“We wanted no ghost to tell us that”: Acknowledging Our Expertise In Information Architecture.
We’re moving offices at The Company this week and consequently I’ve unearthed some dusty print outs of articles I meant to read. Some of these will end up in blog items in the next week or so and the first to prick my conscience has been a 2002 piece by Jesse James Garret. It was written at a time when jobs were being cut left, right and centre as the ‘new media’ industry euphemistically ‘consolidated’ itself.
With this in mind, it is quite an introspective and defensive critique of the purity of information architecture (IA), seeking to justify the role’s existence. What Jesse James Garret astutely points out is the tendency to justify every decision we make as being supported by user research and testable to within an inch of its life. Personally, I have dwelt on this as someone with a rigorous empirical background in psychology where, if it was not testable it was not truly meaningful. Taking a 'step back' (one for you, Matt) it is worth considering the role of a print editor. When they make decisions on layout, content and indeed the customer journey through their product, they are not doing this based on eight people in a room telling them what they think. They are using their insight, experience and gut reactions. They are doing what they are paid to do – to make tough strategic decisions about what they think is the right way of doing things without having the safety net of user-research to fall upon.
I wish I made more decisions like this. I wish I was able to be more bold and say, “you know what, this interaction should look like this because my years of experience tell me it should”. Too often, perhaps, we sit and listen to customers tell us exhaustively what they want and then agonise about what they meant. To go back to the wants and needs argument that Marc McNeill so effectively summarised recently, how many people would have sat in a focus group three years ago and said they would need to upload video and share it with the world? Not many, but then came YouTube … likewise, who would have thought kids would want to be creating web pages, writing daily blogs and sharing their lives with the world on an unprecedented scale? Not a huge amount but then came MySpace.
Now, I’m not saying that IAs would have foreseen YouTube or MySpace ahead of their users – if we all had that foresight IAs would be being paid a lot more – but it does demonstrate that user-research doesn’t provide all the answers.
If we dwell on making the most usable sites as defined as the sites that pass the user test then we become no better than the teacher who schools their pupils in just enough to pass the exam. The exam in this example is no more reflective of the challenges of life than the user test is reflective of the continued experience of a website by thousands upon thousands or users. The trouble is, testing and metrics deal largely with the quantifiable elements of experience, time taken, click volumes, conversion. Unfortunately these are the predominant currency of business and unless you present such information higher up the chain your credibility is called in to question.
To return to the publishing editor analogy, the editor is top of the tree. Their decision is respected not just because of the experienced opinion but because in the hierarchy there are few people above them. Information Architects on the other hand sit lower down, we don’t have that power and all too often we find our work being passed up the chain, interpreted, interrogated and ultimately ignored as the wishful thinking of an idealist. Peppering a report or proposition with research is our way of protecting our ideas, wrapping them in cotton wool to survive the journey to the senior manager’s desk.
At the end of this journey what lands on the manager’s desk is a collection of research quotes and some proposed designs. What is missing is the insight, the explanation of how we leapt from someone saying they wanted to be able to change their address to a panel in a wireframe with ‘change my address’ on it.
Fortunately the assertion by Jesse James that IA as a distinct role would fade away have not proved correct and in many organisations there exist a team of such professionals. Granted, we have had to broaden our horizons and think about technology and business at the same time but, at our core, we can still be information architects.
Nevertheless the advice still holds true. We should trust our abilities more and occasionally eschew the temptation to commission research at every turn to support our work. We must be prepared to act on hunches and still show our working. Explain and annotate why we have made these decisions but not be afraid to say it was our personal decision and not the decision of 83% of survey respondents. These hunches and gut reactions allow us to work and react faster. Take AJAX for example – we have not had the time to test and research this stuff adequately but we are under immense pressure to deploy it sooner rather than later for the UE and technological benefits. We should be brave enough to generate great interfaces using AJAX implementations based on our experience and get this stuff out there without waiting for XYZ Ltd to fling something out and observing how they get on.
With this in mind, it is quite an introspective and defensive critique of the purity of information architecture (IA), seeking to justify the role’s existence. What Jesse James Garret astutely points out is the tendency to justify every decision we make as being supported by user research and testable to within an inch of its life. Personally, I have dwelt on this as someone with a rigorous empirical background in psychology where, if it was not testable it was not truly meaningful. Taking a 'step back' (one for you, Matt) it is worth considering the role of a print editor. When they make decisions on layout, content and indeed the customer journey through their product, they are not doing this based on eight people in a room telling them what they think. They are using their insight, experience and gut reactions. They are doing what they are paid to do – to make tough strategic decisions about what they think is the right way of doing things without having the safety net of user-research to fall upon.
I wish I made more decisions like this. I wish I was able to be more bold and say, “you know what, this interaction should look like this because my years of experience tell me it should”. Too often, perhaps, we sit and listen to customers tell us exhaustively what they want and then agonise about what they meant. To go back to the wants and needs argument that Marc McNeill so effectively summarised recently, how many people would have sat in a focus group three years ago and said they would need to upload video and share it with the world? Not many, but then came YouTube … likewise, who would have thought kids would want to be creating web pages, writing daily blogs and sharing their lives with the world on an unprecedented scale? Not a huge amount but then came MySpace.
Now, I’m not saying that IAs would have foreseen YouTube or MySpace ahead of their users – if we all had that foresight IAs would be being paid a lot more – but it does demonstrate that user-research doesn’t provide all the answers.
If we dwell on making the most usable sites as defined as the sites that pass the user test then we become no better than the teacher who schools their pupils in just enough to pass the exam. The exam in this example is no more reflective of the challenges of life than the user test is reflective of the continued experience of a website by thousands upon thousands or users. The trouble is, testing and metrics deal largely with the quantifiable elements of experience, time taken, click volumes, conversion. Unfortunately these are the predominant currency of business and unless you present such information higher up the chain your credibility is called in to question.
To return to the publishing editor analogy, the editor is top of the tree. Their decision is respected not just because of the experienced opinion but because in the hierarchy there are few people above them. Information Architects on the other hand sit lower down, we don’t have that power and all too often we find our work being passed up the chain, interpreted, interrogated and ultimately ignored as the wishful thinking of an idealist. Peppering a report or proposition with research is our way of protecting our ideas, wrapping them in cotton wool to survive the journey to the senior manager’s desk.
At the end of this journey what lands on the manager’s desk is a collection of research quotes and some proposed designs. What is missing is the insight, the explanation of how we leapt from someone saying they wanted to be able to change their address to a panel in a wireframe with ‘change my address’ on it.
Fortunately the assertion by Jesse James that IA as a distinct role would fade away have not proved correct and in many organisations there exist a team of such professionals. Granted, we have had to broaden our horizons and think about technology and business at the same time but, at our core, we can still be information architects.
Nevertheless the advice still holds true. We should trust our abilities more and occasionally eschew the temptation to commission research at every turn to support our work. We must be prepared to act on hunches and still show our working. Explain and annotate why we have made these decisions but not be afraid to say it was our personal decision and not the decision of 83% of survey respondents. These hunches and gut reactions allow us to work and react faster. Take AJAX for example – we have not had the time to test and research this stuff adequately but we are under immense pressure to deploy it sooner rather than later for the UE and technological benefits. We should be brave enough to generate great interfaces using AJAX implementations based on our experience and get this stuff out there without waiting for XYZ Ltd to fling something out and observing how they get on.
In time, I hope senior management will begin to trust these hunches and – when augmented by the right amount of research – will begin to believe the value of the small collection of experts in their web teams. I’ll paraphrase for my final thoughts: Research data and formal methodologies do not guarantee better architecture. Better architects guarantee better architecture.
I would have found the answers.