Three bids arrive and one is noticeably lower. The temptation is to read that as better value, and occasionally it is. More often it is a different scheme being priced, and the gap closes during the build until the cheap bid has become the expensive one. Avoiding that is the whole purpose of procurement and contractor selection, and it is mostly done before the bids go out.
A tender is a price for a defined scope under defined terms. If the scope is loose, the bids will differ in what they assume, and comparing the totals tells you nothing except who assumed least. The work is in making the bids comparable and then reading what each one has actually priced.
Where the gap usually hides
Four places, in order of frequency. Exclusions, where the bid states that something is not included and the reader skims past it. Provisional sums, where an allowance stands in for work that has not been designed and will be remeasured later at a price nobody has agreed. Qualifications, where the contractor accepts the scope but changes the terms. And quantities, where the bid prices less work than the drawings show.
None of those are dishonest. A contractor pricing an incomplete design has to make assumptions, and stating them is the correct thing to do. The failure is on the other side of the table, where the assumptions are not read, priced and compared.
The consistent pattern is that the lowest bid carries the most assumptions. That is not a coincidence: assumptions are what make a bid lower.
There is a fifth place worth checking, which is the programme. A bid priced on a shorter programme than the scheme can realistically be built in carries less preliminaries cost, and the difference reappears as an extension of time claim with the time-related costs attached. A programme that is weeks shorter than every other bidder’s is a commercial decision by that contractor, not an efficiency.
Provisional sums are the main engine
A provisional sum is money held for work that is not yet defined. When the work is finally designed and instructed, it is valued, and the valuation replaces the sum. If the sum was generous the scheme gains; if it was thin the scheme pays the difference, and the difference is a variation rather than a negotiation.
The distinction between a defined and an undefined provisional sum matters here, because it decides whether the contractor has had to allow for the work in their programme and preliminaries. Where it is undefined, instructing it can carry time and cost beyond the measured work itself. We set that out in defined and undefined provisional sums.
A bid with large provisional sums is not a fixed price. It is a partial price with an open section, and the open section is where the final account grows.
The practical response at tender stage is to count them. Add up the provisional sums in each bid and express them as a proportion of the total, and the comparison changes character: you are no longer comparing three prices but three degrees of certainty. Where the proportion is high in one bid, either the design needs completing before the contract is let or the sums need defining well enough that the contractor has genuinely allowed for the work.
How to make bids comparable
Normalising the bids means putting them onto the same basis before you compare them. You take each bid, add back what it has excluded, bring the provisional sums to a common allowance, price the qualifications that shift risk onto you, and check the quantities against the measured scope. What you then have is three numbers for the same building.
That exercise frequently reverses the ranking, and it always tells you something about the bidder. A contractor whose bid is tightly priced and clearly qualified is easier to work with than one whose bid is low and silent, because the first one has read the documents. The method is described in normalising tender bids.
It also needs doing in writing and in the same format for each bidder, because the comparison is a document you will refer back to for the rest of the project.
Prevent it at the documents, not at the interview
Most of the problem is created before tender. A pricing document that tells each bidder exactly what to price, with quantities measured from the drawings and a clear statement of what is included, removes the space in which assumptions live. Bidders then compete on price for the same work, which is the only competition worth running.
Where the design is genuinely incomplete, say so and structure the tender around it, whether through a two stage approach, a clear schedule of provisional sums or a remeasurement basis. What does not work is issuing an incomplete design as though it were complete and hoping the bids absorb the gap.
Measured quantities are the practical mechanism, and they belong to you rather than to the bidders, which is why quantities and take-off work before tender pays for itself on the first variation.
What else to weigh besides the number
Whether the contractor has built this type and size of scheme, and whether the people who did it are the people proposed for yours. The programme they have offered, and whether it is achievable rather than attractive. Their financial position, because a contractor failing mid-contract is the most expensive outcome available. How they price change, since change is certain.
The contract terms sit alongside all of that, and the form chosen decides who carries design risk and how change is valued. That choice is the subject of which JCT contract suits a small development.
Once the contract is let, the discipline moves to measurement and change control, where the final account is actually decided. That is delivery and cost control, and it runs inside development management.
About to go out to tender?
We prepare the pricing documents, normalise the bids and tell you what each one has actually priced.