XFERDATA
Fifty rows is the first thing a data buyer asks for

What a data buyer asks for before they will bid

The valuation is the easy part. If you take a dataset to auction, here is what a serious buyer will want before they put a number on it, and roughly why.

A sample that represents the whole

Usually around fifty rows. Not your cleanest fifty rows. The point is to show the shape of the data, including the parts that are messy, because a buyer who discovers the mess later will reprice or walk. Include the nulls.

Five years of lagged history

Lagged means old enough to be useless commercially but long enough to be testable. A buyer takes it, runs their model against outcomes they already know, and finds out whether your data would have told them something. This is the step where most deals are actually won or lost, and it is why history depth matters more than any other single attribute.

A data dictionary

Field names, types, units, and what each one actually means. This sounds trivial and it is the thing most often missing. A field called status_2 with values 1 through 7 and no explanation is worth nothing to somebody outside your company.

Provenance

Where each field came from, whether it was captured or derived, and what happens to it between the source system and the export. Buyers are increasingly unwilling to take derived fields on trust, because a derivation they cannot see is a dependency they cannot manage.

The right to collect and the right to share

Two separate questions, and people routinely conflate them. Having lawfully collected data does not mean you may license it. This is where terms of service, privacy notices, and any contracts with the counterparties in the data all get read carefully. It is also where most datasets that fail, fail.

Personal data, handled properly

Anonymisation is not a checkbox. A dataset with names removed can still identify people through combinations of fields, and a buyer who inherits that problem will not buy. Expect this to be examined rather than asserted.

An exclusion list

Your right to say which buyers may not see it. Competitors, customers who would object, anyone you would rather not be in business with. Any process that does not offer you this is not one you should be in.

What this adds up to

It is more work than filling in a form, and most of it is work any serious buyer would ask for regardless of who ran the process. Done once, properly, it can be reused across every bidder rather than repeated for each one.

The order matters too. Doing the compliance work before an auction opens rather than after a buyer asks is the difference between a process that closes and one that stalls in legal for a quarter.

This is the market we run

Xferdata values a company's data, builds the auction pack, and runs the auction to buyers globally.

If you own the data, run a valuation and find out what yours is worth.

If you buy data, sign up and tell us what you are after. You will see packs as they come to market.

Sign up to Xferdata