By 
Sixteen buyers. One seller. One afterthought line on the invoice. That is the whole billing argument in a sentence, and almost nobody sees it coming until the quote lands.
Here is the claim I will defend: virtual data rooms bill by the project because buyers multiply like weeds while the project itself stays one thing. Price the room by seats and you punish every seller who runs a decent process, since a competitive auction is exactly when your review team of five swells past four hundred signatures. Price it by the deal and the number stops moving while your deal does its work. If you want to see how that logic plays out in published rate cards, look at typical virtual data room price tiers, where the variable being sold is the shape of the transaction, not your guest list.
So the useful question is not which vendor is cheap. It is which billing unit matches what your deal actually does when it hits the market.
What exactly are you renting when you open a data room?
Not software, or not mainly software. You are buying a gatekeeper, a numbered archive, and someone who can put deal documents in front of a buyer in Tokyo at 2 a.m. without you personally clicking anything.
That is why the meter behaves so strangely. A room holding four hundred documents costs almost the same to host as a room holding four thousand. Hosting is not the cost driver. Attention is. Every added bidder means more permission sets, more questions routed through the Q&A queue, more export logs somebody may eventually have to defend, and a support team that stays awake across time zones. Storage on a typical entry tier runs somewhere in the low gigabytes, so picture a folder with a few thousand PDFs, not a media library.
Seats look like the fair measure because they are easy to count. They are also the worst possible proxy for what the room actually spends on you, since a bidder who logs in once and a bidder who downloads everything cost wildly different amounts to serve.
Three billing models, and who quietly loses under each one
Almost every quote you receive collapses into one of these shapes. Mixing them up is how teams end up comparing numbers that were never describing the same thing.
| Model | What moves the price | Who it fits |
|---|---|---|
| Per project, tiered | Complexity, administrators, service level, storage band | Auctions, cross-border deals, recurring sellers |
| Per seat or user | Headcount with access | Narrow processes with a fixed buyer list |
| Selected pages | Page count uploaded | Small, tightly scoped diligence reviews |
Per seat punishes the seller with the most competitive process, which is upside down. Page counting sounds tidy until you realize legal teams upload in bundles and nobody trims a binder mid-negotiation to save a few dollars. The per project tier flips the incentive where it belongs: your price is set by how complicated the deal is, and then you can invite every credible bidder without wincing at the counter.
I would take the project tier nearly every time. The exception is a genuinely fixed process with three known bidders and no chance of a late entrant, and even then I would want an overage clause spelled out in ink rather than in a sales email.
A pricing walkthrough you can run this week
The fastest way to gut a bad quote is to pressure test it against your own deal rather than a competitor’s feature list. Work through these in order and disagree with me where you have real evidence.
- Write the bidder count in pen, then write your worst-case bidder count in pencil next to it. Quote against the pencil.
- Ask what a second administrator costs and what triggers an upgrade to the next tier. Headcount inside your own team is usually the real trigger, not outside users.
- Check whether storage is capped as a number or described as a band. Bands are fine, unbounded promises of unlimited storage usually come with a clause attached further down.
- Ask what disappears when the deal closes. Access expiry, final export format, and archiving terms belong in the contract, not in the FAQ.
One pattern holds across nearly every vendor I have looked at: the expensive quote is rarely the one with the biggest number. It is the one where the scope drifted after signature.
Rights and records are the line item everybody skips
Two questions decide whether your room was actually worth paying for, and neither shows up in the price table.
The first is control. Every document you circulate to a bidder carries confidentiality terms, and the ability to pull access, revoke a download, or prove who saw what after the fact is a legal instrument as much as a technical one. The U.S. Copyright Office is a useful reference point for anyone selling assets where content rights travel with the deal, because rights paperwork has a habit of arriving late and slowing a signature you already scheduled.
The second is whether you can show your work. If the buyer asks how you handled personal data in the room, logs, retention rules, and access reviews are your answer before any regulator asks. The Federal Trade Commission is a reasonable starting point for how seriously American authorities treat the handling of consumer information, and a room that quietly keeps every bidder’s download history forever is not the safe default people assume it is.
My old boss had a phrase for this and it stuck: pay for the exhibit, not the filing cabinet. The exhibit is the part you can prove.
What to negotiate instead of the headline number
Discount hunting on the sticker is the least productive hour you will spend. Vendors hold the line on rate and fold on scope, so ask for things with expiry dates attached.
- Bidder headroom written into the contract, so a late entrant does not trigger a re-quote.
- A stated service level with a name behind it, not a general assurance of support.
- A clean exit: export format, retention window, and deletion confirmation after close.
- A cap on the number of administrator logins before the next tier activates.
Access control sits at the center of all four, which is why security frameworks make useful negotiation language. The National Institute of Standards and Technology publishes the security guidance most enterprise buyers already expect their vendors to follow, so a room that can speak that language tends to survive procurement review without a second round of questions.
Here is the scenario I use to explain this to juniors. Picture a mid-market seller with four serious bidders in the first week. Under per seat pricing, that seller pays four times for the privilege of running a better auction. Under project pricing, the invoice holds steady while the bidder count does the work of pushing the multiple up. The billing model is quietly deciding who gets rewarded, and it should not be the vendor collecting more for your competitive process.
Pick the model that rewards the deal you are actually running
Look at your last three deals, not the glossy comparison chart. If bidder counts swung unpredictably, project pricing is your insurance policy. If your process is genuinely narrow and repeatable, seat pricing can work, provided the contract covers a late entrant at a defined cost. And if you sell assets where rights and records travel with the deal, the paperwork questions matter more than the tier name on the quote.
Before you sign anything, ask the vendor to price your worst-case bidder list rather than your optimistic one. The gap between those two numbers tells you everything about how the contract will behave six weeks in.
What does your own invite list look like today, and what happens to your budget if you double it next month?
Comments