What a one-page data manifest is, and why buyers price from it
A manifest is an inventory, not a dataset: what records exist, in which systems, over how many years, with what rights. It is the page buyers price from, and the reason you never send data to get a number.
Owners who get an inbound offer for their data usually assume the next step is to send some of it. It is not. The document that moves first, and the one buyers actually price from, is a single page that describes the data without containing any of it. That page is the manifest.
A manifest is an inventory. It says what records exist, in which systems, over how many years, in what volume, and under what rights. It does not include a single customer name, email address, or ticket. It is closer to the schedule of assets in a sale agreement than to a data room. If you can describe your company's systems to a new IT hire, you can produce one.
What is on the page
A complete manifest answers eight questions, usually as eight rows of a table.
Systems. Which platforms hold the records: the helpdesk, the CRM, the email tenant, the chat tool, the document store, the billing system, the call platform, the engineering tracker. Named by product, because buyers know what each export looks like.
Record types. What each system holds: tickets with resolution notes, deals with stage history, threads, recordings with transcripts, proposals, invoices and disputes.
Date range. The earliest and latest records that still exist, per system. Migrations matter here. If the helpdesk moved in 2021 and the old data was not carried over, say so.
Volume. Approximate counts: tickets, messages, documents, hours of calls, closed deals. Round numbers are fine.
Format. Whether a native export exists, whether attachments are included, whether transcripts already exist for audio.
Outcome signals. Whether records carry results: resolution codes, satisfaction scores, closed-won and closed-lost, dispute outcomes, merged or rejected. This row changes the price more than any other.
People and rights. Who appears in the records (employees, customers, third parties), what your contracts and notices say about use of that data, and whether any of it is licensed from someone else.
Exclusions. What you will not license: privileged material, regulated categories, a specific customer, a specific period.
That is the whole document. Eight rows, one page, no data.
Why buyers price from it
Buyers price from a manifest for the same reason a lender prices from a rent roll rather than walking every unit. The manifest carries the variables that drive value, and a buyer who has read many of them can benchmark a new one quickly.
Depth and continuity come from the date range and volume. Learnability comes from the outcome signals and the record types. Usability comes from format and rights. A buyer building a customer-service model this quarter reads the ticket row first; a buyer working on sales agents reads the CRM row. The same page is priced differently by different buyers, which is exactly why it should go to more than one.
The manifest also lets a buyer say no early, which is a feature. If the rights row shows that half the records belong to a customer under a contract that forbids reuse, the buyer learns it on page one instead of after a sample, and no one has wasted a month.
What a manifest is not
It is not a dataset. Nothing in it can identify a person.
It is not a sample. Samples come later, are small, are de-identified, and are approved by you by category or record by record.
It is not a data room. There is no upload, no shared drive, no connector. The manifest is a document you write and approve, and it is the only thing that passes through a seller-side agent.
It is also not a promise. A manifest describes; diligence verifies. A first offer based on a manifest is indicative, and binding terms follow sample review.
Building one in an afternoon
Start with the systems list. Walk through the tools your team logs into every day and write down each one. Ask the administrator of each for two numbers: how far back the data goes and roughly how much there is. Note which systems have been migrated and what was lost.
Then mark the outcome signals. For each system, ask whether a record shows how things ended. Tickets with resolution codes, deals with stages, disputes with rulings. If a system only shows activity, say so. It still has value, just less.
Finally, read the rights row with your counsel. Customer agreements, employee policies, privacy notices, and any content you license from a vendor. Most companies find they can license most of what they have, with specific exclusions. Write the exclusions down.
A manifest that honestly says six years, 40,000 tickets, resolution codes on most of them, and two excluded customers is worth more than one that claims everything and survives nothing.
What makes a manifest weak
Vague systems rows, such as the single word email with no tenant, range, or volume. Volumes that do not survive a sample. A rights row left blank. Outcome signals left unmentioned when they exist. A single line for ten years of data that actually lives in three systems with two migrations between them.
Weak manifests do not produce low offers so much as slow ones, because every buyer has to ask the same questions before pricing, and some will not bother.
From manifest to offer
Once the page is complete, it is benchmarked against what vetted buyers are paying for comparable records right now. That is what makes a first offer in 24 to 48 hours possible: the pricing work happens before any buyer sees your name, and the buyers see a manifest, not a dataset. The offer that comes back is a range, labeled as indicative, with binding terms to follow after the buyer reviews a sample you approve.
If you want to know what your records are worth, the manifest is the first step, and the only document you will need to produce to get a number.
Questions worth asking
Do I have to give anyone access to my systems to build a manifest?
No. A manifest is written from what you already know about your own company: which systems you use, roughly how many records each holds, how far back they go, and who appears in them. Most of it comes from an afternoon with whoever administers your email, helpdesk, CRM, and file storage. Nobody connects to anything. The dataset stays where it is until a contract is signed, and then it moves directly to the buyer.
How accurate do the numbers on a manifest need to be?
Accurate enough to be honest, not exact. Round figures are normal: about 40,000 tickets, roughly six years, a few hundred recorded calls a month. Buyers expect approximations at this stage and verify them during diligence against a sample. What hurts is overstating. A manifest that claims ten years when a helpdesk migration kept only the last four will be found out at sample review, and the offer will move down.
What happens after the manifest is approved?
The manifest is benchmarked against what vetted buyers are currently paying for comparable records, and a first offer comes back as an indicative, buyer-benchmarked range within 24 to 48 hours. If you want to proceed, buyers review a small de-identified sample you approve, complete diligence, and return binding terms covering price, exclusivity, field of use, term, and deletion. You sign with the buyer you choose. The first offer is a process target, not a guarantee.
Measured from a complete data manifest. A first offer is a process target, not a guarantee; buyers decide eligibility, price, and terms. Indicative value ranges mentioned on this site are illustrative, not quotes. Nothing here is legal advice; have your own counsel review any agreement before data moves.