Al Power

Back in 2009 I attended the user experience conference UXLondon organised by Clearleft. The format was simple (as you would expect from a UX conference!) with a solid day of keynote talks, followed by two days of half day workshops. I didn’t note all talks, and any errors are likely my notes - but a few highlights from them are below: Eric Reiss talked about e-Service, focusing on companies such as eBay, wine.com, and BA, who all came in for a grilling.

His main point was that service was completely about the user experience, and that it touched on all the points of customer contact, whether air, ground, correspondence or cyberspace. He demonstrated that unhappy customers are dangerous (using himself as the case in point), saying that a happy person tells 3 people about their experience, and an unhappy one about 17. Service happens at the moment of experience - it’s a perception and cannot be stored.

More often than not companies try and measure service through the use of wonky metrics - “we have 83% satisfaction rating!” - not ideally good!, and “we fly on time!” - well duh! He also touched on the fact that expectations are directly related to levels of satisfaction - you would expect less of a low cost airline while expecting a “premium” labeled one like BA to have higher standards. He closed by saying that there are really 3 types of service - Helping, Enhancing and Fixing, and even if a company does something wrong service-wise, if they fix it in a wonderful way the customer can still come out of the experience happy.

Luke Wroblewski - the man responsible for some of the redesigns of the Yahoo.com homepage, talked about the design process relating to the architectural concept of the parti, the central idea or concept behind what you are trying to achieve. By using this high level concept to evaluate decisions, they were able to cut through a lot of the hard decisions when so many stakeholders were involved. He also mentioned how Windows 7 is seen as much more of a success due to it similarly following design principles, and how bad it was to patch to try and fix a bad idea.

Dan Saffer then talked about behaviour driven design - “Designing from the inside out - behaviour as the engine of product design”. A reason so many products don’t get it right is that people often think that “the interface is the product” - its easier to focus on form, colour, size than behaviour, much as its easier to focus on the mechanics (technology,CSS, AJAX etc) Products built around behaviour are usually the best, being aesthetically pleasing, ‘plug and play’, offering clear instructions and ‘moments of delight’ to the users, respecting their time and effort. All these things are achievable if we focus on the users behaviour. But if the ‘interface is the product’ how do we focus on the behaviour?

#1 Behaviour as a design strategy - behaviour can be the product differentiator. Behaviour is one defence against featur-itis. People and companies love features, but its a poor long term strategy as it can be easily copied, but it’s much harder to copy how features behave.

#2 Behaviour as design research - Dan talked about the (mis)use of personas in design. There are three things to focus on when doing design research - Motivations, Expectations and Actions, and if you don’t look at the best and the worst of these when creating design personas then you are likely to veer off path.

#3 Behaviour on product structure - when using a product, feedback and transitions are vital. Whats the core set of activities? The “hero Task” - what’s the most important thing the user has to get done, and does your product focus on this? E.g. a radio - the volume knob is usually most prominent. What behaviour do you want to discourage/encourage? Its hard to change learned behaviour.

One of the most interesting points Dan made was that you should think about metaphor when thinking about what you are trying to do - the product as an action. Looking at the two briefs - “design me a shower” or “Design me something to clean a person” - its obvious that the second one will be more focused on behaviour. He ended with talking about functional cartography, looking at whether controls should be digital/physical, and in what priority/platform, and the notion that you can tinker with things to make them better.

Jared Spool then came on to talk about “what makes a design intuitive”, with his blend of ‘usability stand up’ - he sure knows how to entertain the crowd while educating - one of my favourite speakers. He talked about how intuitive design is personal - based on past experience. He mapped out a concept of one’s current knowledge and target knowledge, and the ‘knowledge gap’ in between, and to have a successful product one must design to ensure this gap is as small as possible.

If a product allows us to learn things as we need them, and then lets us forget these things, it’s pretty intuitive. He then closed by talking about tools you can use to make things intuitive, such as field studies to identify current knowledge, and usability studies to help identify target knowledge and the gap. If this can be captured in personas for the team, then all the better. To make this intuitive you have to reduce the target knowledge, and provide user clues.

Jeff Veen came on with a highly visual presentation on ‘designing our way through data’. His main idea was that

“statistics can be anaesthetising, and that as designers we need to find a story in statistics and communicate it. We need to create tools to allow users to manipulate data, and by providing filters to enable clarity, we give users the capabilities to discover.”

Jeff Veen: Video - Designing for Big Data is here (in summation from the Web2.0 expo): Don Norman, the ‘don’ of UX design, closed things out in true Sensai style by talking about complexity, and how we look at it wrong. The problem is not complexity but understanding.

“Complexity is good, but complicated is bad.”

Life and the things we do in it is complicated, so we do need systems that portray that, and people often get hung up on trying to simplify things as much as possible, which is wrong. I thought his talk was pretty good, but he did lift a lot of slides from the other speakers (in a summation of the day) so I found what he was talking about to be a little to abstract to be of practical value (and he did confuse people with talk of faucets which to those not familiar was a little confusing).

He did however plot a neat X/Y axis graph of Difficulty going up against Skill going across, and then mapped out three diagonal zones of frustration at the top, boredom at the bottom, with a flow zone in the middle. This neatly highlighted how we have to cater for all users, as as one’s skill level increases with a product, we are able to cope with more difficult things, and we have to cater to both of these users to keep them in this diagonal middle ‘flow zone’ (the most productive) as their skill increases.

A nice evening was spent out at a Japanese restaurant with Andy Budd, some delegates and some of the Clearleft people talking shop, before knuckling down to the half-day workshops on days two and three. Most of the workshop speakers had initially been under the impression that they were giving a whole day workshop (ClearLeft -take note for next year!) and so ran through things at break neck speed, which was fine for me (and I loved, as often these things can go too slowly), but I know that from general discussion attendees from all workshops who were not native English speakers found it hard to keep up at times.

I started out day two with “Information Architecture - just the essentials” with Donna Spencer, who was excellent, and presented in a very no-nonsense practical Aussie style. We started out with some group exercises which led into classification schemes nicely. We covered a high level of IA, then delved into categories and category definition, organisation schemes and classification schemes, card sorting and analysis, and then finally designing navigation.

In the afternoon I attended ‘Design Strategy and Planning Tools’ workshop with Peter Merholz from Adaptive Path, who talked about the whole product design process, from Focus (what’s worth doing) -> Definition (what are we creating) -> Customer Value (what value does it provide) -> Scope (and how we deliver it), and how we can work with the customer at all stages to ensure both designer and customer are working towards the same goals. We did a lot of team planning exercises, and there was a big focus on scope with regards to how a company differentiates itself from competitors.

The prize for the most colourful handouts goes to Peter without a doubt, and anyone who can deliver a highly valuable and useful workshop with notes featuring both lots of ducks and MC Hammer gets my vote. Day three was a very much hands on experience for me, with ‘Quick Sketching for Interaction Design’ from Mark Baskinger (from Carnegie Mellon School of Design) & William Bardel in the morning, and ‘Wireframing Web 2.0 for Design and Definition’ with Richard Rutter & James Box (Clearleft) in the afternoon.

‘Quick Sketching for Interaction Design’ focused on design sketching, and how one can use sketches to both develop and present ideas effectively. It started off as a design sketching 101, and swiftly progressed to more advanced concepts like drawing people and industrial sketching techniques and tips, and by the end of the day we had produced some pretty neat storyboards, considering our skill level a few hours earlier.

‘Wireframing Web 2.0 for Design and Definition’ was very useful, as it both covered paper prototyping (which I am familiar with) and then High-Fidelity prototyping (which I had previously dismissed as too much work). We looked at a few example sketches out there (notably Soxiam from Vimeo), and some design patterns and anti-patterns. Digression - Soxiam UX sketches in more detail: The HiFi Prototypes were most impressive, using PolyPage and JQuery to maximum effect to simulate server interaction, while also reducing the need to create 1000’s of different screens to show different states. Here’s an example of poly page in use - a demonstrative fictional site for those who love pie (a group severely under-represented on the web IMHO :) ), by the chaps who gave the workshop. Some key messages that came out of the session were that:

“we need to be aware of accessibility (but you can use fancier less accessible stuff on closed/defined systems if need be) and that functional specifications should ask not dictate.”

Hi-Fi Prototyping looks like it can be really useful if used for usability testing before development starts in earnest. The day rounded up with some networking/drinks in the bar, before heading back to Oxford. All in all this was one of the most interesting and useful conferences I have been to, and look forward to attending more if I get the chance, as compared to other conferences the quality of talks/workshops was very high, and I got a lot out of it practically.

Big thanks to the ClearLeft chaps for running it so smoothly!