Building my homepage

 by Martin Belam, 16 June 2006

During April and May 2006 the BBC ran a competition to redesign their web homepage, called Accompanying the competition was a blog, which I was a guest contributor to.

My main series of posts on the blog described the way I developed a theoretical entry into the competition, and compared it with how the BBC's New Media department develops new products and services. Here on currybetdotnet, rather than duplicating the whole thing, I have gathered together links to each post, and a synopsis of the content that can be found there.

#1: The must haves
This post looks at making sure the scope and constraints of a project are understood right from the very start - in this case the elements that any BBC homepage design would be obliged to retain.

The first thing that I want to establish, as part of my 'understand' phase, are the things that absolutely have to be on my page - so I started by making a list of those. In any project it is good to get these things defined up-front, so you don't suddenly get surprised at the end when you have all the designs signed off, and then somebody pipes up with "but where is the...".

#2: Generating ideas
This post looked at the ways in which the BBC might run an idea generation workshop for a project like re-designing the homepage.

In the BBC's New Media department, at this stage of a project it would be usual to hold some sort of workshop. For a small team it might just mean booking a meeting room for an afternoon, for a larger or strategic project it might involve moving people around the country, and booking a large space in one of the BBC's buildings. Re-designing the site's homepage is pretty much as big and as strategic as they come - so if I were running the project at the BBC I would be expecting to run a large workshop for between 25 and 50 people.

#3: Research
This post looked at the kind of research that would be done by the BBC in this kind of project, whether it was running a survey on the site, doing internet leg-work, or commissioning some specific research.

What I would want done is some one-to-one interviews with people who don't currently use, and with people who do currently use but don't have it set as their homepage. I'd be looking for the research to help me understand:

* Why these users have chosen the homepage they use.
* What kind of emotional engagement have they got with that page - looking at things like loyalty, trust, or whether it surprises them or is always predictable content.
* What are the common tasks they perform when online and starting from their homepage - i.e. do they generally search, check their email, follow the stories on the page and so forth
* Why don't they use the homepage as their homepage - what features is it lacking that they want or need from their choice of homepage.

#4: Personas
In this entry on the blog, I looked at how the BBC uses the design technique of personas to help development teams focus on the audience they are trying to serve.

A real example of this from the BBC was the work done to re-launch the search service in May 2002. The service was targeted at a fictitious user called Gail (or was it Mandy?) who was a married mother of two from Bromsgrove in the Midlands. The point of using her as the main target persona was to shape a service that would suit her needs, not just what was the most fun thing for the team to build.

We knew that people like her wanted to use the internet, particularly to help their children learn, but were very wary of it because of the kind of internet scare stories they had regularly read in the tabloid press. The point was to then find the things that the BBC could offer her that would appeal to her.

#5: Feature prioritisation
Having generated some ideas, done some research, and arrived at a target audience persona, it was time to prioritise the features that would appear on the page. This entry looked at two techniques common within BBC New Media projects to do this feature prioritisation.

#6: Technical, Interaction and Visual design
Exploring the different disciplines involved in the design of a product like the BBC homepage was the theme of this post.

When people talk about 'design' on the web they very often fall into the trap of only thinking about the visual design of a service or web site. In fact, you can see some of that in some of the entries to the competition, where people have re-coloured and moved around the existing blocks of content on the BBC homepage, without necessarily looking further at the functionality or the type of information on offer.

Absolutely key in this kind of re-design project is the technical design and the interaction design. In fact, in some ways they are more important to the success of a project than the pure visual design. If you've got a nicely funky valid and semantic XHTML/CSS driven site you can experiment with different visual approaches until you strike gold. If you've fundamentally got the design of your technical architecture or your interaction patterns wrong, then no amount of window dressing will make a system perform better.

#7: User Testing
This post looked at how the BBC goes about user-testing new products and services. For major project this can often be done before a single thing has been built.

The BBC does user-testing either by putting the work out to usability testing agencies, or conducting the tests in-house. Putting the work out to agencies has a strong advantage that people don't have to visit a BBC building. This can be really important if you are testing a new service, and you want to gauge how the user reacts to a vanilla version without any preconceptions about the BBC attached to it. For testing done in-house, the BBC sometimes invites people via message boards or via newsletters - making sure it gets to test new features on the exact people who will be using them.

The key thing with any user testing is to make sure that the tasks you ask the user to perform really show whether your solutions have met their goals.

The next series of posts detailed the eight elements that I had selected for my final design, and explained the functionality and thinking behind each in turn.

#8: Blending the News
#9: Embedded iPlayer
#10: Search
#11: A-Z
#12: My Bookmarks
#13: My Conversations
#14: My Inbox
#15: BBC Today

The final two posts in the series looked at how such a project might get built within the BBC's New Media department. The first looked at software engineering methodology, particularly the BBC's use of Scrum.

#16: Scrum

Essentially it is a process where producers and designers specify a list of what they want software to do in descending order of priority. They then leave the programmers alone to get on with for the length of a scrum cycle. At the end of the cycle the programmers present back what they have developed, which should always be working software. It may not have all of the features yet, but it should be good enough to work. During the development phase this process will be reiterated until the product is ready to launch. At the beginning of each cycle the producers can re-prioritise what is important, remove features or add new features.

The last post rounded up three key areas - a final round of user-testing whilst the project nears the completion of the build phase, QA and testing, and the arranging of an excellent launch party by the project manager.

#17: User Testing, QA and Launch Party

As a general rule of thumb it is best not to party on the actual night of launch. For one thing, people are liable to be very tired and possibly quite worked up having rushed to meet deadlines, and may not be in the best of moods to enjoy a good party. Secondly, you should always give a new system time to bed down before celebrating. Otherwise, there is a risk that something could go wrong just after launch, and there is no more thankless task than trying to drag software engineers away from a party in order to get broken things fixed at short notice...

My theoretical redesign of the BBC homepage ended up looking like this:

My entry

You can see all of the entries to the competition - many of which are considerably better than my effort - at the gallery.

Keep up to date on my new blog