair, and in public procurement open to challenge.
- A price formula nobody tested. The ranking then depends on the formula, as in the worked example, and the team only finds out after the award.
- One evaluator for everything. Nobody can check the scores, and one person’s view becomes the decision.
- Scores without reasons. The losing bidder asks why, and the buyer cannot say.
- A low price accepted without questions. The contract starts well and then fills with change requests.
- Clarification used as negotiation. One bidder gets a second chance and the others do not.
Most of these are decided before the tender is published: the criteria, the weights, the price formula, the scoring guide and who will evaluate. Writing them down in one evaluation plan, and testing the plan on sample bids, prevents most of the problems above. A bid comparison template that holds the scores and the formula is planned in this toolkit’s tools.
Questions
Is bid evaluation the same as tender evaluation?
Yes. “Bid”, “tender”, “offer” and “proposal” describe the same thing in most procurement rules, and the evaluation stage works the same way. US federal buying uses “bid” for sealed bidding and “proposal” for negotiated contracts; UK and EU public buyers usually say “tender”.
What are common bid evaluation criteria?
Price or whole-life cost, and quality criteria such as the technical solution, delivery plan, service levels, staff and experience, risk, and social value or sustainability. Criteria must relate to what is being bought and be published with their weights. Pass or fail conditions, such as insurance or certificates, are checked before scoring.
Does the lowest bid always win?
Only when the published method is lowest price. With a quality threshold or a price and quality weighting, a more expensive bid can win if its quality is worth the difference under the published weights. UK guidance notes that the most advantageous tender may not be the lowest cost bid.
How long does bid evaluation take?
It depends on the number of bids, the stages and the people available. The UK guidance note says applying the criteria and finalising the award can take as little as a month for a simple process and often longer for complex ones. Plan the evaluators’ time before the tender is published.
Can bid evaluation be done in software?
Yes. E-sourcing tools keep bids sealed until the deadline, let each evaluator score alone, apply the price formula and keep the record. They do not choose the criteria or the formula; the evaluation plan still has to be written and tested by the team.
New guides on the strategic sourcing line
One short email when a guide on this line is added or revised. Nothing else.
Coming soon: sign-up opens with the first email.