The licence stack: what you buy when you buy a data business

Buyers of data businesses tend to value the revenue line and the renewal rate, both of which are visible and neither of which is the asset. Underneath the product sits a stack of agreements: source licences that permit collection, redistribution rights that permit resale, client contracts that define permitted use, and obligations about deletion, audit and territory. The company owns the stack rather than the data. Where the stack has a hole in it, the revenue is temporary and the buyer inherits the hole.

Source licences are read for three things. The first is scope: which fields, which territories, which purposes. The second is derivation, meaning whether an output computed from the source may be sold without the source, which decides whether the business has a product or a distribution agreement. The third is change of control, because a licence that lapses or reprices on acquisition hands a negotiation to the buyer at the worst possible moment. Our desk has withdrawn from two processes on the third point alone.

Client contracts matter in the opposite direction. A data business that has promised perpetual internal use to its twenty largest clients has sold its future pricing power for present revenue, and the renewal rate that looks so attractive is measuring an obligation rather than a preference. We model the revenue line twice: once as reported, and once as it would stand if every contract repriced at renewal on current market terms. The gap between the two is the real negotiating position.

Provenance sits above all of it. The firm requires that every dataset it holds or acquires has a documented origin and a lawful basis, and that requirement applies to a target holding before completion rather than after. A collection whose origin cannot be evidenced is not discounted. It is excluded from the model and, where it is material, from the transaction. That position has cost us assets. It has also kept the firm out of two businesses whose principal holding proved undocumented.

Prices in this sector are high for reasons everyone understands. Data businesses have low marginal cost, high switching cost and a customer base that budgets annually and cancels rarely. Those properties are genuinely attractive. They are also widely understood, which means the advantage in this sector is not identifying the model but reading the stack faster and more sceptically than the next bidder. That is a research capability, and the Information and Data division has been building it since 2019.

Deployment sectors are where origination proceeds go, and data analytics has taken the largest single share of ours since the first realisations. The logic is symmetry. Origination sectors earn returns because a regulator or a biological process makes entry slow. Data businesses earn returns because a contract stack and a switching cost make exit slow for the customer. Both are structural. One is bought with patience and the other with reading, and the firm intends to do both.

Published 2021-05-19 by the Information & Data division. Research is prepared for eligible counterparties and does not constitute advice.