Low bid is simple to lose. You either had the lowest compliant price or you did not, and the lesson afterwards is arithmetic.
Weighted evaluation is different, and firms coming from a low-bid habit misread it consistently. They treat the quality sections as a formality standing between them and the price envelope, write them the night before, and then conclude the process was subjective when the scores come back low.
The structure, in the order it happens
Under Quebec's public contracting regime and most comparable frameworks, quality and price are submitted separately and opened separately. The evaluation committee scores the quality envelope without knowing any bidder's price.
Each criterion carries a stated weight and is scored against a described scale. A minimum passing mark per criterion is common — seventy percent is the usual figure — and it is a threshold, not a target. A submission that falls below it on a single criterion is typically eliminated regardless of how strong the rest is and regardless of price.
Only then is price opened, and only for the submissions that passed. The final ranking combines the quality score and the price through a formula stated in the tender documents.
What that formula is really doing
The precise arithmetic varies, but the shape is consistent: the quality score improves the price you are effectively judged on, so a higher-scoring bid can win at a higher number.
Two consequences follow, and they are the whole reason to read the formula before writing anything.
The first is that the value of a quality point is calculable. The tender tells you what the formula is, so you can determine what a five-point improvement in quality is worth in dollars against a realistic competing price. That converts "should we invest in this response" from an instinct into a number.
The second is that above the threshold, quality points and price are genuinely tradeable. Below it, they are not tradeable at all — a criterion under the minimum ends the bid. Effort belongs first at clearing every threshold, and only then at maximizing the total.
What a committee is permitted to reward
This is the part most often misunderstood. Committee members score what is written in the submission against a grid they did not design, and they are generally not permitted to credit what they happen to know about your firm.
That has a blunt implication. Your reputation with that owner, the project you delivered for them two years ago, and the fact that everyone in the room knows you are competent are all worth zero unless they appear in the document as evidence responsive to a criterion.
Firms with strong reputations score worst on this, and for a predictable reason: they write as though the reader already knows them.
Why sections score badly
Almost always because they assert rather than evidence. "We have extensive experience with projects of this nature" is unscoreable — there is no scale on which a committee can place it.
The alternative is specific and boring. Name the project, the owner, the value, the year, the constraint that made it comparable, and the person from that project who is assigned to this one. An evaluator can locate that on a scale. They cannot locate an adjective.
The same applies to methodology. A generic method statement describes how construction works. A scoring one addresses the specific constraint in this tender — the occupied building, the winter schedule, the phasing around an operating facility — and says what the firm will do about it.
Read the grid before writing anything
The grid is published. It lists the criteria, their weights, and usually the elements each one looks for.
Answer it in its own structure and its own language, criterion by criterion. An evaluator scoring fifteen submissions against a grid should not have to search your document for the thing they are scoring, and a response organized around the grid rather than around your own preferred narrative consistently scores higher for the same underlying content.
That mapping — grid criterion to the evidence in your own records that answers it — is work Offra is built to shorten: extracting each criterion and its weight from the tender, checking it against your company profile, past projects and certifications, and showing what is already covered, what is partially covered and what is missing before the writing starts. The judgment about what to claim, and how, stays with the team making the bid.
