Community polls: who raised your price this year?

Every procurement lead has had this call. The account team says the increase is standard across the base this year, that everyone is absorbing it, that you are actually being treated well relative to comparable accounts. You have no way to test that sentence in the moment. Your benchmark data describes closed transactions, which is exactly what you want when you are arguing about a rate, and exactly the wrong instrument when you are arguing about a direction of travel that started eight weeks ago. So you nod, you say you will come back, and you spend the next fortnight asking around. We shipped a board that does the asking for you.

The named problem: the uplift you cannot source

Call it the sourcing gap on current cycle behaviour. Pricing evidence is strongest where it is oldest. A signed contract is a fact, an invoice line is a fact, and the transaction record behind our benchmarks is the strongest thing a buyer can put on a table. But a publisher who decides in January to push a mid single digit uplift across the installed base does not appear in signed contract data until the renewals actually land, and by then your own renewal has landed too.

Buyers fill that gap informally. A message in a user group Slack. A quiet question at a conference. A former colleague who now runs IT sourcing at a similar sized firm. The information is real and it is often accurate. It is also unsourced, unrepeatable, and useless in front of a CFO who wants to know how many companies you actually spoke to. We wrote about the same tension in buying blocs, where the value of pooled leverage keeps colliding with the very reasonable refusal to expose your own numbers. Polls are the low commitment version of that trade. You give up one bit of information, whether a named vendor raised your price at your most recent renewal, and you get the aggregate back.

The fourth board, and why the results stay veiled

The trading floor now runs four boards. The newest asks a single question: which vendors raised your price at your most recent renewal? You select all that apply from the 20 tier one publishers, the names that account for the bulk of enterprise software spend and the bulk of the arguments about it. Multi select matters here, because the interesting pattern is rarely one vendor in isolation. It is the buyer who marks four of the twenty and realises their whole estate repriced in the same twelve months.

Ballots are anonymous. Nothing on the board ties a selection back to a company, an account, or a person. The more consequential design choice is the veil. Results stay hidden until your ballot is in. You cannot browse the tally, form a view about what the answer should be, and then vote in a way that agrees with it. That is not a courtesy to us, it is a defence of the data you are about to rely on. Anchoring is the single cheapest way to ruin a crowd signal, and a poll that lets you peek first is a poll that slowly converges on whatever the first fifty voters happened to say.

The cost to you is one click before you get the answer. The benefit is that when the board tells you eleven of the last set of respondents marked a given publisher, that number was not manufactured by people agreeing with each other.

Walking the floor without losing your place

A fourth board makes the floor more useful and also longer. Three changes shipped alongside it to keep the whole thing walkable in a single sitting.

First, a board index. It marks which boards are still waiting on your ballot and follows you as you scroll, so you always know how much of the floor you have actually covered. Before this, the honest failure mode was a buyer voting on the first board, getting absorbed in the results, and never reaching the third. Second, the coverage board's full vendor list is now searchable, with your existing picks pinned above the search field. If you are looking for one specific publisher among a long list, you type three letters instead of scrolling, and you never lose sight of what you already selected. Third, every live board now hands you to the next open one when you finish. Vote, read the tally, continue. The floor behaves like a sequence rather than a page you have to re navigate.

Where this sits in the renewal workflow

Polls are not a replacement for benchmarks and should never be used as one. They sit earlier, at the point where you are deciding how hard to push and what story to tell internally. The sequence that works is roughly this. You see a renewal approaching. You cast a ballot on the price increase board and read the tally to understand whether the uplift you have been quoted is a base wide move or a move aimed at you specifically. Those two situations call for completely different tactics, and until now you were guessing which one you were in.

If it looks base wide, you pull the benchmark and argue about where you sit in the distribution, because arguing that the increase should not exist is a losing position when the publisher has already applied it to thousands of accounts. If it looks targeted, you go looking for what changed on your side, and the year on year compare usually finds it. Either way the poll result is a routing decision, not the argument itself. Pair it with Terms Watch and you get both halves of the picture, what the published list did and what buyers actually absorbed.

Honest limits, stated plainly

This is self reported sentiment from a self selected group, and you should treat it that way. Four limits are worth naming before you rely on it in a negotiation.

It is directional, not quantitative. The board asks whether a vendor raised your price. It does not ask by how much, on which product line, or against what baseline. A buyer who absorbed a two percent uplift and a buyer who absorbed twenty percent both register as one mark. If you need the magnitude, that is what the benchmark record and the comparable deals behind a given analysis are for. Do not quote a poll where a percentile belongs.

Participation skews. People who just got hit with a large increase are more motivated to vote than people whose renewal was uneventful, which pushes any tally upward. Read the shape of the result rather than the absolute count, and be more interested in the gap between two publishers than in either number on its own.

Sample size varies by vendor. The tier one publishers with the widest install bases will accumulate ballots quickly. A specialist name in the same list will move slowly, and a tally built on a small handful of responses is a conversation starter, not evidence. The board shows you the count. Use it.

And a mark on the board is not a like for like comparison. A renewal that repriced because the publisher moved its list is not the same event as a renewal that repriced because the buyer added seats, changed metric, or walked into a co term that removed their leverage. The poll cannot tell those apart. You still have to read your own paper.

None of that makes the signal worthless. It makes it a first instrument rather than a final one. Cast your ballot on the trading floor, read the tally against your own estate, and then go get the numbers that will actually sit in the counter offer. The board tells you which fight you are in. The benchmark tells you how to win it.