What is data monetization?

Monetization is the widest word in this family. Researchers at MIT's Center for Information Systems Research describe it as converting data into financial returns, and they warn against two misreadings: treating it as only selling datasets, or stretching it to any benefit at all from using data. The test is whether money, as revenue or savings, actually shows up.

Commercialization and licensing

Commercialization is narrower: it covers only the external route, bringing data to paying outsiders. Licensing is narrower again: it is the legal instrument used in most external sales. Keeping the three apart avoids muddled plans where a team says monetization but means a sale to third parties.

1. 1

   Improving

   Using data to make internal work cheaper or faster, such as better stock forecasting. The return is a saving.

2. 2

   Wrapping

   Adding data features to a product the company already sells, such as a usage dashboard for customers, so the product earns more or keeps customers longer.

3. 3

   Selling

   Packaging data or insight into an offer that outsiders pay for directly: a dataset licence, a feed, a report or an API.

Only the third route involves a buyer outside the company, and only the third route needs a broker. The first two are usually run by internal analytics and product teams.

Suppose a maker of commercial refrigeration units collects temperature and compressor readings from units in shops. Improving: its service team uses the readings to schedule repairs before units fail, which reduces warranty costs.

Wrapping: it adds an energy report to its customer portal and charges a higher tier for it. Selling: it licenses de-identified, region-level failure patterns to an insurer that prices equipment cover.

Each return is real, but they ask different questions. The first two stay inside existing customer contracts. The third needs the maker to check whether its contracts with shop owners allow the readings to go to an insurer at all. In this illustration, that third step is where a data broker would get involved.

When a company says it wants to monetize its data, ask which route it means. Many executives picture selling, when their quickest return may be internal. A broker adds value on the selling route only, and should say so. Being clear early saves weeks spent preparing a sale that the company's own contracts or strategy rule out.

• Counting value that was never realised. A dashboard nobody pays for is not yet monetization.
• Assuming data collected for one purpose can be sold for another without checking contracts and privacy notices.
• Ignoring the cost side: cleaning, documenting and supporting a dataset for buyers takes ongoing work.
• Treating any market-size figure as a forecast of what one company's data is worth.