How Tenderline Calculates Its Numbers
Published 19 September 2026 · revised 13 October 2026 · applies catalogue-wide
Every figure on Tenderline is built from official UK procurement notices. This page explains, in plain terms, what we count, how we keep the numbers honest and what the figures are not. The on any figure says what that number is and how many records it rests on. Who we are and where the data comes from: About Tenderline.
1. Three units, never blended
A public tender has three layers, and each answers a different question. We count them apart and every figure says which one it uses.
- Procedures: the tender itself, with all its notices. Used for activity: how much a buyer, sector or region publishes. A notice that names several sectors counts in each, so sector figures are never added together. Region pages count each notice once, in the region of its buyer’s office (see below). Planning notices count as pipeline, not as competitions.
- Lots: the part suppliers actually bid for. Used for competition: a tender with seven lots is seven contests. A tender without lots is one.
- Regions: two different facts, kept apart. Region pages and supplier region bars use the region of the buying organisation: the office address on its buyer profile, placed by the official ONS postcode lookup, one region per notice. The tender page and the search region filter use the place of delivery exactly as the notice publishes it, which can be several regions or the whole UK. A national body therefore counts in the region of its office on region pages, and its UK-wide notices show under every region in search.
- Awards and contracts: used for money. Values are shown without VAT where the notice gives them that way. A framework or dynamic purchasing system ceiling is a cap, not a price, so it never enters a sum or a typical value.
Each organisation has one profile. Different spellings and registration numbers of the same authority or company are merged first, checked against official registers such as Companies House where one exists. When a central purchasing body runs a tender for other authorities, the process is credited to the body that ran it and the money to the authority that pays.
2. Competition figures
Typical bids per lot is the middle value of the bid counts that award notices publish, one per lot; single-bid lots is the share with exactly one bid. Both use real competitions only: direct awards are counted separately, and framework and dynamic-market places are shown apart, because joining a panel is not the same contest as winning a contract. Notices under the Procurement Act 2023 report the bids assessed at the final stage, older notices the bids received, so every bids figure says how its sample splits between the two.
Competed is the share of notices run as an open or selective competition, among notices that say which route was used. Abandoned counts tenders published as cancelled, unsuccessful or withdrawn, so it is a minimum. Deadline to award is the typical number of days from the submission deadline to the first award, on tenders that publish both dates.
On a tender page two views sit side by side and are never merged: the authority's own record and the wider market for the same kind of work. Where the local picture is too thin, a figure widens to a broader market, and its names the scope it used. An empty tile means the notices published nothing usable.
3. Award against estimate
An award-against-estimate pair is one lot: the estimate the notice published and the single award on that lot, on the same VAT basis. Shared awards, framework places and pairs too far apart to be a genuine estimate and award are left out. Many buyers publish the award at the estimate, so "at the estimate" is a common result. The gap is shown in words, as a percentage below or above the estimate.
The likely winning price on a tender page is our forecast of where the award may land against the buyer's estimate, based on how similar past awards compared with their estimates, shown with a range. It is a benchmark from the published record, not a recommended bid price and not advice.
4. Who wins, and money
Supplier tables rank by the number of separate tenders a supplier won, not by award records, so a supplier with hundreds of call-offs under one framework does not outrank one that won hundreds of separate competitions. An authority's top-three supplier share counts each awarded tender once.
Won locally and SME winners use contract awards whose winner published a registered region or a company size. A registered address is usually a head office, so a national firm with a local branch reads as non-local.
Award values are published values without VAT, on awards to a single supplier, with framework ceilings and repeated republished totals removed. They are not spend, turnover or a budget, and the largest awards are always listed so any total can be checked against its notices.
5. Time windows and sample size
Every figure states its own time window: market figures usually cover the last three years, authority profiles the last two, and the retender radar reads contracts by their published end dates. When a figure's window differs from its neighbours, its says so.
Every figure shows how many records stand behind it, with a plain label — robust, moderate or indicative, so you can see at a glance how far to lean on it. A figure built on too few records is muted or replaced by a simple count, never filled in. Where a narrow category is too thin we widen to the next one and say so, rather than blending. Figures about named bidders exist only where Procurement Act 2023 award notices list every bidder.
6. Possible competitors and match scores
"Possible competitors", "what this replaces" and the match score are our estimates from the published award history: similar past contracts and the suppliers with a track record in this kind of work. The score ranks how closely a record fits; it is not a probability of winning. Official notices do not say who is bidding, and where the evidence is thin the list stays short rather than guessing.
7. What the figures are not
- Award values are published values, not spend, turnover or a budget. A framework or dynamic-market ceiling is a cap, not a contract's worth.
- Forecasts, possible competitors and match scores are statistical estimates from past awards: not advice, not a recommended bid and not a chance of winning.
- No figure rates an organisation's reliability, credit or payment behaviour.
- Contact details on a tender page are those the official notice published, shown on request (see the privacy notice).
- The legal regime shown (Public Contracts Regulations 2015 or Procurement Act 2023) is the one the notice declares.
8. Sources, freshness and corrections
The catalogue is built from official UK public procurement notices reused under the Open Government Licence v3.0: today Find a Tender, with Contracts Finder and further sources being added. New notices arrive through the day, amendments and awards update the same tender, and the figures are rebuilt nightly. Coverage changes as sources are added, so a figure describes the record as it stands.
Not every notice publishes a bid count, a value or a delivery region, so each figure shows its own count rather than assuming full coverage. If a number looks wrong, the notice it came from is one click away, and tell us: we recheck against the source and correct our figure. The official notice stays as its publisher wrote it.