FAQ and glossary

Tender Navigator glossary

The vocabulary of Tender Navigator, one term per entry, in the product's own wording.

Last updated

Tender

A public procurement opportunity: a contracting authority's published invitation to bid for goods, services or works. In Tender Navigator every tender is represented by a notice ingested from a procurement platform.

Notice

The published record of a tender on a procurement platform: title, description, buyer country, dates, classification codes and other fields. Tender Navigator holds a copy of the notice; the original on the platform remains the source of truth. Also called a tender notice.

Buyer country

The country of the contracting authority that published a notice. Searches and filters select tenders by buyer country.

Source system

The procurement platform a notice was ingested from, shown as Source on a tender. Examples: TED, Find a Tender, SAM.gov, GOV.UK.

External ID

The identifier of a notice on its source system. Together with the source, it locates the original notice.

CPV

The Common Procurement Vocabulary, the eight-digit classification used by European Union and United Kingdom procurement portals. One of the three classification systems Tender Navigator supports.

NAICS

The North American Industry Classification System, the six-digit classification used by United States and Canadian procurement portals.

UNSPSC

The United Nations Standard Products and Services Code, the eight-digit classification used by international bodies and portals outside Europe and North America.

Classification code

A CPV, NAICS or UNSPSC code attached to a notice to say what is being bought. In Tender Navigator, codes are entered from a built-in dictionary and combined with OR.

Pre-search tool

Tender Navigator's live search across the tender database, available on the website with a five-result preview and in full inside the dashboard. Also written Presearch or Pre-Search. Its results are not evaluated by AI.

Not evaluated

The badge on every Pre-search result, meaning the tender has not been judged against any filter.

Filter

A saved tender search that keeps running: keywords, countries, classification codes, an optional budget rule, an AI prompt, an AI model and an email schedule. Only an active filter searches, evaluates and sends digests.

Active and paused

The two states of a filter. Active filters run daily; paused filters are stored and do nothing. New filters start paused.

AI prompt

The free-text description on a filter of what makes a tender a good or bad fit for your company. Every tender the filter picks up is evaluated against it. Also called business criteria or business context.

LLM model

The large language model a filter uses for evaluation, chosen per filter from the list in the filter editor.

AI evaluation

The daily process in which each tender picked up by an active filter is judged by the filter's AI model against the filter's prompt, producing a verdict and a reasoning. Also called AI matching or tender evaluation.

Evaluation

One AI evaluation of one tender by one filter. Plans include a number of evaluations per billing cycle.

Matched

The verdict that the AI judged a tender a fit for the filter's prompt.

Unmatched

The verdict that the AI judged a tender not a fit for the filter's prompt. Unmatched results are kept so the prompt can be corrected.

Reasoning

The AI's written explanation of a verdict, shown under "Evaluation completed" in the Tender Info panel, in digests and on shared tender pages.

Result

An evaluated tender: the notice plus its verdict, reasoning, evaluation date and a snapshot of the filter it was evaluated by. Results live on the Results page.

Filter snapshot

The copy of a filter's criteria, prompt settings and model stored with each result at evaluation time, shown as Filter info in the Tender Info panel. Editing the filter later does not change it.

Stuck tender

A tender a filter picked up but could not evaluate because the plan's evaluations for the cycle, or the filter's own evaluation limit, were used up.

Evaluations limit

An optional per-filter cap on the number of tenders the filter may evaluate in a month. Called hard limit in some older material.

AI summary

A short AI-generated overview of a notice, generated on the first open of the notice and shown in the Pre-search side panel and on the notice page. Also called AI overview.

Tender Info

The panel that opens for an evaluated tender on the Results page or the Overview, showing the verdict, reasoning, details, original link and filter snapshot.

Tender Detail

The page or side panel showing a notice opened from the Pre-search tool.

List

A named collection of saved evaluated tenders under My Lists. Favorites is the default list. Also called Tender List.

Pipeline status

The per-list status of a saved tender: Unprocessed, Reviewing, Qualified, Bidding started, Proposal submitted, Won, Lost or Archive.

Digest

The email an active filter sends on its schedule, daily or weekly, covering the tenders it recently evaluated.

Business Profile

The private, structured description of your company on your account, used to generate draft filters. Never public.

Supplier Profile

The public company listing you can publish in the Supplier Directory: identity, logo, descriptions, tags and public contact details.

Supplier Directory

The public page listing published Supplier Profiles.

Plan

The subscription tier of an account: Basic (free), Standard, Premium or Corporate. A plan sets the active-filter allowance, the evaluations per cycle and the extras.

Billing cycle

The monthly period of a subscription, with a period start and end shown on the Subscription page. Evaluations are counted per cycle.

Partner

An organisation that offers Tender Navigator's search on its own website through the embedded widget and registers visitors through a co-branded page. Accounts created that way are affiliated with the partner.

Platform tour

The four short guided tours inside the dashboard, started from the Product tour button or a tour bar on the page.

Getting-started checklist

The five-row card on the dashboard Overview that tracks account creation, plan activation, the first filter, switching it on, and the first evaluated result.

AI assistant

The chat assistant in the bottom-right corner of the website and the dashboard, offered at three levels: Pre-Sales Agent, Q&A Agent and Full Assistant.