Aims and Objectives

The Royal Institute of British Architects (RIBA) is a membership organization which champions better buildings, communities and the environment through architecture.  As an organistion Its probably best known to the public for its architectural awards such as the annual Stirling Prize.

In 2012 RIBA commissioned VML to undertake a redesign of the website which had been in its current state for a number of years and was no longer fit for purpose. Their existing website was aimed very much at the professional audience and had grown somewhat organically over the years meaning that a lot of good quality material was buried deep within the website. It also suffered from having dozens of microsites which various departments had developed themselves mostly out of frustration with the inadequate website. The main goal for the website however was not just to redesign the existing website but to realign the focus towards the public, capatalising on a renewed public interest in Architecture from TV programs such as Grand Designs while still maintaining its usefulness as a resource for Architects.

Approach

Website analytics research

Some work had already been done to look at the public audience from a marketing perspective but none for the web. The first step was to look at the analytics for the website to try and establish who the audience is, where they are coming from and what they are looking for.  Looking at the search logs helped a lot and told us that once I had separated out the more professional search queries a lot of people were coming to the site to look for Architects to commission and also for inspiration for their building projects.

Content Audit

To get a sense of the size of the site I undertook a content audit which entailed looking at all of the content areas on the site then looking at the traffic and bounce rates for each. This was useful as it highlighted large areas that were rarely visited because the limited value to the site visitors and also more good quality content that was buried 8+ levels deep.

Survey

I wanted to get a better understanding of why people were visiting the site to flesh out some of the findings from the search logs so I designed a site survey.  The survey ran for a month and we had a good response rate which confirmed some of the findings from the analytics research but also raised more goals which we could use.

Stakeholder Interviews

Having got a better idea of the user goals I wanted to understand the organizational needs which meant talking to stakeholders from all the departments and members of the board. I ran through a series of exercises that got them to think not only of the present state of the site but also their aspiration for the future. I was looking to find shared agreement about the direction of the site and also how they saw their department fitting into it.

Personas and User Journeys

By this stage I had built up a good understanding of the user from my work and also from the previous research so to ensure that the team at VML and RIBA understood who the audience I designed a series of personas that included the public, students and professionals. I also designed a series of high level user journeys based on the user goals which would be useful for when we moved into the design phase of the project to sense check the decisions that we would be making further down the line.

Domain modelling the Site Map

The main challenge with designing the website was that we were starting from scratch. I needed to design a framework for a site with limited existing content but which could be built on over the coming years.

Building upon my experience with the BFI I decided to use domain modeling to design the framework for the new site. I chose a subset of Dublin Core metadata as it was already a well established set of terms and would be flexible enough to fit our model for the new Explore section of the website.

Out of the set of 15 Dublin Core terms I chose Title, Creator, Date, Location and Subject. The thinking being that once you are looking at a page about a building you could see linked information like the Architect (Creator) which in turn could link to their buildings (Titles). The Buildings would also have been built at a date in time in a location and also have subjects attached like for example building style, materials etc. Initially we would be representing Building, Person and Subject pages but in the future there is no reason why there couldn’t also be Date and Location pages, all automatically generated as new content is added.

The Dublin Core taxonomy would also be used for populating the explore section pages  with aggregated content. RIBA already have a lot of media content that could be used on the website to populate pages. We decided early on that we weren’t trying to develop an encyclopedia of buildings at this stage but launch with a subset of pages we could be built on over the years. The existing media could quite easily have its metadata bought inline with the new taxonomy then and poured into the CMS generating pages where they didn’t previously exist and adding to pre existing pages. This would work by the CMS aggregating terms and grouping them together on a page automatically, for example all of an architects buildings on the architects page.