innovation – give it ten years (girly geeks london)

Microsoft T-Shirt

So, I went to the Girl Geeks Dinner in London last evening. It was an interesting night. The first thing you need to know if you’re thinking of going, is that it’s not a dinner. It’s drinks and a talk. But it’s still good.

I went there knowing absolutely no one, and ended up meeting a few people (hooray to those girls who were brave enough to introduce themselves to people they’ve never met… this happened about three times throughout the night, I did it a few times but not as bravely as some!)

One thing I’m taking away from the evening is that I need to find a way to talk about what I do that sounds as exciting as I think it is. As you do when you don’t know anyone, you find yourself explaining what you do with your time at work. You’d know by now that I’m pretty enthusiastic about my work – but I know that when I talk about it, it doesn’t have that zing. That’s something to work on.

Someone who does much better at it is Abigail Sellen. She’s been involved in amazing HCI work for ages. At the moment, she’s working with Microsoft. Abigail gave a really interesting talk to the Geeky Girls. I loved her relaxed presentation style. Abigail has been doing this work and talking about it a lot. She has such an understated approach, but her CV is so incredibly sexy, I suppose it’s easy to be understated.

Abigail says – if you’re going to *really* innovate – really do something out of the square – then be prepared for a ten year wait to see it go to market- otherwise be prepared to engage in taking it to market (getting out of the research lab and going out for lunch with product managers, engaging with the economics and the politics of the organisation outside of the research lab). She was talking about projects they were working on ten years ago that we’re looking at today and thinking ‘how sexy’. Seen that two handed desktop interaction? That kind of thing. They were working on it ten years ago and now the market is almost ready to find a place for it.

If you want to take innovation to market quickly, then focus on tweaks. Find ways to make existing technology work better. And this is no small task. Abigail gave the example of the mobile phone and the way that SMS completely revolutionised what that device meant to people and how they used it. That’s a reasonably small innovation that came to market reasonably quickly (depending on what market you’re in) and made huge changes.

At Microsoft they’ve been looking at the home technology market. Their thinking is that up until now, home technology has been divided into two areas: time saving and time wasting. This is a pretty simple breakdown, they say, and there must be some more interesting opportunities for technology in this environment – like for using it to allow people to express themselves, to emote, and for supporting families.

Really interesting stuff – enough to turn some of us green with jealousy, I’m sure. Sometimes I really like the idea of working in a research lab. But then, they too have frustrations – such as the ten year wait, and the products that are designed but never get to market, and getting IP Patents for all your ideas can’t be that much fun either.

It was definitely worth the effort to make it to Geek Girls and I’d recommend it to other London gals. Get along and check it out!

Meanwhile – check out Sarah Blow’s great t-shirt (picture above). It’s a customised XXL Mans Microsoft .NET tshirt. Microsoft has never looked so cool. Mash-up of the year I reckon :)

Technorati Tags:

Check out a non-crunchy version of the photo here.

Attack of the killer Assumptions (and how to overcome them)

assume the position

Assumptions are something we battle in kinds of ways. I know when I was doing more project management, trying to get a handle on project assumptions and documenting them was a necessary challenge. Understanding and documenting assumptions was critical to managing my client’s expectations, and making sure that it was actually possible for me to deliver a project on time and on budget.

These days, I’m more likely to make assumptions about the way that people will understand an interface and what they’ll find easy to use. Even though I continually try to train myself NOT to bring assumptions to the table when I’m designing or testing designs – or at least, to position my assumptions more as hypotheses than as a personal truth.

I often learn as much about my own inbuilt assumptions as I do about how people interact with particular interfaces… even now when we are all conscious of the new challenges created by different kinds of novel interface element, it’s a constant challenge to keep assumptions under control (which is – in my opinion – to make them conscious assumptions).

I’ve been thinking about this subject for a few years now and have asked lots of people along the way about their experiences so it was reassuring to see Kathy Sierra sum up my quandary so succinctly in her recent post on Assumptions (and their use by dates):

The really big problem is the assumptions which are so ingrained that we don’t even know they’re assumptions. They become an accepted Law of Physics, as good as gravity.

For me, assumptions are something that you usually become aware of after they’ve bitten you in the butt. Once they’re known, conscious and documented they’re not so scary… in fact, they’re not scary at all.

It’s kind of like being afraid of the dark… when you can’t see what’s under the bed, you imagine all kinds of hideous things. Once the light is on, you wonder how on earth you let your imagination run away with you so crazily.

Kathy is right – once you’ve recognised your assumptions, you can’t just leave them sitting there. You need to pull them out and re-examine them every now and then and make sure that they’re still as they should be, or update them if you need to. (Or, potentially throw them away as irrelevant).

But here’s my question – what do *you* do to try to expose these really dangerous assumptions? The ones you don’t even know that you have? How do you bring them to light and make them known and not dangerous?

Come on. Help me take out some of these killer assumptions.

:)

Image Credit: Kayaness @ Flickr

how do you analyse your user research data?

Affinity diagram?

Of course I’ve just finished a week of asking users lots of interesting questions and getting a vast amount of even more interesting information in response. On this particular project we asked quite a few people (15) lots of questions over quite a broad spread of topics. So, now I’m trying to work out what I’ve learned.Over the years, I’ve used a range of different methods for analysing data. The ‘simplest’ yet least able to be reproduced/backed up is a combination of memory and gut feel (not recommended), then there are a range of more or less physical tools from Excel Spreadsheets, to Post It Notes (which seem to be in vogue at the moment), to Mind Mapping (my current pet approach).

I like Mind Mapping because I think it’s a fairly efficient way to push the data around into sensible groups and to also keep the ‘authentic’ user voice in the mix for as long as possible. I tend to type quite a bit (especially the really interesting parts) verbatim, and I like that even though the users have started to meld together in my analysis, their voices are still there – it is quite powerful in taking me back to the conversation we were having and the context in which their statement was made… something that I think can get lost in other methods.

Mind Mapping is also a lot more space efficient! Where I’m working now (more about that soon), Affinity Diagrams using vast quantities of all different coloured PostIt Notes are very popular… to the extent that wall space is at an absolute premium :)

This is a method that I’ve really enjoyed using in the past. In particular, I think it’s a strong method to use when you are working as part of a team doing the data analysis (whether that ‘team’ is you + colleagues or you + client… both useful). Mind Maps do tend to fall down in a screaming heap where you’ve got more than one person doing the analysis.

Interestingly, the IA Wiki (where I liked to for a definition of Affinity Diagrams above), includes both Post It notes and a Team as pre-requisites for doing an Affinity Diagram… I’m not expert in terminology, but that’s not my interpretation. Anyways, that’s a tangent. (I think!)

I wouldn’t call myself a MindManagerPro power user, but I can see that there are opportunities to further streamline my process (perhaps) through integration with Excel (where I capture my raw user data) and Visio (where the design solutions are ultimately outputted). I need to explore this integration with MS Office some more (unless someone out there has and can tell me what’s worth exploring and what’s not!)

Another thing that I really like about MindMaps is that they allow you to spend quite a bit of time ‘working on’ the data and starting to make some meaningful and interesting conclusions, which you can then bring to your client, and you’re then able to really focus their minds on what problems need to be worked through, workshopped and resolved – but with all the data to hand, and organised, and illustrating/illuminating the points that you’re discussing in your workshop.

Of course, my choice of tools is also heavily influenced by the fact that I tend to do a lot more qualitative style research then quantitative (I’ve never been one for maths) – so statistical applications and graphs I approach with caution and generally a fair amount of resistance… :)

I’d be really interested to hear about what techniques you like to use for data analysis and why you use them. Or others that you’d like to try that you haven’t yet…

Come on then, share with the people :)

PhotoCredit: RR and Camera @ Flickr

To content inventory or not to content inventory? (continuing conversations)

Late night at work

Recently I suggested that starting a project with a content audit was not necessarily the best approach.

There’s been a bit of discussion around that since then, most notably over at Donna Maurer’s blog, and also as a Question of the Week at the IA Institute.

The overview of responses from the IA Institute probably give the best idea of current concensus:

The responses to this question gave a nice blend of ideas, mainly that the initial runthrough of the content at the start of a project can be thorough, but likely should not be the final, detailed audit.

Also, there is a desire to clarify the terms at work here. One person’s “content survey” is another’s “content inventory.” Or, one person’s “content inventory” is another’s “content audit.”

The responses to this question suggest the following continuum for the level of detail:

(Least detail) Content survey > content inventory > content audit (More detail)

I have to say – I think that there are plenty of projects where a content audit/inventory *is* probably a good place (or sometimes the ONLY place to start a project). The reason for my post was to make the point that this should become a de facto ‘standard’ approach to all IA projects.

As it happens (and possibly via karmatic consequences from posting what I did) I’ve had to do two content inventories since I wrote that post. In one project I did it because the client specifically requested one at the outset of the project, and in the second case it was because the content was so extensive and so poorly structured that there was no way to get a good idea of what content was involved by taking a top level survey.

I hope to not see an excel spreadsheet for at least a few weeks….

Image credit: WorkIsPlayIsWork @ Flickr

Technorati Tags: , ,