The cost of data

Much has been written and discussed in recent years about the many social and economic benefits resulting from open data, but not about the total cost that these open data initiatives may incur.

There are also frequent statements that say that the cost of an open data initiative is very small, especially compared to the potential benefits but, however, it is very difficult to find nowadays actual figures which serve as a proof of such statements.

On the other hand, in some occasions attempts have been made to make a more accurate estimation in terms of cost nationally and through similar analyzes in other countries. In these cases we find highly variable results ranging from a few hundred euros to almost 400,000. Moreover, in general, it is common to make the mistake of associating the cost of an openness initiative only with the cost of development and maintenance of portals that serve to hold open data catalogs and support the initiative, often leaving aside other key and very significant aspects, such as the cost of managing the internal organizational change assumed generally by the entities.

This current scenario still leaves the big question unanswered: What is the total cost of launching an open data initiative? Fortunately we also have other resources that can help us make more complete and accurate estimations:

1. The price list of services offered by Socrata (one of the international companies in the sector) for US market in 2014 that can help us know ​​how much may increase the costs of such initiatives depending on the complexity of the systems used, with services ranging from a few hundred dollars for the most basic services to more than $ 30,000 per month for the most complex;

2. Some case studies as the one of the Danish Government, in which the joint publication of address databases in an open format was carried out with a minimum investment of 2 million euros, achieving a direct benefit of 62 million euros over a period of 5 years thanks mainly to the improvements in efficiency;

3. Internal audit exercises of the governments, as in the case of the audit conducted in the UK that shows some interesting overall figures: 2 million pounds per year costs the maintenance of  the data catalog (less than 0.5 % of the annual budget); added costs between 53,000 and 500,000 pounds per department for adaptation of existing data or other examples such as:

a) more than 500,000 pounds dedicated to the creation and improvement of crime maps (plus another 150,000 pounds of annual maintenance);

b) 400,000 pounds for the publication of data on international development projects (plus another 64,000 pounds a year for maintenance);

c) estimated cost of 275 million pounds (and benefits 446 million) for the publication of clinical trial data for a period of 10 years.

However, we have the possibility to make our own estimates using the basic guide to estimate the cost of open data initiatives in their planning proposed by the Open Data Institute which includes technical aspects, administrative capacity and teamwork as marginal costs due to the maintenance, coordination and development of the initiative. An example of application of these estimates is that made by the city of Gold Coast in Australia, which gives a figure of between 6.5 and 16 working days for each set of published data.

It is easy to conclude how complicated it can be to specify a single reference quantity for any case, since the cost of each specific project depends on many variables ranging from the existing infrastructure to the amount and quality of data or the complexity of internal management procedures.