When is a Data Utility Not Really a Data Utility?

As more vendors announce their plans to overhaul traditional approaches to data management, the terms ‘utility' and ‘managed service' are increasingly being used interchangeably. However, there is a distinction between the two.

A utility is a multi-tenant environment in which data is aggregated and cleansed once, and the same golden copy is then provided to multiple users. This model is attractive because it is more efficient and therefore cheaper than the common scenario, in which a vendor does the same data management work separately for each of its clients— who, in fact, all have the same requirements.

Like a utility, a managed service has a central, shared platform and operations team, which help to create economies of scale. However, the difference is that the data produced by a utility is completely standardized—exactly the same for all consumers— whereas a managed service produces a bespoke golden copy, tailored to the needs of individual users.

As discussions about utilities progress, it is becoming clear they face a number of obstacles. Utilities are well suited to the management of public-domain data, which they can redistribute freely, but they face a much harder time when dealing with data that is subject to a license.

Some data vendors are happy for the operators of utilities to take their data, manage it and redistribute it to the users of the utility. But the data vendors will, of course, require the users of the utility to pay them for the data they receive. As a result, the utilities have to install metering systems and controls.

Some observers believe that to really qualify as a utility, its operators must source the data themselves, so that it is not subject to redistribution licenses—a requirement that certainly raises the bar for all would-be data utilities.