Over the last few months, I've been thinking about open data, partly at where I used to work in my old job, and in my own time.
The field is currently very young, so I thought it worth sharing some examples that I’ve found particularly interesting.
Four ways of making money around open data
Selling access based on timeliness
In some industries, timeliness is the difference between making a decision that makes millions, or one that gets you fired. Bloomberg have made a lot of money from selling the same data they make available publicly, available a few milliseconds earlier to paying customers on trading desks, and across newswires.
In the open data world, the musicbrainz database, used by the BBC for its programming, and its live data feed product is a good example of the same open data being sold on timeliness, rather than content.
Selling more detailed versions of the data
Elsewhere, giving away data, to sell access to more detailed data is common, and a tried and tested approach, used when selling business information.
While this is hardly new, one of the most interesting examples would be Duedil - you can retrieve a wealth of information on their public website, (the kind you’d easily pay hundreds, if not thousands of pounds for access to), but sell premium services for further data, beyond what is available on their pages, and for certain kinds of API access.
Letting people pay to go private
In some cases you might want take openly licensed data, and either build upon it, or incorporate it into an existing body of info, to give your business an edge over competitors.
In this case, you’d license the data under different terms - this approach is often called the dual licensing approach, and is common in the software industry. Oracle and 10Gen both take this approach, licensing versions of some their respective products, MySQL, MongoDB differently.
In the world of open data, Open Corporates is another good example; you’re free to use their data, as long as you share it under the same terms, and credit them as the source. If you want to use their database without sharing the changes back (say, in a product you’re building that uses the data that you want to charge), the same data is available under different terms.
Selling services around the data
Another common use of this would be providing services that tailor how existing data is presented, to provide specific insight more quickly, more cheaply, or in ways simply not possible for a customer to access themselves.
Another good example of turning open data into money like this would be the work done by the [Prescribing Analytics] project for the NHS.
A group of doctors and technologists took anonymised, open data provided by the NHS, about which drugs were prescribed each year around the UK, processing it to find patterns in how the NHS buys drugs.
Among other findings, they found patterns where branded drugs, costing 10 times as much as generic, off-patent equivalents, were routinely being prescribed, for the same therapeutic benefit to patients.
So far, this has uncovered potential savings of around [£200million in 2013, just when looking at single drug so far].
Hopefully, I've shown that while open data is an interesting idea, it's not so new and different that all existing knowledge about how to run a business around information ceases to apply.