Read ICANN Blogs to stay informed of the latest policymaking activities, regional events, and more.

Addressing the Findability Problem on ICANN Sites

12 May 2015
By Chris Gift

In addition to the ICANN Languages, this content is also available in

The most common complaint I get about ICANN's digital presence is as simple as it is fundamental: 'I can't find stuff'.

Finding information on ICANN.org is difficult. The role of any web presence is to provide information and function to users and, while both may be present, it's not especially helpful if users can't find the content or function that they need.

The problem is, in part at least, born out of the distributed nature of ICANN as an organization.. That has led us to have an ecosystem of websites (rather than a centrally-run single entity) which means that content resides within  a range of different websites and domains - so, for example, GNSO working group content is on the GNSO site and the community wiki. It is not hosted on ICANN.org.

An Information Architecture for such a variety of domains run by such a range of people and organisations is quite the conundrum, making it difficult to create useful and usable user journeys

So users need to rely on search to find content - but searching on a range of domains is a challenge too, with fewer solutions. We use a Google enterprise solution as the least-worst option, but it's a black box product and can't be tailored to our needs. We've investigated more bespoke solutions, including one called Solr, which could have brought useful functions like facetted search but, again, that's not possible in a multiple domain presence. We could only search the one domain (ICANN.org), and not others such as newgtld.icann.org, community.icann.org. Which is a big hole.

Once we get beyond the domain factor, we run into (search) issues around the content itself, where we have no shared idea on standardization around the key components of search-friendly content:

  • Titling
  • Versioning
  • Format
  • Missing meta data
  • No shared glossaries around both meta data and text

Having a shared approach on all the above gives any web presence the sort of content discipline which allows site managers to organise and present content and makes that content show up on searches. Without that, which is where we are, things get difficult.

That lack of a shared approach comes because, for very good wider organizational reasons, there is no editorial ownership, no single approach to the creation and curation of content. Hence there is a lot of content which is stale or out of date (and when people say they can't find what they need, they also mean they can find stuff they know has been usurped). And the content that is there, doesn't share standards on terminology and meta data especially, which is essential if its to be presented properly and findable in search.

Arguably the worst example of all these issues is the Resources section on ICANN.org - we have a site map (newly posted on the Community Wiki), but as it stands it does (at best) betray it's 'organic' beginnings and has become a dumping ground for content that doesn't-quite-fit anywhere else. We made an attempt on redesigning it in the last redesign process but the complexity of the issue, on limited resources, defeated us. It's still deeply unsatisfactory

So what next?

We're working with the various structures to support them as they redesign their websites. - so now might be a good time to discuss bringing SO-AC sites  under the ICANN domain to help us break that search/domain issue; and begin addressing the idea of shared approaches to titling, meta-data and terminology..

We're hiring an Information Scientist (a librarian in old parlance) to help us with that standardization and content management. Our attempts to enforce improved internal standards on meta-data has been enforced less rigorously that I would have liked, so hopefully the new librarian can crack the whip a little stronger.

And, now that we have our first ever user experience specialist, we can look to improve those user journeys and begin to tackle the rag-bag that is the Resources section, although always (of course), with full consultation with the community.

There's some big questions in here about domains, content and editorial standards (and, by implication, control) so we need to consult widely on this. But the findability of content is a basic function that we cannot afford to neglect. So we need to talk widely and often about the changes we intend to make - conversations to be held by myself and the rest of the Digital Services team and, in this case especially our new UX person and the incoming Information Scientist. As ever, I'd be interested in your thoughts - by email (chris.gift at icann.org), or in the comments below, or just collar me in our next conference call or meeting. This is too important for us to not get it right.


Chris Gift

Chris Gift