Tactics for government websites

You hear bad things about government websites. The £585 icon at the Information Commissioner’s Office. BusinessLink’s £2.15 cost per visit. The website for Birmingham City Council that cost three million pounds.

Easy targets. And never the whole story…

I ran the websites for a government agency called CABE and the truth is that it’s a difficult job to do well. The main problem is that the big decisions are often made by people who don’t use the web, so projects end up driven by whims, internal politics or even something someone saw on TV.

It took me many false starts to work out how to get things done. Now that I’m moving on, here are my top 5 tactics for running better government websites.

1. Be an editor that people want to hug

Most government sites rely on content from people who aren’t trained writers. Their work needs to be edited, which is a brutal experience for them if they’re new to the process.

You cannot alienate these people – they are the foundation of your site. So you need to explain how the editing process works, show them why your edits are necessary and be prepared for some give and take. It takes more time but it builds trust, helps them improve their own writing and creates a lasting bond.

If you don’t believe me, read the acknowledgements at the front of any book. As Frank Chimero says, ‘writers, hug your editors’.

2. Stop writing big reports

It was clear that our web strategy needed an update. We meticulously researched the issues and wrote up recommendations in beautifully formatted Word documents. After all, civil servants love Word. How could this fail?

It failed because nobody read the reports.

The breakthrough came when we presented the same data using information graphics shamelessly ripped from Information Is Beautiful. Senior managers could now see the research data visually. They understood what was going on and quickly adopted our recommendations.

3. Each one, teach one

There are people in your organisation who love technology. Harness their enthusiasm by helping them understand the issues that matter on the web.

My favourite weapons are books, particularly Don’t Make Me Think by Steve Krug and Letting Go Of The Words by Ginny Redish. Get copies, lend them out and watch how the quality of the conversations you have is transformed.

And don’t be afraid to teach! People want to learn techniques, even complicated ones, if they make their life easier. I worked with a string of events and marketing officers who now wield Excel techniques like concatenation, IF statements and VLOOKUPs with unabashed glee…

4. Don’t build it!

Avoid the temptation to build functionality from scratch at all costs. After all, if software companies struggle to make good software (they do) it’s not surprising that government wastes huge amounts on trying to do the same.

Most of what you need already exists. It’s cheap, well-supported and has incredible usability because it’s already being used by millions of people. Yes, this is about using best-of-breed web applications and services.

At CABE we needed a website (Drupal), email marketing (Campaign Monitor), event booking (Eventbrite), surveys (SurveyMonkey), custom forms (Wufoo), file transfer (Yousendit), video (Vimeo) and analytics (Google, 4Q, CrazyEgg). The savings from using existing products rather than building your own are huge – and you deliver better services for users.

5. Design your design process

Design can get very expensive because everyone loves to get involved.

My first web project went through six costly iterations because designs were sent to stakeholders without proper context. They made poor decisions because they had no information about why decisions had been taken. As soon as they were given the reasoning behind designs, their decisions improved.

Another problem is that it’s standard to ask for three different design concepts. Agencies get to bill more and stakeholders feel that they’re making big decisions, but what inevitably happens is that the starting points get merged into one horrible mess. Instead, insist on a single design route and then iterate.

Finally, it can help to leave deliberate mistakes in the text and images used in design concepts. This is because some managers always feel the need to change something in the design to justify their pay grade – and this way they get to make a change without destroying the fundamental design.

And finally…play politics

This isn’t so much a tactic as a statement. In large hierarchical organisations it matters who you know and what you know about how they operate.

If you want to change anything, you have to play politics – whether it’s doing people favours, making people feel good about their work or knowing how to phrase a request in just the right way. It’s not evil, it’s not manipulative and you don’t have to be nasty. But to get things done you need to get involved.

Despite the challenges, running the CABE sites was an incredible experience. Although being made redundant by an inept coalition was less enjoyable.

Now I’m off to the private sector – how many of these issues will I find there?

I’d love to know what you think about this – you can reach me on @myddelton. I want to thank the government web people who inspired me, particuarly Steph Gray, Dave Briggs, Simon Dickson and Neil Williams. And last but never least, Andrew Travers, who showed me the way when I knew nothing…