How much do the UK's biggest suppliers spend bidding for government work?

What this page estimates

The annual cost of bid pursuit — every bid written, winning and losing — across the 37 largest strategic suppliers to the UK government. Three independent methods, built from entirely different starting points, all land in the range £0.8bn–£1.0bn a year. No single number is precise; the convergence is the finding.

This page sets out each method, the figures behind it, and — for every figure — whether it is measured, taken from a company's own published accounts, or estimated. Three status labels are used throughout:

• MEASURED — a hard count from contract-award data, or a figure read directly from an audited report.
• PRIMARY-SOURCED — taken from a company's own published accounts or results, with some scope judgement applied.
• MY ESTIMATE — a judgement made because no source discloses the figure. These are the soft levers of the model, and they are all named below.

Reference date for all figures: 9 July 2026. Contract-award counts cover the 12 months to that date.

What the estimate covers — and what it does not

• It IS total pursuit spend — every bid these suppliers write, winning AND losing. To win one contract at a typical 20% win rate a supplier writes roughly five bids; this number counts the cost of all five. The losing bids are not excluded — they are the majority of the number.
• It is NOT the cost of only the bids that won. That is a different, much smaller figure — roughly a third of this one.

The one-sentence version: "The 37 largest strategic suppliers to the UK government spend roughly £0.8bn–£1bn a year on total pursuit activity for public-sector contracts — including the bids they lose — of which around 70% is sunk on unsuccessful bids." The "including the bids they lose" clause matters: without it, the figure reads as if it covered only winning bids, which would be roughly three times too small.

Three independent methods

The estimate was built three separate ways, deliberately from different raw material: government contract-award data, bid-team economics, and company financial accounts. Each method stands alone; their agreement is what makes the range quotable.

Method 1 — the awards-led build (£1.01bn)

This is the primary method. It starts from contracts that can actually be counted, and works backwards to the bids written to win them.

1. Count the wins. In the 12 months to 9 July 2026, the 37 suppliers won 722 de-duplicated contract awards of £25,000 or more. MEASURED
2. Split by route: 672 open-market awards and 50 via framework agreements.
3. Open-market bidding cost: for each win, multiply the cost of writing one bid (scaled to contract size) by the number of bids written per win (1 ÷ win rate). Result: 3,262 bids written → £604m. Built on counted contracts with estimated win rates.
4. Framework bidding cost: award data sees framework activity poorly, so this part is inferred. If frameworks carry 80% of activity, framework wins run at roughly four times the visible open-market wins → ~2,688 wins → ~8,960 bids at £45,000 each → £403m. MY ESTIMATE
5. Total: £604m + £403m = £1.01bn. Roughly 60% of it rests on measured contract counts; 40% is inferred.

Two measured facts support the win-rate assumptions: contests in the award data attract a mean of 4.0 real bidders (median 2), consistent with win rates in the 20–33% band. MEASURED

The cost-to-bid scale is anchored on three real bids where the cost is known: a £12m bid for a £1.2bn programme (1%), a £25m deal bid at roughly 2.5% of value, and a framework mini-competition at £45,000.

Method 2 — bid-team economics (£1.07bn)

A completely different construction. Instead of counting wins, it models the market by contract size tier × bid-team size × number of bids per year — what the bid teams across the market actually cost to run.

The tier inputs are themselves estimates (MY ESTIMATE), which is exactly why this method matters: two entirely different constructions — one counting wins, one modelling team costs — land within 6% of each other (£1.01bn vs £1.07bn). That agreement is far stronger evidence than either number alone.

Method 3 — the accounts-led build (£0.77bn)

The third method builds the market up company by company, from real accounts. For 21 of the 37 suppliers, UK revenue and UK operating margin were taken directly from audited annual reports and Companies House filings, each figure page-cited in the underlying research. PRIMARY-SOURCED

The scaling principle: bid cost tracks the gross profit a firm defends, not its revenue. A high-margin consultancy defends far more profit per pound of revenue than a low-margin outsourcer, so it bids harder. The ratio — roughly 36p of bid cost per £1 of margin defended — comes from the single-supplier cross-check below.

Because published UK revenue covers a firm's whole UK book — banking, retail and commercial clients as well as government — a public-sector share is applied to each company before calculating. This is the method's biggest judgement call:

Of the £0.77bn, £0.34bn rests on primary-sourced accounts; £0.43bn is still estimated. This method sits at the lower end of the range because it is the only one that strips private-sector work out of company revenue, and its estimates for the uncovered firms are deliberately conservative.

A single-supplier cross-check

To sanity-check the market total, one of the largest of the 37 suppliers was built bottom-up from its own published full-year results: UK-division revenue of £2.6bn, annual UK order intake of £3.7bn, a disclosed win rate of 60% and contract retention above 90%. PRIMARY-SOURCED

Two builds bracket its bid spend: a cost-of-sale ratio of 1–2% on UK revenue gives £26–65m; working back from order intake (won £3.7bn at a 60% win rate means bidding for ~£6.2bn, at 1–2%) gives £62–123m. One further adjustment narrows it: not all bids cost the same. A renewal or extension costs about a third of a new competitive bid and is almost always won; new competitive work costs full price at 3–4 bids per win. A retention rate above 90% marks this firm as strongly renewal-heavy.

Central estimate for this supplier: ~£45m a year (range £40–55m). The revenue and win-rate inputs are primary-sourced; the renewal share is inferred from the retention rate, not disclosed (MY ESTIMATE).

Does it fit the market total? £1.01bn ÷ 37 suppliers = ~£27m average. This supplier at ~£45m is ~1.7× the average — right for one of the largest firms in the cohort. Consistent.

Where the three methods land

All three sit around £0.8bn–£1.0bn. They were built from entirely different starting points — government contract data, bid-team economics, and company financial accounts. Their agreement is the real result. Quote it as a range, and keep the "including the bids they lose" clause: without it the figure reads roughly three times too small.

Sensitivity and range

Moving the two softest inputs in Method 1 — the framework share of activity and the assumed win rates — to their reasonable extremes swings that method's answer between £0.7bn and £1.6bn. The £1.01bn is a central estimate, not a precise figure. The published headline is therefore the cross-method range, £0.8bn–£1.0bn a year, not any single method's point value.

The assumptions register

Every judgement call in the whole build, graded:

Limits, and what would tighten the estimate

1. Public contract data records winners, never losers. Every bid-spend figure therefore rests on an assumed win rate — no supplier publishes its bid costs or its full bid count.
2. The public-sector share of each firm's UK revenue is the single biggest soft lever. Almost no firm discloses what proportion of its UK revenue is government work; the shares used here are set from market knowledge.
3. Around 14 of the 37 firms are not yet covered by primary accounts. Their contribution to Method 3 is estimated, conservatively.
4. Framework activity — the majority of central-government buying — is inferred, not counted. Award data sees it poorly.

The highest-value refinements are extending the primary-source coverage to the remaining firms, and locating the minority of firms that do disclose a public-sector revenue split in their segment reporting.

Sources

The figures on this page draw on three classes of source: UK government contract-award data (the measured contract counts and bidder numbers); audited annual reports and Companies House filings for 21 of the 37 suppliers, each figure page-cited in the underlying research; and an independent procurement data provider's published estimate of the framework share of central-government buying. Where none of these covers an input, it is labelled MY ESTIMATE above rather than presented as fact.

Status: methodology estimate as at 9 July 2026. The three-method convergence on £0.8bn–£1.0bn is stable; the precise point within that range depends on the named assumptions above, all of which are graded on this page.