Monday was a whole set of keynotes from some of the big names in user experience – the rock stars if you will. Eric Reiss from fatDUX (and author of one of the books that got me started in this stuff) talked about e-service (sometimes called service design) – thinking about all the touch points between an organisation and customers, not just websites. Luke Wroblewski brought some architectural thinking to interaction design projects in his talk titled Parti and the Design Sandwich. Dan Saffer made us think about how behaviour and function should drive the design of products and services rather than getting hooked up on form or mechanics (how often do we think/hear “we should build an ajax site for this”?). Jared Spool entertained after lunch with his take on making things intuitive, Jeffrey Veen shared insights from his work on the information design for Google Analytics and (the) Don Norman wrapped things up by reviewing the day and arguing how complexity is a good thing.

Then we had two days of smaller, more practical focused workshops. I spent Tuesday morning in Donna Spencer‘s session on Information Architecture. Billed as “No filler, no fluff, just pure IA”, it was just that. For me it was helpful in reassuring me that we’d been doing some of the right things (e.g. card sorting) but that we need to think a little more holistically. For example, you can’t just take the results of a card sort and build a site structure from it, it’s a tool to help you think about tricky parts of a site or get some initial ideas. In the afternoon, I stayed with Donna for a session called Designing for People. Again a fairly introductory session, for me this reinforced some of the things I learnt about people in HCI modules of my computer science degree and put them in the context of the web (which barely existed when I studied them the first time around). So we learnt to think about and leverage things like human memory and visual attention in the web design and development work which we do day-to-day.

Wednesday morning saw me plumping at the last minute for a session on sketching. Not something I’d have normally picked, but I’d heard such good things from those who’d attended it the day before that I decided to give it a go. The session was called Quick Sketching for Interaction Design and was run by two really nice chaps called Mark Baskinger and William Bardel. They took us back to basics with pencil and paper sketching. We drew lines, we drew squares, we drew circles, we added perspective, then we drew arrows. We brought these things together to illustrate problems and sketch process flows and design ideas to address them. My drawing was terrible – it was  pretty embarrassing having my squares corrected in red by NatBat – but I’m going to buy a sketch book and test Mark’s assertion that we can all get better with practice. The session concluded with a group sketching exercise where we explained the many pains of laundering a duvet visually (where are the pictures, Al? Not good enough for your portfolio?!). The chaps are working on a book on sketching and I’m definitely going to keep an eye out for that one coming out.

Finally on Wednesday afternoon I went along to Leisa Reichelt‘s session on design research. Leisa discussed the range of techniques which can be used to bring users into the design process and focused particularly on user interviewing. For me this was great as it’s something we’ve been trying out at White October to try to get useful user insights for little cost. It was reassuring to see that the process I’d pretty much made up for interviewing was close to that which Leisa has been practicing for ages.  I wish she’d had more time to go through the analysis side of things, but I’ve got some good pointers of what to think about and where to learn more. Leisa is a prolific live tweeter and the dumps of her tweets on her blog give some good insight into the various sessions she made it along to – well worth a scan through.

Other sessions I would like to have made it along to include Richard Rutter and James Box’s session on wireframing. I heard good stuff from developer-y folks who went along to that to learn how Clearleft use jquery and other tools to produce rapid prototypes for demonstration, discussion and testing.

Just a quick story which I think sums up my overall feelings for the event. At the beginning of Leisa’s session, she asked us to introduce and tag ourselves BarCamp style (no mean feat with 30-40 people in the workshop). The most common tag by far was ‘inspired’. We were nearing the end of a packed three days so the fact that so many were inspired rather than tired has got to be good news for the attendees, the speakers and Clearleft alike. I met lots of friendly, interesting and helpful people and learnt so much so a big thanks to all those involved. The UK UX scene is really picking up now and it’s an exciting time.

Like a lot of web apps the site has pages where the user interacts with data on a page, completing their action by clicking a button.  For example a list of products that they browse with an ‘Add product’ button at the top:

Controls at the top of the Magento product list page

Often the page that you’re looking at will be long with lots of information on it, or with lots of information for you to fill in, and as you scroll down the page you lose the controls at the top.  What Magento does is: as the controls reach the top of the page (where they would normally disappear), they detach from the page and follow you down as you scroll, staying constantly visible at the top of the page.

Magneto controls follow you down the page as you scroll down

This is a neat application of DHTML, used to make life easier and more intuitive for the user, and not just for the sake of fancy graphics.  We murmured our delight in the office when we saw it…

BUT a few weeks ago they released their new website.

I think to truly appreciate how big a step back they’ve taken with this, you needed to have seen their old website, but unfortunately I was too late to get a screenshot of this [DF - since writing they may well have rolled it back - for clarity the new design is khaki and the old design is blueish!]. However, over the weekend, I must have spent at least 10 minutes swearing at the new one.

Amongst its biggest failings is its wizards : they try to walk you through common tasks. The idea is that they make life easier for the user who is unaccustomed to the site. I think that there has been a backlash against wizards for almost a decade now, so I’m a little confused as to why Postcode Anywhere have decided to go this route. In any case most of my frustrations were vented attempting to use one of their wizards that didn’t give me the options I needed to complete my task. And there was no viable alternative to the wizard, which left me completely stuck.

It is stupid to employ a wizard which dumbs down the options to the user with a list of idiot, non-comprehensive options that second guess the user’s task, particularly when that user is a developer, more likely than anyone to want a plain, straight-forward option that doesn’t obfuscate the underlying data.

Apart from this impossible interface, there are few more problems that leapt out at me:

• Instructions in some wizards that tell the user they can avoid the wizard by clicking a non-existent link (particularly annoying)
• Account codes displayed as images so that you can’t copy and paste them (something you will certainly need to do if using the web services)
• Copy on the ‘Become a reseller’ page that tells the user that you really should think about becoming a reseller. Clearly you are, you’ve clicked on the ‘Become a reseller’ link… so how do you become a reseller?… I’m still none the wiser…

And then there’s the new colour scheme. I suppose it’s fairly trendy and the design is in keeping with all things web 2.0, but honestly, khaki and dirty orange? The old site was blue, fresh, exciting (still after 5 years), engaging and truly lovely. These new colours make me feel boxed in and oppressed.

HOWEVER dismayed I was with this website, Postcode Anywhere have displayed a truly admirable quality in starting to hold up their hands and admit they’ve got it wrong. I wrote a pretty disgruntled series of emails to them over the weekend, and got immediate response from their IT Director, Jamie Turner, acknowledging my feedback, that of others, and indicating that they’d be asking their customers for feedback this week with a view to fixing the problems or rolling it back.

Indeed today their customers got an email asking for feedback which said “You may have noticed that we’ve launched a new website recently to reflect our new branding and, to be honest we’ve had mixed feedback…”, and as I was writing this blog post I got an email from Jamie Turner telling me they have decided to roll it back. Hurray!

So, what have we learnt kids? That’s right :

1. Always, always, always test your new ideas with your users, however bright you think they are.
2. If you get it wrong, admit it, listen to your customers, and put it right.

Lesson over.