The 2025 proposal-tool hype cycle, mapped

The Gartner hype cycle has five phases: trigger, peak of inflated expectations, trough of disillusionment, slope of enlightenment, plateau of productivity. Vendors rarely travel it on the published schedule. In 2025, the proposal-tools category produced one clear peak, two clear crashes, a plateau, and a genuine slope-of-enlightenment moment. This post maps each public actor against the five phases using only publicly observable signals — review trends, pricing posture, AI claims, and product-launch cadence.

Everything here is drawn from public sources. We are a category participant; the obvious bias is disclosed, and the method is the same we used in Wave 1 of the State of Proposal Tools report.

Methodology in two paragraphs

For each vendor we tracked in Wave 1, we collected quarterly review counts from G2 and Capterra, public pricing signals (where posted), public product announcements, and any earnings commentary from parent companies that filed publicly. We then placed the vendor at a point on the hype-cycle curve for each of the four quarters of 2025.

A peak is a quarter where the vendor’s public discourse dominated category attention — launch announcements, conference keynotes, broad press pickup — without a commensurate improvement in review sentiment. A crash is a quarter where review sentiment moved meaningfully negative, pricing posture tightened, or a public departure (product lead, CEO, major customer) became visible. A plateau is a quarter where nothing changed measurably. A slope is a quarter where review sentiment moved positive in a way that tracked a specific product-level change, not a marketing push.

The map

Responsive — plateau, with sentiment drift. 2025 was a quiet year for Responsive at the marketing layer. Behind that was a continuing accumulation of negative G2 reviews on search quality and UX — the “search is terrible” meme, the “clunky, impossible to locate” complaints, the pattern of a legacy CMS whose AI features pale beside ad-hoc generative tools. The vendor did not crash in 2025. It plateaued on the curve while sentiment quietly drifted, which is the pattern that precedes the kind of quarter that looks like a crash in retrospect.

Loopio — slope of enlightenment, conditional. Loopio entered 2025 carrying the reputation captured in autorfp.ai’s review synthesis — the “overpriced document repository” framing, the “magic doesn’t work well” meme. Over the year, Loopio shifted communications meaningfully toward content-freshness as a feature. Whether the underlying product has moved in step is a Wave 2 question. The posture changed in 2025; the evidence that the posture is backed by product change will settle in the first half of 2026.

AutogenAI — peak, descending. AutogenAI had a peak-of-inflated-expectations 2025. Their own hallucination-mitigation blog was the most-cited vendor piece of 2025–2025, and their conference presence through mid-2025 was the category’s loudest. By late 2025 the signal had shifted: the AI-drafting category that AutogenAI pioneered had commoditized enough that differentiating on “we have an AI” no longer landed. The vendor’s 2026 posture — we expect — has to be grounded-retrieval-centric or it loses the narrative high ground it earned in 2024.

Qorus (QorusDocs) — trough, visible. Review aggregation on Capterra showed consistent complaints about speed, dashboard limitations, and search relevance through 2025. The vendor has a real customer base and a real product footprint, but the public-review surface suggested that the category’s expectations moved faster than the product. Whether Qorus climbs toward a slope-of-enlightenment in 2026 depends on whether the performance complaints see concrete product-level responses in the next two quarters.

Upland Qvidian — plateau on the long tail. Qvidian, Upland’s legacy RFP-response product, sits on a plateau that does not much matter to the category’s forward motion. G2 reviewers continue to describe a UI that “could be more modern,” AI performance that is inadequate by 2025 standards, and a product that is expensive relative to what it delivers. Upland’s portfolio strategy keeps the product alive without investing into the kind of rebuild that would reposition it on the curve. Plateau forever is a real position. Some buyers want exactly that.

The generalist AI-assistant wave (1up and adjacent) — trigger, still rising. 1up’s own critique of the category captured the 2025 positioning of the generalist-first class of tools: RFP software has been “mostly just knowledge management,” which the current AI wave can refactor into something closer to retrieval + drafting without a monolithic vendor in the middle. Whether the generalist-first wave reaches a peak in 2026 or flames out against the trust bar that regulated buyers apply is an open question. In 2025 they were firmly in the trigger phase.

What nobody reached

No vendor in the public category is, by the end of 2025, on a clear plateau-of-productivity — the phase where a tool has crossed from “emerging” into “boring utility the market depends on.” The tools in that phase in adjacent categories (a CI/CD product, a logging tool, a CRM) have fifteen-year track records, boring-but-reliable reviews, and a customer base that renews without examining the decision. No proposal-tool vendor is yet there. The category is young; the AI cycle is resetting posture faster than vendors can deliver proof. Wave 2 of the State of Proposal Tools will return to this question.

What our own posture is

PursuitAgent launched publicly inside 2025. By the definition above, we are in the trigger phase — on the curve, but not yet tested by a full cycle of renewals and reference checks. The honest placement is: we have a thesis (grounded retrieval, KB-backed drafting, closed-loop post-mortem writing), and the market has not yet had enough time to produce the reviews that tell us whether the thesis holds at scale. Next year’s map will include the answer. This year’s does not.

Two predictions for 2026

First: at least one of the incumbent vendors whose review sentiment drifted negative in 2025 will see a CEO change, a pricing reset, or a major customer departure made public. The signal in the review base is too consistent to hold without a correction. We will not name which vendor; the evidence points in more than one direction.

Second: the generalist AI-assistant wave will produce exactly one breakout in 2026 — a tool that wins a regulated-market customer on a grounded-retrieval posture and stays — and three or four that do not survive the trust bar. The generalist wave will not displace the incumbents at the enterprise layer. It will erode the bottom of the market, where the incumbents’ pricing has always been the softest.

Both predictions are opinion. They are sourced to the review-signal patterns above, not to inside information. The January Wave 2 report will grade how they held up.