Synthesis as an analysis activity in designresearch
30 Mar 2009 - 4:20pm
4 years ago
There's Synthesis and synthesis.
You always do some synthesis along the way, you can't help it. But the Synthesis is a really important step that should be distinct from Research. This is where you are really putting the whole picture together. And checking assumptions: "Did everyone really say x? Or do I just think that?"
Finally, I find that your audience won't hang with you if you don't have a great story about what you did and what you found. And I don't mean the story is "we did this, and we think this." You need to really craft your findings to capture all the nuances that lead you to a design direction. This is Synthesis.
Thank you for the detailed view of your process. You raise one point that
I'd like to tease out: although you have a separate stage that is solely
focused on synthesis activities - Consolidation; your Research stage
includes other, smaller but no less significant, synthesis activities. Did I
understand that correctly?
Things like: "visually sketch out other relevant models (sequence, flow,
One of the things I'd like to understand better is the way in which these
smaller tasks - right across design research - are intertwined. It's that
intermingling which makes them so difficult to identify, understand and
> For me, the work is broken down a little differently. (I work at > InContext Design and so use the Contextual Design methodology created > by Holtzblatt and Beyer). Using this methodology, the process is > broken more into Research and Consolidation (or synthesis), with > analysis being part of Research. > > The Research phase consists of gathering information: we talk to the > client and other stakeholders to understand the business needs and > technical constraints, and we do Contextual Inquiry interviews with > users. As part of this Research phase we have an interpretation > session after each interview%u2014this is our analysis. We recount > the interview and capture the details that are relevant to our focus. > This includes capturing notes to later build an affinity diagram, and > visually sketch out other relevant models (sequence, flow, physical, > etc.). We do this so everyone on the team can have a shared > understanding about what happened during the interview. For me, this > analysis is just part of the research%u2014but it is separate from > synthesis as Steve initially suggested. > > After enough interviews are completed, we then consolidate each model > across all users. Using our process, we take each individual sketch > and combine them to create new consolidated sketches. This is where > the synthesis takes place and you begin to see the larger picture of > the work across all the users. > > The sketching that we do in these phases is different than the > sketching that Brad Nunnally discussed, but similar to what Dave > Malouf raised. In these phases, we use sketches to understand the > data and to share and communicate that understanding to the team and > eventually to the client. (Yes, the sketching here is synthetic, but > that's not the main purpose.) We don't sketch solutions until > consolidation is done and we have a full picture of the work across > the user population. > > Once into the design phase though, I agree wholeheartedly that > designers should be sketching their ideas. We have a saying, "If > nothing is being captured, then you are just talking in the air." > Without a shared representation, it's hard to build a shared > understanding and make a decision. Personally, I find it very > difficult to even think about design without sketching. > > I suspect that our process may be different than most. If so, I'd be > curious to hear how other processes differ in terms of research, > analysis, and synthesis. > > David Rondeau > Design Chair > Twitter: dbrondeau >