RFP tools: four different jobs, one word
RFP tools is a category label covering four separable jobs — finding the opportunity, writing the response, knowing the buyer, and signing the document. Very few products are good at more than one of them.
This page is the map, not a ranking. It sorts the category into its four jobs, shows which vendors claim which, and prints what each one publishes about its own coverage and price. For a ranked recommendation see best RFP software; for a head-to-head see the comparison.
RFP tools: the short answer
Buying in this category goes wrong in one specific way: you buy for the job that is loudest in the marketing rather than the job that is actually failing in your week.
The four jobs are genuinely separable, and they fail separately. A team that never hears about the right contract has a discovery problem, and the best writing tool in the world will not fix it. A team drowning in questionnaires has an authoring problem. A team bidding hard and losing to the incumbent every time has an intelligence problem — it is aiming at contests it was never going to win. A team that has none of those has a document-handling problem, which is the smallest and cheapest of the four.
Almost every product in this category is strong at one job, adequate at a second, and marketing copy at the other two. That is not a criticism of the vendors — the jobs need different data, different pipelines and different skills. It is a warning about the word "RFP tools", which flattens all four into a single shopping decision.
So the useful first question is not "which tool is best". It is "which of the four is broken". The decision panel below asks three questions and names the category; it recommends no product.
The four jobs RFP tools actually do
Each of these is a different product with a different data problem behind it. Reading them as one category is what produces the wrong purchase.
Finding the opportunity
Watching the publication routes and surfacing notices that match what you sell, early enough to bid properly.
Writing the response
Extracting requirements from the pack, drafting answers from a content library of what you have written before, and keeping one voice across a long document.
Knowing the buyer
Who holds the contract now, what they were paid, when it ends, and how often this buyer actually changes supplier.
Getting it out of the door
Assembly, version control, approvals and e-signature — the mechanics of turning a finished answer into a submitted document.
Two of these four are about information the buyer publishes, and two are about work your team does. That split matters more than it looks: the first two are limited by what UK procurement law requires to be published, and the second two are limited only by the software. Which is why the coverage question below has a ceiling that no vendor can engineer past.
Discovery tools, and the only question that matters
Every discovery product makes a coverage claim. Almost none of them make a checkable one.
Under the Procurement Act 2023 there is a statutory central digital platform. s.95(4) requires that "a Minister of the Crown must make arrangements to establish and operate an online system for the purpose of publishing notices, documents and other information under this Act" [1] — that system is Find a Tender. s.95(5)(a) requires that it "make notices, documents and other information published under this Act available free of charge" [1], and the Cabinet Office restates it for suppliers in plain words: "Find a Tender is free to use for everyone. There are no charges and you should not pay to view opportunities in the public sector." [2]
That single sentence should shape how you buy discovery software. You are never paying for access to the notices. The notices are free and openly licensed — Find a Tender publishes them under the Open Government Licence and offers them in Open Contracting Data Standard JSON through its own API. [4] What you are paying for is filtering, matching, alerting and the devolved routes: Scotland, Wales and Northern Ireland run their own portals under their own arrangements.
Contracts Finder is the other name you will see, and its status has changed. PPN 019 is explicit that its Contracts Finder guidance "only applies to procurements commenced before 24 February 2025 under the PCRs", and that "there should be no circumstance where this would be required as no new procurements will be conducted under the PCRs after 24 February 2025". [3] A discovery tool whose source list still leads with Contracts Finder is describing where notices used to go.
What does it actually cover?
Pick a vendor and read what it says about its own sources. The distinction being drawn is not good against bad — it is named against unfalsifiable.
The pattern is the finding. One vendor in the set names its sources; the rest publish a number or an adjective. And naming them is not a weakness — it is the only thing that makes a coverage claim mean anything, because a list can be checked and "every public sector portal" cannot.
There is also a hard ceiling, and it applies to us as much as to anyone selling against us. s.94 of the Act permits an authority to withhold information where publication "is necessary for the purpose of safeguarding national security", or where the information is "sensitive commercial information and there is an overriding public interest in its being withheld". [10] No tool can see what was never published. Total coverage is not a claim anyone in this category is in a position to make, and a vendor who makes it is telling you something about their marketing rather than their pipeline.
Market intelligence: knowing before you bid
The smallest of the four categories, and the one that changes the most decisions per pound.
Intelligence products are built from award data rather than notice data, which is a harder problem: awards are published later, less consistently, and with more variation in how the winner is named. Tussell describes itself as "the all-in-one UK public sector market intelligence platform" and claims "over 80,000 key decision-makers" [18]; BidStats sells an Insights tier separately from its discovery product, "From £5k per annum". [17]
The National Audit Office's work on this is the best public evidence of why the category exists at all. Its 2024 report found that "government does not know how many frameworks exist, but third-party data service providers have identified more than 21,000 from published data" — and the three providers it names disagree substantially, with Tussell identifying 8,990, BiP Solutions 11,194 and Delta Market Analytics 21,561. [8]
The providers disagree, and the gap is the product
Three commercial datasets built from the same published record produce counts that differ by more than a factor of two. That is not one of them being wrong — it is what "a framework" counts as being genuinely unsettled. Ask any intelligence vendor how it counts, not just what it counts.
Fragmentation raises bidding costs
The NAO is direct: this fragmentation "prevents the government from acting as a single buyer across numerous categories of common goods and services, resulting in duplication of effort and increasing bidding costs for suppliers". [8] The cost of not knowing the landscape is paid by the bidder.
One more figure worth having before you sign anything: framework levies charged to suppliers run from 0% to 11% of revenue, with the Crown Commercial Service's typical range 0–2% and an average of 0.7%. [8] A framework place is not free money, and an intelligence tool that shows you the levy alongside the opportunity is doing something useful.
Document and signature tools
The mature end of the category, where general-purpose products compete directly with bid specialists — and win on price.
PandaDoc and Proposify are document products: assembly, templates, tracking and e-signature. Neither claims UK public-sector notice discovery and neither should be read as an RFP response tool in the authoring sense above. What they are is dramatically cheaper: PandaDoc publishes a free tier, Starter at $19/seat/month and Business at $49/seat/month [16]; Proposify publishes Basic at $19/user/month annually and Team at $41/user/month annually, with Business "Starts at $3,900/year". [15]
If your problem is that finished answers take three days to become a submitted document, this is the category to shop in, and you should not be paying authoring-tool prices to solve it.
There is a UK-specific reason to care about the mechanics, too. The Act sets a mandatory standstill period of "eight working days beginning with the day on which a contract award notice is published" [5], and the old standstill letter has been replaced: "the function of the standstill letter in providing information about assessed tenders is now provided by the assessment summary". [6] Both of those are deadlines that land on your document process, not on your writing process.
RFP tools compared, criterion by criterion
Twelve products, four criteria, and every cell either a quote from the vendor's own site or the word “unstated”. Nothing here is inferred.
| Tool | Primary job | Public pricing | UK notice discovery claimed? |
|---|---|---|---|
| Responsive | Authoring | Tiers listed, no prices [11] | No claim |
| Loopio | Authoring | Tiers listed, no prices [12] | No claim |
| AutoRFP.ai | Authoring | Yes — $899–$1,299/mo yearly [13] | No claim |
| Qvidian | Authoring | No pricing page found [14] | No claim |
| Proposify | Signature | Yes — from $19/user/mo [15] | No |
| PandaDoc | Signature | Yes — free, $19, $49/seat/mo [16] | No |
| Stotles | Discovery + authoring | Inconsistent on its own site [19] | Yes — “every public sector portal” [19] |
| BidStats | Discovery + intelligence | Yes — £0 / £350 / £800 / from £5k [17] | Yes — and it names three sources [17] |
| Tussell | Intelligence | No — “Talk to our Sales team” [18] | No portal-aggregation claim |
| Mercell | Discovery + bid delivery | No public pricing [20] | Yes — Europe-wide [20] |
| Tracker Intelligence | Discovery | Pricing page did not load [21] | Yes — names FTS, OJEU, PCS, eTendersNI [21] |
| Opportuni | Discovery | Yes — £150/mo, £1,500/yr [22] | Claims “over 3,000 portals” [22] |
Read the last column against the third. The two vendors that are most specific about their sources are also among the most specific about their prices, and the vendors that will not publish a price are generally the ones whose coverage claim cannot be checked either. That is a correlation in a set of twelve, not a law — but it is a reasonable thing to notice before a sales call.
Who will tell you the price?
Half of this category does not publish one. That is checkable in a click per vendor, and no comparison page currently says it.
Stotles is listed as inconsistent rather than unpublished, and the distinction is deliberate: its own site renders its entry plan as "from £75/month", "from £99/month" and "£50/user/month" on different pages. [19] There is a published price; there just is not one published price.
None of this makes an unpriced vendor a bad product. Enterprise software is often genuinely quote-shaped, and a tool sized to your volume may cost less than a published sticker. It does mean the comparison you can do from your desk stops at the paywall, and it is worth knowing which half of the category you are about to enter.
Which RFP tool do you actually need?
Three questions. The first “no” decides. No product is recommended and nothing is scored.
The reason to answer in that order is that the jobs sit in a dependency chain. Discovery failures make every downstream tool irrelevant — there is nothing to write. Authoring failures waste the opportunities discovery found. Intelligence failures waste well-written bids on contests that were never open. Fixing them out of order is the most common expensive mistake in this category.
RFP platforms versus RFP tools: does the word mean anything?
Mostly it means the vendor covers more than one of the four jobs. Occasionally it means nothing at all.
There is no definition of "platform" in this market and no standards body policing it, so the honest test is structural rather than semantic: count the jobs. A product that does discovery and authoring is doing two of the four, and "platform" is a fair description. A product that does authoring with a reporting module attached is a tool with a reporting module.
Of the twelve above, two make a genuine two-job claim on their own sites — Stotles with Sales Studio for discovery and Bid Studio, which it describes as "AI tools to streamline tender management… draft stronger bids" [19], and Mercell with "Tender Discovery" alongside "Bid Delivery". [20] BidStats spans discovery and intelligence, which is a different pair.
The useful caution is that breadth is bought with depth somewhere. A two-job platform is competing against single-job specialists on both fronts, and the question worth asking in the demo is which of the two the company actually invests in. The answer is usually visible in which one has the better documentation.
Free RFP software, and what free actually costs
The notices are free by statute. Some of the software has a free tier. Those are two very different kinds of free.
Start with the part that is genuinely free and will stay that way: Find a Tender is required by s.95(5)(a) to make published notices available free of charge [1], the Cabinet Office tells suppliers "you should not pay to view opportunities in the public sector" [2], and the data is published under the Open Government Licence with an OCDS API. [4] If a vendor's pitch is access, the pitch is to sell you something the state already gives away.
On the software side, PandaDoc publishes a free tier [16] and BidStats publishes a £0 tier alongside its paid ones. [17] What free normally buys you in this category is the search and the alert, with the volume, the seats, the history and the export behind the paywall — which is a reasonable trade, and worth testing against your actual weekly volume before upgrading.
The real cost of free is usually time rather than money: a free alert that fires on a keyword rather than on your capability produces a daily list you stop reading within a fortnight. Precision is the thing worth paying for, and it is the thing hardest to evaluate in a trial.
For a fuller treatment of the free end of the market, see free RFP software.
Best RFP software for large teams and high volume
Above a certain volume the buying question changes shape entirely: it stops being about features and becomes about governance.
A team running a handful of bids a year should buy for the job that is failing. A team running security questionnaires weekly alongside major tenders is buying something else — a system of record. At that volume the things that decide the purchase are content governance (who owns an answer, when it expires, who approves the update), permissions, audit trail, and integration with wherever your evidence already lives.
This is where per-project pricing and per-seat pricing genuinely diverge. AutoRFP.ai's Scale plan covers 24 projects a year with unlimited users [13] — a shape that suits a small team doing large bids and penalises a large team doing many small ones. Loopio's Foundations tier bundles 10 seats for "smaller, sales-led RFP teams" [12], and Qvidian's "70+ available reports" [14] is aimed squarely at proposal functions large enough to need reporting on themselves.
The honest caveat: none of the four unpriced enterprise vendors publishes what its enterprise tier costs, so a like-for-like comparison at the top of this market cannot be assembled from public information. The price ladder shows exactly where the public record stops.
For the ranked view at this end of the market, see best RFP software.
Who this page is not for
The most useful half of any category map is the part that tells you that you are in the wrong place.
If you searched for an event RFP tool
You want hotel and venue sourcing — a different market with different vendors, where the RFP goes out to properties for room blocks and meeting space. Nothing on this page applies to it. The overlap is the acronym and nothing else.
If you only bid to private companies
Everything above about publication routes, standstill and thresholds is UK public procurement law and does not touch you. The authoring and signature halves still apply in full; the discovery and intelligence halves largely do not, because private RFPs have no register.
If you are the buyer, not the bidder
You need e-sourcing and evaluation software, which is a separate category bought by authorities rather than suppliers. This page is written for the supplier side throughout.
If you are bidding for everything
If the honest diagnosis is that you bid for too much and win too little, no tool in any of the four categories fixes it. That is a qualification discipline problem, and buying software first usually just produces more losing bids faster.
The named alternatives
We sell a tool in this category. Here are the others by name, described as they describe themselves.
If your problem is discovery: BidStats, Stotles, Tracker Intelligence, Mercell and Opportuni all sell notice discovery to UK suppliers. BidStats is the one that names its sources [17], which makes it the easiest to evaluate honestly; Tracker Intelligence separately names FTS, OJEU, PCS and eTendersNI [21], which is the widest checkable UK list in the set.
If your problem is authoring: Responsive and Loopio are the established response-management platforms, Qvidian is the reporting-heavy incumbent in large proposal functions, and AutoRFP.ai is the one publishing its prices. [13]
If your problem is intelligence: Tussell is the UK specialist [18], and BidStats sells an Insights tier. [17]
If your problem is documents: PandaDoc and Proposify are cheaper than anything in the other three categories and are not trying to be bid software. [15] [16]
And where we fit, stated plainly so you can discount it: RFP Quest is a discovery and intelligence product built on Find a Tender and Contracts Finder data under the Open Government Licence. We do not do e-signature, and we are not a replacement for a mature content library if authoring is your bottleneck. If the panel above routed you to authoring, the sensible shortlist does not start with us.
Questions people actually ask
These are real search strings, kept in the words they were searched in.
Content organisation is the authoring job, and the established products for it are Responsive, Loopio and Qvidian, with AutoRFP.ai the newer entrant that publishes its pricing. All four sell a content library: your previous answers, stored, versioned and searchable, so a new questionnaire is answered from what you have already written rather than from scratch. “GovCon” is a US framing — for UK public-sector bidding the same tools apply, but none of them find the opportunity for you, so if the opportunity is the problem you are shopping in the wrong half of the category.
There is no neutral published benchmark of AI answer quality in this category, so anyone ranking it is ranking an opinion. What you can test in a trial, in an afternoon, is provenance: give the tool five questions you have answered before and check whether each generated answer names the past response it drew on and lets you open it. A tool that produces fluent text with no traceable source is producing something you then have to verify by hand, which is the work you were trying to remove. Content library management is more comparable — look for answer expiry dates, named owners and an approval workflow, because a library without those degrades into a search over stale text within about a year.
We do not publish a reliability ranking, because we sell in this category and any ranking we published would be marketing. For peer reviews the usual sources are G2 and Capterra, where Responsive, Loopio and Proposify all carry substantial review volume from exactly that segment. What is worth doing alongside the reviews is checking the two things reviews rarely cover: whether the vendor publishes a price at all, and whether it will name its data sources. The price ladder and the coverage panel on this page both answer those from the vendors' own sites.
At high volume the deciding features stop being drafting quality and become governance: answer ownership, expiry, approval workflow, permissions and audit trail. Security questionnaires in particular are repetitive and compliance-sensitive, which is the strongest possible case for a content library with a single approved answer per control rather than per-bid rewriting. Pricing shape matters here too — a per-project plan such as AutoRFP.ai's 24-or-50-projects-a-year tiers is built for fewer, larger bids, and a high-volume questionnaire team is usually better served by seat-based pricing.
See the section above. The short version is that above a certain size you are buying a system of record rather than a writing aid, and the deciding criteria are governance and integration. The practical obstacle is that four of the twelve vendors surveyed here publish no enterprise price at all, so a like-for-like comparison at that end of the market cannot be built from public information — you will be running procurements to compare procurement software.
This is the authoring category's hardest genuine problem, and it is where the products differ most. The capability to ask about is import: whether the tool can take a buyer's own spreadsheet or PDF, find the questions inside it, let you answer them in the tool, and then write the answers back into the buyer's original file in the buyer's original format. Plenty of tools do the first half well and hand you a document you then have to re-key. Test it with a real pack from a real buyer during the trial, not with the vendor's sample.
There is no single leader, because there is no single job. Responsive and Loopio are the most established in response authoring; Tussell is the UK specialist in market intelligence; BidStats and Tracker Intelligence are the most checkable in discovery because they name sources; PandaDoc and Proposify lead on price at the document end. Anyone naming one overall leader is either selling it or has not noticed that the category contains four different products. The panel above will tell you which of the four you are actually shopping for.
Different category. An event RFP tool is hotel and venue sourcing — you send an RFP to properties for room blocks, meeting space and rates, and the vendors are event-sourcing platforms rather than anything on this page. None of the twelve products discussed here does it, and no part of UK public procurement law discussed here applies to it. If that is what you were looking for, this page will not help and it is better to say so than to sell you something adjacent.
The glossary
The words this category uses, defined against the statute where the statute defines them.
The sourced record
Every numbered mark above resolves here. Legal claims cite the legislation or the guidance; vendor claims cite that vendor's own page, with the date it was read.
What the public record says about scale
Two official figures are worth having, and they measure different things — which is exactly why they are given separately rather than blended. The House of Commons Library reports gross public sector procurement spending of £461 billion in 2025/26 across the UK, sourced from PESA, and separately that "the latest data is currently for 2024/25 when £395 billion was spent on procurement", sourced from Whole of Government Accounts. It also notes that procurement accounts for about a third of public sector spending — 35% in 2024/25. [9]
The National Audit Office reported that "the public sector spent some £393 billion on goods and services in 2022-23". [8]
A caution about one of these numbers that is easy to get wrong. The Commons Library briefing's PDF at researchbriefings.files.parliament.uk serves the October 2021 edition, with 2020/21 figures. The current numbers are on the HTML landing page. Anyone citing the PDF is publishing five-year-old data with a current-looking citation.
And one figure that does not exist. There is no official UK statistic for the number of procurement notices published per year. The only official cumulative figure is the Cabinet Office's 2022 statement that more than 347,000 notices had been published on Contracts Finder since 2015 [23] — which predates the Act and is history rather than a current rate. Any per-year notice count you see quoted in this category is derived, and the derivation should be stated. This page does not quote one.