I was at the alphagov pre-launch event last week as a representative of the ‘social media types who work with local govenment’ category (i think!) – very interesting!!
For those of you who don’t know – alphagov is the project answer to Martha Lane Fox’s entirely sensible suggestion that you need one government website and not hundreds.
As you’ll see below this is a real alpha release and is in no way complete – and for those of us who have been saying that government should get more agile and release early and get feedback this is a great step.
The project has actually taken the brief in a slightly different direction and rather than create another leviathan of a site they have looked at creating a set of tools that can sit on top of content and provide a single interface. The overall approach is spot on – they have made the whole thing about search and about taking people straight to the content that people want. You don’t want a relationship with a government website, you want to get in and get out with the minimum feeling of contamination from the experience. There are some brilliant principles in play here as they design for the majority case and not the edge (example given was driving licences and the fact it’s crazy to pretend that a specialist hgv licence should be treated the same as ‘my first driving test’ given the differences in volumes of requests).
There is also an attempt to clean up some of the madly complex language that information is presented within and to use forms rather than complex explanations (for example calculating sick pay based on dates and offering a sample answer, rather than making you read through and figure it out). You can see an example below:
They have also made the bold, but I think entirely practical decision to deal with accessibility once they have a clear view of user behaviour rather than making it part of the alpha release. This is a decision that will be debated a great deal, but at least they have been clear and up-front about what they have achieved in this area. I also salute their rather clear decision to ignore the existence of IE6 altogether – I wish more people would do the same.
The team has been working quickly and agilely to get an alpha build together – and they have done an excellent job. If anything, parts of the site look more beta than alpha, which is a credit to whoever is doing design (though may, of course, bite them on the arse when people think it’s finished – you really can’t win!!!). I think we should also really appreciate the fact that they are releasing an alpha at all, given government’s obsession with the launch and the (false) idea that you can completely finish a website before launching it. It offers hope that government can ‘do digital’ differently and, before we all wade in to explain how we would do it better, I think we should all be applauding the fact that they have got this far with a very different project approach.
So: no criticisms, but a few observations and suggestions at this point.
- They are going to get a lot of flak – some will be the usual waste of government stuff, some will be differences of opinion and some will be useful suggestions. I think they need to have this debate in public and not do the government huddle behind closed doors. I think it was quite right to keep things private until this point, but they need to be open and welcoming of feedback for this next stage – and then hide themselves away again for the next build phase. Don’t fiddle with the alpha – put time into explaining thinking (and not getting bogged down in defending it), getting feedback and talking to people at all different levels of technical knowledge, experience and influence. Do some usability testing with real people and then go back to the room for another couple of months, synthesise it and do your next version. A user forum would work well so they can have the debate largely on their own turf.
- I personally don’t think you need to segment the audience. Instead, I think you should be returning choices where audience matters, but in most cases you should offer up the simplest information – and provide a clear route to drill down and refine the search. And I’m not sure it helps to use different language for the business community – you should use the simplest language you can, no matter who the audience is.
- I think my greatest concern is the faith that the approach shows in the ability to predict what people want from search terms – and to predict what audience they are from. The concern is that by trying to make things easier for people you frustrate them enormously when you get it wrong. I think I am saying that if the tech is not perfect tell people and involve them in the solution. Ask them if you have the right results and then offer them choices, both in terms of your best guess, but also the chance to refine the terms. I find this confidence in search a bit (worryingly) technologically evangelical – but this is something that should get resolved in user testing.
- It was interesting to see the approach to departmental websites and grouping themes like ‘policy’ or ‘data’ across departments – rather than in the drill down. Central government spends way too much on their basic web presence and I think this approach is spot on – though it will clearly need some very strong leadership to rip these different websites (and their budgets) out of individual department hands.
- Really good approach to presenting policies – and design head-on with the fact that ‘whose policy is it?’ may not have an obvious answer. Dealing with the grey areas is the most difficult part of any project like this and the team has made lots of really sensible decisions about this kind of stuff.
My last point is from the perspective of someone who is focused on trying to get people involved in decision making. I think the site should make it clearer who is accountable for the bit of government you have found. I’m not talking about a complaints process, which as Will Perrin pointed out quite rightly you may need at a transaction level, but an opportunity to react to bits of government policy that you understand but don’t agree with. If I have one criticism of the project, it’s the fact that it seems to have a customer, not a citizen view of the user. This is an expedient and practical decision as with many interactions we are customers and given timescales I can see why it’s this way. But I wouldn’t want this model to be embedded as I think this is another important aspect of doing things differently.
At the point where you end up at a dead end within ‘The System’ and you think the process is just wrong then you should have the option to do something about it – otherwise your next click is to post a cross update on your Facebook page. Give people a ‘I want to change this’ button and then offer them the chance to email their MP if it’s personal or make a suggestion if it’s process change. You could point them to related consultations and you could tell them how things could change. This could be an elaborate tease like the futile experience of complaining to a mobile phone company or it could point out to people that ultimately, as citizens, we have a little bit of responsibility for the way things are.
I often go on about the need for government to (urgently) become more agile. So, big congratulations to alphagov for getting government to do that and quickly releasing a really good alpha version of a complex proposition – and then asking people what they think. It’s not perfect and it’s not finished – and that is progress. Looking forward to seeing what happens next.