Bid Evaluation Matrix

What this tool does

CM8-274 is a working bid evaluation matrix for the quotes and tenders you receive — a subcontract package, an equipment purchase, a service contract, an annual supply agreement. You fix the criteria and their weights, score every bid against every criterion with the evidence written next to the score, and the tool builds the comparison: totals by bidder, what drives each total, prices side by side with scores, and an award summary you can put in front of a director, an auditor, or the bidder who lost.

The cheapest quote is a number. The best bid is a judgement — about capability, quality, capacity, risk and service as well as price — and a judgement is only defensible when it is structured: criteria set in advance, weights fixed before the envelopes were opened, scores tied to facts. That structure is what this matrix records. It protects you twice over: from buying the wrong bid, and from being unable to explain why you bought the right one.

Everything runs inside this single file. There is no account, no upload and no network request of any kind, so bidders' prices and your scoring of them — commercially sensitive in both directions — never leave the computer you are using.

Set the criteria and weights before you open the bids

This is the integrity rule, and it is not negotiable. Decide what matters and how much it matters before you see a single price. Once you have read the bids, every weight you choose is contaminated: you know which weighting makes your favourite win, and so does anyone who later checks the dates. Weights adjusted after opening are the single most common way an evaluation stops being an evaluation and becomes a justification — and it is precisely what an auditor, a losing bidder's lawyer, or your own finance director will probe first. Write the criteria, fix the weights, record who agreed them, and only then open the bids. If a genuinely new criterion emerges from what the bids reveal, record the change and the reason openly — do not slide it in silently.

Criterion groups — and the price-weight honesty check

Every criterion belongs to one of four groups: commercial (price, payment terms, total cost), technical (capability, quality, capacity), risk (financial health, single-source exposure, location) and service (lead time, flexibility, support). The groups drive two of the charts. The stacked bars show what actually drives each bidder's total — two bids on 240 points are not alike if one earned them commercially and the other technically. The donut shows where the weight sits, and it exists to keep you honest: if commercial criteria carry 80% of the available weight, this is a price competition, and the right response is to say so openly, not to pretend a rounded evaluation happened. Either answer can be correct. Hiding the answer never is.

Scoring with evidence

Score each bid 1–10 on each criterion, and write the basis in the evidence field: the quote figure, the audit result, the reference call, what you saw at the site visit. The tool refuses a score of 8 or more, or 2 or less, with an empty evidence field — strong opinions are exactly the ones that need facts behind them. Define what a 10 means per criterion before scoring (for price, the lowest compliant quote is the usual anchor), score one criterion across all bidders rather than one bidder across all criteria — it keeps the scale consistent — and keep the same weight for a criterion on every bidder's rows. The weight belongs to the question, not to the answer.

Compliance is a gate, not a score

A bid that does not meet the specification is not a cheaper alternative; it is an offer of something you did not ask for. If you let a high score on price compensate for non-compliance on substance, you have not run a competition — you have changed the specification for one bidder without telling the others, which is unfair to every compliant bidder who priced the real requirement. So compliance here is a tick, not a number: untick it on every row of a non-compliant bid, and the tool will refuse to mark that bid as the winner. Keep the scores — they tell you what the market can do, and whether your specification is demanding more than it needs to. If a non-compliant bid is genuinely attractive, the honest route is to revise the specification and re-invite everyone.

Total cost, not unit price

The number on the quote is rarely the number you will pay. Payment terms have a financing cost; tooling and set-up charges land once but land on you; carriage, minimum order quantities and price-break thresholds shape what you actually order; a longer lead time is stock you must hold. Before scoring the price criterion, normalise every quote to the same basis — the same annual volume, the same delivery point, the same term — and score the normalised figure. Enter that figure in the quoted-price field and say in the evidence field how you built it. Comparing one bidder's ex-works price against another's delivered price is the oldest error in purchasing, and the matrix will faithfully amplify it.

The value-for-money index and its limits

The award summary computes points per 1,000 units of price for every bid with a quoted price — a quick check on whether the premium bid earns its premium. Treat it as a lens, not a verdict. It is only comparable within one RFQ; it inherits every weakness of the scores and the weights beneath it; and it quietly double-counts price when a price criterion is already in the matrix. The tool restricts the value-for-money tile to compliant bids, because a bargain you cannot award is not a bargain. When the index and the total disagree about the leader, that disagreement is the discussion your award meeting should have.

Debriefing losing bidders professionally

Every bidder who lost spent real money preparing an offer for you. A short, honest debrief — where they scored well, where they fell behind, without revealing competitors' figures — costs you twenty minutes and keeps them bidding next time. A market that stops quoting you is a market that has priced you as a waste of estimating time, and you will feel it as fewer bids and higher prices within a couple of cycles. The matrix makes the debrief easy: the criteria and weights were fixed in advance, the scores have evidence behind them, and "you were strongest on lead time but the gap on quality systems decided it" is a sentence you can say with the printed report in your hand.

Related tools

Three tools share a family resemblance; they face different directions. The Weighted Decision Matrix (CM8-230) is the general-purpose version for any decision — choosing a machine, a site, a strategy. The Tender & Bid Register (CM8-67) tracks the bids you submit to customers — your pipeline, win rate and estimating effort. This tool evaluates the bids you receive: it adds the things supplier selection specifically needs — quoted prices alongside scores, a compliance gate, criterion groups with the price-weight check, and an award summary written for the file.

FAQ

How many criteria should an evaluation carry? Four to eight. Fewer and one criterion dominates by default; more and the weights stop meaning anything because everything matters a little. If two criteria always score together, merge them.

Can I evaluate several RFQs in one file? Yes — spell the RFQ name identically on every row of the same enquiry. The tiles, charts and matrix table focus on the RFQ with the most rows; the award summary lists every RFQ. For heavy parallel use, one exported file per enquiry is tidier.

Should more than one person score? Ideally, yes — commercial criteria by the buyer, technical by the engineer, each recorded under their own name in the evaluator field. Where scores diverge sharply, the conversation that follows is the most valuable part of the process.

What if two bids finish within a few points? The leading-bidder tile flags a close call when the gap is inside the threshold you set. A matrix cannot split a tie that fine — the scores are not that precise. Go back to the evidence, or to the risk criteria, and decide with your eyes open.

One bidder is far cheaper than everyone else. Good news? Sometimes. It can also mean they misread the specification, plan to recover margin through variations, or are in financial trouble and buying work. An outlier price is a question to ask before award, not after.

Saving your work

Scores, settings and the report header are written to this browser's local storage as you type, and the toolbar shows the time of the last save. That storage belongs to one browser on one computer: another browser, a private window, a second machine or a clean-up tool that clears site data will not have it.

Treat Export .json as the real save — one file containing everything, which Import .json restores anywhere. Export CSV gives you the register for spreadsheet work. Reset asks twice, then erases everything this tool has stored. There is no undo. An evaluation file contains bidders' prices and your judgement of named companies — treat exports as confidential on both counts.

Accuracy & disclaimer

The arithmetic here is deliberately simple — weight times score, summed — and the tool does it faithfully. Everything that matters sits underneath the arithmetic: whether the criteria are the right ones, whether the weights were fixed before the bids were opened, whether the scores reflect evidence rather than preference, and whether the prices were normalised to a comparable basis. The matrix documents the evaluation; it cannot detect scores entered after a favourite was already chosen — though an empty evidence column usually gives that away to anyone who looks.

Procurement rules — especially for public bodies, regulated industries and funded projects — differ by country and by sector, and may prescribe evaluation methods, publication duties and standstill periods this tool knows nothing about. This is a record-keeping and comparison aid, not procurement, legal or financial advice.