Top Design Tips for Improving Open Data Sites

In a blog post for Scientific American author Cesar Hidalgo has reflected on the key problems with Open Data sites. He summarises that despite the rapid uptake of sites such as, there remain barriers to their effective use and application.


So, what are the top ways to make sure open data sites offer the highest amount of potential they can?


1. Aesthetic Design.

It sounds simple, but a lot of sites manage to get this very simple element wrong. If the site is difficult to read, has no explanation, or isn’t obvious as to how it’s used then people ultimately won’t use it. This is less of a problem for companies and startups sites as often designer are employed to consider aesthetics carefully before anything is created. With larger, governmental or research based sites, aesthetics needs to be a higher priority to ensure users aren’t put off before they even get around to using open data.


2. Organisational Design.

Hidalgo firstly explains that most open data sites are poorly designed in terms of the conceptual model used to organise and deliver data to users. This should be improved to provide data in the way that it is used, as opposed to the way in which it’s collected. Secondly, datasets shouldn’t be organised in the order of their creation. Sites need to consider how they are delivering data back to the user as all too often they throw irrelevant secondary data at search queries, despite not fully fitting the request.


3. Usefulness Design.

Finally, sites shouldn’t bury data in the deep Web to ensure that search engines can easily get hold of it. This is important to bear in mind as search engines aren’t clued in to knowing exactly what piece of information a user is looking for. Hidalgo instead states that “to make open data truly open, we need to make it searchable, and for that we need to bring it to the surface of the Web.” There are many potential solutions to improving existing sites. The creation of sites that merge multiple datasets and transform them into “stories” is one such method endorsed by Hidalgo himself. In this, profiles are created rapidly which feature visualisations and text thus making user engagement quick and informative.



Using open data should be easy. The easier it is, the more effectively it can be applied to countless scenarios and jobs internationally. Ultimately, the goal of open data should not be simply getting hold of some files, regardless of their potential, but it should stimulate our understanding of the systems that the data describes. Design in each of the above iterations is the key to mastering this.