The cheapest number and the best bid are different things, and the difference only becomes visible on paper. Every buying decision on your project — scored, compared, and defensible two years later.
Illustrative engagement telemetry — live engagements display from the actual record.
Every tender allocates risk. Which party carries design development, who owns the ground conditions, what happens when information arrives late, whether a provisional sum is a genuine allowance or a placeholder for an argument — these are decided in the tender documents and then lived with for three years. Most owners approach that moment with a document set inherited from the last project and a recommendation from whoever is administering the process.
The bid comparison then compounds it. Submissions arrive scoped differently, qualified differently, and programmed differently, and they are compared on the bottom line because that is the only figure that appears in all of them. The lowest number wins, and the exclusions surface as variations from month four onward. HKA's research finds scope change among the leading causes of claims globally — much of it originating in exactly this gap between what was tendered and what was assumed.
Long-lead items are the quiet version of the same failure. Nobody disputes that the chillers take thirty weeks. The order is simply placed against a programme that assumed twenty, and the delay is discovered as a fact rather than prevented as a decision.
Strategy before documents. Packaging and tender sequence are designed against the delivery programme rather than after it, with the risk allocation in each package made explicit and deliberate — including the question most tender documents leave implicit, which is what happens when design information arrives late.
Bids levelled like-for-like. Every submission is normalised to a common scope: exclusions priced back in, qualifications valued, provisional sums tested, programme assumptions compared. Only then are commercial, technical, and qualification criteria scored against weightings agreed before the bids were opened.
Due diligence before award, not after failure. Financial standing, current workload against your programme, and delivery record on comparable work — verified rather than assumed, because a contractor's capacity is a project risk long before it becomes a commercial one.
| Bidder | Submitted | Exclusions priced | Levelled | Technical score |
|---|---|---|---|---|
| Bidder A | AED 384M | AED 31M | AED 415M | 72 |
| Bidder B | AED 398M | AED 9M | AED 407M | 81 |
| Bidder C | AED 402M | AED 14M | AED 416M | 78 |
| Bidder D | AED 428M | AED 4M | AED 432M | 84 |
Bidder A submitted the lowest price and is not the cheapest bid. Once its exclusions are priced back to a common scope it sits above Bidder B, with a materially weaker technical score. That inversion is ordinary — it appears on most tenders — and it is invisible to any comparison that reads the bottom line.
Scope that has been quietly excluded, provisional sums that hide the real number, qualifications buried in a covering letter, programme assumptions that differ between bidders, and rates that only look competitive until you price the exclusions back in. Levelled like-for-like, the cheapest submission and the cheapest bid are frequently different bidders.
Because a recommendation is an opinion and a score is an argument. Weighted criteria agreed before bids open, applied consistently and recorded, produce a better decision and a defensible file. If the award is ever questioned — by a board, an auditor, or a disappointed bidder — the reasoning already exists in writing.
From the moment the delivery programme exists, which is usually before the package is tendered. Long-lead procurement is the delay that everybody could see coming: the item was always going to take thirty weeks, and the order was placed in week twenty.