Docs

How to evaluate tender bids

Scoring bids without ending up with the cheapest supplier who cannot deliver — comparability, weighted criteria, and abnormally low bids.

Evaluation is where a tender either earns its structure or throws it away. The failure mode is not usually corruption. It is drift: criteria that shift once the bids are on the table and one of them is unexpectedly cheap.

Before you open anything

Your evaluation method should already exist, in writing, published in the tender document. If you are deciding how to evaluate after seeing bids, you are not evaluating — you are justifying.

Step 1 — Compliance check, pass or fail

Before scoring, check each bid against eligibility criteria and submission requirements:

  • Eligibility met — turnover, experience, registrations, certifications
  • All required documents present and current
  • Bid security or EMD furnished in the required form
  • Bid validity at least as long as required
  • Submitted before the deadline

This is binary. A bid that fails is out, and it is not scored. The temptation to let a promising bid through on a technicality is exactly what makes criteria meaningless for every future tender.

One caution: distinguish a material deviation from a clerical one. A missing signature page is usually curable by asking for it. A missing EMD, or a bid that proposes different scope, is not.

Step 2 — Normalise the bids

Bids are rarely directly comparable even when the request was structured. Before scoring, put them on the same basis:

  • Landed cost, not ex-works — freight, insurance, packing, unloading included
  • Taxes stated consistently across all bids
  • Currency, at a stated conversion basis if bids came in different ones
  • Payment terms — a bid at net 90 is worth less than the same number at net 30, and the difference can be quantified
  • Scope inclusions — if one bid includes commissioning and another does not, add the cost of commissioning to the second

Document the normalisation. It is the part most likely to be questioned later.

Step 3 — Score against published criteria

Price-led evaluation. Technical is pass/fail; among compliant bids, lowest wins. See L1. Appropriate when the specification is tight and capability genuinely does not vary.

Weighted evaluation. Technical and commercial scores combined at a published ratio. Score technical before opening financial bids — this is what the two-bid system is for.

A workable technical scoring frame:

CriterionWeightWhat you are assessing
Relevant experience30%Similar work, similar scale, verifiable
Proposed approach25%Do they understand the problem
Team and capacity20%Who does the work; current load
Timeline credibility15%Is the schedule achievable
Compliance depth10%Beyond the minimum

Score each criterion independently, ideally by more than one evaluator, before combining. Discussing a bid holistically before scoring it produces consensus around the first opinion voiced.

Step 4 — Interrogate the abnormally low bid

A bid materially below the others is the most important signal in the set. It usually means one of:

  1. The bidder misread the scope
  2. The bidder is buying the work and will recover through variations
  3. The bidder has a genuine structural advantage
  4. Your own estimate was wrong

Only the third and fourth are good news, and you cannot tell which without asking. Ask the bidder to confirm their understanding of scope in writing, and to break down the rate. A bidder who has genuinely found efficiency will explain it happily.

If the answer is unconvincing, most tender frameworks allow either rejection or a demand for enhanced performance security.

Step 5 — Check the gap between L1 and L2

A large gap is diagnostic. If L1 is 40% below L2, and L2 through L5 are clustered, the outlier is probably wrong rather than brilliant. If bids are evenly spread, the market is telling you your estimate was reasonable.

Step 6 — Record the decision

Who bid, what they bid, how each scored, what was normalised and why, who won, and on what basis. This record is the deliverable of the evaluation, not a by-product of it.

Then tell the unsuccessful bidders. Suppliers who receive a straight answer bid again; suppliers who hear nothing stop, and your field shrinks every cycle.

Common mistakes

Introducing a criterion mid-evaluation because a bid raised something you had not considered. Note it for next time; do not apply it now.

Negotiating with L1 after award. If price was negotiable, the tender was theatre.

Scoring price and technical together in one holistic judgement. The whole point of weighting is that the two are assessed separately and combined transparently.

Awarding to a non-compliant bid because it was cheapest. This is the single fastest way to make every future tender you run unserious.

On Bidancer

Bids arrive in the structure the host defined, so normalisation is largely done before evaluation starts. Contact details stay private until a bid is accepted, which means evaluation happens on the submitted bid rather than on which bidder has been calling most. The award and its outcome are recorded against both businesses, and feed the response score.

Was this page helpful?