Published on May 20, 2026
Greg Ellis spent years on both sides of the enterprise RFP. The most expensive habit he sees is treating the proposal team as a centrifuge: work in, document out, team stands down. Here is what the teams that win do instead, and what a practitioner can do this quarter to break the pattern.
✹ Proposal teams that operate as an administrative function inside the sales process tend to lose influence on the deal once the document is submitted, even when the document itself is strong.
✹ The most consistent win-rate gap between proposal teams sits in the final four hours of editorial polish and commercial review before submission. Time stolen from those four hours by reinstated no-bid bids is the single most reliable predictor of a missed deal.
✹ Lost bids are 18-month sleeper pipeline. Practitioners who run a win-loss review and a renewal-watch process can convert a no-decision into a qualified opportunity 12 to 18 months before the next procurement window opens.
✹ Procurement increasingly reads the response, the demo, the negotiation track, and the implementation handover as a single document. Cross-references, dropped commercial terms, and inconsistent language between phases now register as broken trust.
✹ The proposal practitioner who treats the document as the deliverable is becoming easier to replace. The practitioner who treats the document as one moment in a continuous record of trust is becoming harder to replace.
The centrifuge model describes the way many enterprise sales organisations structure their proposal function: sales runs discovery, the proposal team produces the document, the solution consultant runs the demo, and each function stands down at the end of its stage. The metaphor comes from the way work spins through a centrifuge and is ejected. The model worked when buyer evaluation was siloed. It is now breaking down because the buyer's record of the deal is continuous and searchable.
For most of the last two decades, proposal teams have been organised as a written-word service inside a sales organisation. The salesperson runs the discovery. The proposal manager produces the document. The solution consultant handles the demo. Each function owns a stage, and the deal moves between stages in a relay.
That relay model worked when the artefacts of the sale lived in different files. The proposal was a PDF. The demo was a recorded call. The implementation kickoff was a separate meeting. A buyer scoring the response read the response. A buyer scoring the demo watched the demo. The functions could stay siloed because the buyer's evaluation was siloed too.
Two things have changed.
First, every spoken and written word now lives in a single searchable record. Call transcripts, email threads, demo notes, written responses, and meeting recordings are all indexed and retrievable. Buying committees pull on those threads. As Greg Ellis, Senior Solution Consultant for enterprise deals at Zendesk, put it: every word is encapsulated and searchable within seconds. The buyer is no longer scoring a proposal. The buyer is scoring a continuous record.
Second, AI-assisted drafting has compressed the document phase. The work that justified a relay model, the labour-intensive writing of a 200-question response, is now compressible. The bottleneck has moved upstream into qualification and downstream into trust.
The result is that proposal teams who continue to operate as a centrifuge have lost the part of the deal where their judgement matters most.
Ellis describes the pattern this way. On larger enterprise deals he worked on as a proposal manager, the document itself accounted for about 20 percent of the elapsed time. The remaining 80 percent (negotiation, proof of concept, contract schedules, set-piece reviews) happened after submission. In the centrifuge model, the proposal team had no formal role in any of that.
The teams who win are the ones whose proposal function stays involved in some structural way through those later phases. That does not mean every proposal manager sits in every demo. It means the team treats the proposal as one moving part of a single conversation, and is structured to inject continuity at handover points.
There are three handover points where most proposal practitioners can build influence without renegotiating their job description.
Most content libraries are not connected to the company's current product marketing, pricing, or competitive posture. Ellis describes content libraries as islands. They sit outside the outward-facing site, the technical roadmap, and the sales-led messaging. By the time the RFP arrives, the library is six months behind the rest of the company.
The practitioner move: schedule a one-hour quarterly working session with product marketing and the lead solution consultant. Review the top 20 most-used responses in the library. Mark any that conflict with current pricing, positioning, or roadmap. Update them. This is library management dressed as cross-functional discovery, and it pays off the next time a response is farmed out to an SME under time pressure.
This is the trust gap. The proposal lands. The salesperson and solution consultant pick up the negotiation. Three weeks later, a commercial change happens. The price drops 40 percent. Ellis tells the story of a deal where this happened, and the champion lost the deal because she could not justify the price cut to her CFO. The company had re-engineered the solution and removed service levels, but the customer only saw a number move. Trust evaporated.
The practitioner move: any commercial change between submission and contract should be accompanied by what Ellis calls a commercial breach. A short, structured document that shows time A versus time B, what was added, what was removed, and what changed in scope. The proposal team is structurally well-placed to produce this because they already hold the submission baseline. The work takes an hour. It protects the deal.
Most companies have no formal handover from the proposal team to the implementation team. The submitted promises sit inside the deal folder. The implementation team starts from a clean slate, often re-interviewing the customer about scope. The buyer notices.
The practitioner move: produce a commitment register at submission time. This is a one-page extract of the response that lists every operational, commercial, and SLA commitment in the proposal. It goes in a permanent home that survives the salesperson moving territory. AI-assisted summarisation handles the first draft. The proposal manager validates it. The implementation lead inherits it.
The reason most proposal teams cannot do any of the above is that they are running too many bids.
Ellis calls this the dilution effect. The same team can run 10 proposals and win three, or run six proposals and win four. The dilution comes from spreading the final four hours of editorial polish, commercial review, and message alignment thinly across too many bids. Win rate is fragile in those last four hours. Anything that takes capacity out of that window compresses the win-rate curve.
The instinct to bid on everything is institutionally driven. Sales leadership treats every opportunity as worth a response. Strategic bids are added late. No-bid decisions get overruled. The practitioner is not in a position to change the company's bid culture overnight. They are in a position to instrument it.
What to track
Hours per bid, broken down by stage
Hours spent on bids that were no-bid in qualification and then reinstated
Win rate on bids where the final four hours were uninterrupted versus interrupted
Win rate by bid count per quarter
The data does not need to be precise. It needs to be enough to build a business case. Ellis points out that most companies have nowhere to file a win-loss report. The proposal team can be the function that builds that home, because no one else is structurally placed to do it.
A useful reframe for any practitioner under deadline pressure is that the buyer is not scoring the document in isolation. The buyer is scoring trust over time.
Ellis describes the procurement decision as having two layers. The first is whether the vendor can do the work. The second, which Ellis paraphrases from Rory Sutherland, is whether the decision can be retrospectively defended when put under scrutiny. The procurement lead needs a decision that holds up if the project goes sideways in 18 months.
That second layer is built across the entire interaction, not the document. It is built through consistency of language between the salesperson and the proposal manager. It is built through speed of response to follow-up questions. It is built by the team showing up to the negotiation with the same vocabulary it used in the submission. It is built by the implementation team inheriting the commitments that were made.
The proposal team is one of the only functions in a B2B organisation that touches every one of those layers if it chooses to.
For a practitioner who wants to break the centrifuge pattern without renegotiating their role, four moves are tractable inside the next quarter.
Track hours per bid. Even rough estimates are enough to demonstrate the dilution cost on bids that should have been no-bid. The data unlocks the harder conversation about pipeline discipline.
Build a one-page commitment register for every submitted proposal. AI tools handle the first draft. The register lives in a shared location with the customer name and a 12-month renewal alert.
Set a 12-month and an 18-month calendar reminder on every lost bid. At 12 months, do a win-loss review with the buyer if they will take the call. At 18 months, restart the nurture cycle ahead of the renewal procurement window.
Run a quarterly library health review with product marketing and the lead solution consultant. The 20 most-used responses are the ones that decide the next 10 bids. They should not be six months out of date.
None of these moves require new tooling. They require a structural choice to stay involved in the deal after the document leaves the building.
The proposal practitioner who treats the document as the deliverable is becoming progressively easier to replace. The practitioner who treats the document as one moment in a continuous record of trust is becoming harder to replace, because the value sits in the surface area no AI is currently mapping. The category will pull in that direction. The next two years will reward proposal practitioners who behave as analysts of their own deal record.
The dilution effect describes how taking on too many proposals reduces a team's collective win rate. The mechanism is the final four hours of editorial polish, commercial review, and message alignment before submission. Working on fewer bids at full focus produces a higher win rate than working on more bids at partial focus.
A quarterly review of the top 20 most-used responses is sufficient for most teams. The full library can be audited annually. Library decay tends to be concentrated in the responses tied to pricing, positioning, and roadmap, which change more frequently than the operational and security responses.
At submission, not at contract signature. The commitment register should be produced when the proposal is submitted and held in a permanent location. When the deal converts, the implementation lead receives the register on day one rather than re-interviewing the customer on scope.
A commercial breach is a short document that explains a price change between submission and contract by showing time A versus time B, what was added, what was removed, and what changed in scope. It is produced by the proposal team because they hold the submission baseline. It protects the buyer's champion when defending the deal internally.
Track hours per bid by stage, win rate by bid count per quarter, and hours spent on bids that were no-bid in qualification then reinstated. The data does not need to be precise. It needs to be sufficient to build a business case for declining strategic escalations that compress the final four hours of editorial review.