After Chris Khalil's enthnography session it was time for Matt Balara (@MattBalara and MattBalara.com) and the third session of the day. And this is where things began to heat up a bit. Matt's presentation was on More, Better Faster! Agile designs for fun and profit.
(Update: this presentation is now on Slideshare).
This was a great session, very practical. "Agile is the buzzword dejour" said Matt's session description, and from the off it really captured people's attention - maybe it was the video holding slide, and even the flame fx midway through - certainly the 'don't fetishise your deliverables' slide caught peoples' attention (hint: there was leather rubber involved).
So... key elements of an agile design process include flexibility, allowing a design to emerge and develop, not necessarily sticking directly to 'the plan' (because most of the time things do not go to 'the plan'), redesigning, improving and reiterating your designs over and over as you develop them and a whole heap more.
There's a good overview of Agile Design here: http://www.agilemodeling.com/essays/agileDesign.htm and actually another good summary of Matt's presentation over on Rook in the Garrett.
Matt also focused in part on a major web project he did with Ecco, as lead of a 40+ team of designers redesigning a site containing over 1500 products. The cost was in the millions of Euros and the developemtn shots, before and after etc. were great to see. Within it, there were a few things that really stood out for me:
- Ensure you have good communication with your client and you know what you're both working towards. Sounds obvious, but to hear Matt's experience of a client and an agency that were seemingly working to different ideas with the homepage was interesting to hear. He and his team got back on track with some impromptu workshops and many design iterations thereafter
- Be flexible in your process. Recognise your constraints, embrace them, and don't be afraid to shake things up.
- Sketch it, get the ideas out of your head: Sketching is a process. If you think something might work, sketch it out, try it out, push through the bad or confused ideas to get to the good ones (iterate and reiterate). My only problem with sketching is that I can barely write my name eligibly with a pen, let alone draw anything of any value. Over the course of the conference quite a few people also said their drawing skills were poor, but were keen to get over this hump.
- Build things, get fast results. Matt said wireframes, photoshop mockups - they're all things you do before you give the designs to other people to build. Why not actually build the thing, build a prototype site with clickable buttons, and go out and test it (even do Guerrilla-style usability testing with… a Geurrilla (a great live testing app going by the name of Silverback). "Go and sit in a cafe and do random live user testing with real people!". I love that idea.
This session was eye-opening for me for other reasons too. I came out of it thinking, "We do several things like that already".
For example, on one project I'm in at the moment we're ditching a couple of sections of paper- and card-based testing in favour of an on-platform html prototype. It's going to let us claw back some time, and get some good results faster. It may not be as 100.00.000% comprehensive, but it will definitely be enough together with all the other research we've done (more than we'd planned), and we need to meet our deadline and the freed time is going to be useful elsewhere. Is that a little bit agile? Perhaps.
Another thing Matt said was that not enough agencies have the balls to say to customers, "Do you know what you're asking me/us to do? I don't think you do."
While not in those words exactly, more than a few times we've questioned projects that have been presented to us and we've delayed or lost the chance of work because of it. From the perspective of doing things properly, that's no bad thing (of course, we also frequently help organisations get a much better project scope and understanding of what needs to happen.. ;-))
Lastly, Matt was a class presenter. It's fair to predict he'll be doing more presentations at future UX related events. This session could have gone on for twice the time and it still would have been packed by the end.
This was the type of presentation I went to UX Australia for.
Next up, some 10-minute talk summaries, or click to see all related UX Australia posts and reviews
