A Manifesto For Experience Design
For most of my working life I’ve worked in some way or another on the productivity of knowledge workers, people who do the work that’s difficult to encapsulate in process and/or automate. Knowledge work is hugely important and with some colleagues at my current employer I’ve developed an approach to analysing their needs which we call Experience Design. In our framework for Experience Design I’ve always envisaged that it would be most powerful when applied to very specific high value roles, but I’ve never had an opportunity to do that, until now.
Over the last couple of months I’ve been fortunate enough to coach a team looking into our support business. This is a complex business with several thousand people undertaking a wide variety of support tasks. I’ve had the chance to see how these people are managed, measured, analysed and assisted with knowledge and tools, it’s been a fascinating and enlightening experience. This Manifesto is a ‘watered down’ view of the way forward for this activity, which is transitioning from it’s analysis phase towards action.
What’s amazed me in this work is that to a large extent when we think of task workers we don’t start with the people, even though the people are 90% of the cost of running the business. The people are treated as an extension of a tool, as a step in the process, the tool rules, the people are driven by the tool. We spend many millions of dollars on the tools, there’s little left for the people. We spend millions more analysing the process, but very little understanding the people. I think that needs to change, the tools are there to support the people and the people do the activity.
People are not like cogs. They are not identical interchangeable components, if we treat them like that we are missing an opportunity.
Let’s step back a bit and consider the future of this type of work. What we know for sure is that we are living in the second machine age. The first machine age automated physical work, the second is automating mental work. The work that’s left is going to be the work that’s too complex, too interpersonal, too unpredictable to automate, or it’s going to be dealing with exceptions that the automation failed to handle, or bugs in the automation. This is the work we should be preparing for today, because if we don’t design for it today we won’t have time to learn how to do it well. All of our competitors will have the same automation, so our competitive advantage will derive from how well we do the work that can’t be automated.
This work that’s left is going to look at a lot more like knowledge work. Its going to be dealing with exceptions, applying knowledge to diagnose complex failure modes, dealing with people, bridging the gaps between customers and automation that they don’t understand, capturing new knowledge, analysing. Of course tooling/automation is still going to be important, but we need to start putting people first.
Lets take a look at people, because in that first phrase we have already made a mistake. People are not the same, any psychological profiling test will reveal significant differences in IQ, memory and personality between people doing the same task. So in our world of standard processes and tools we need to provide for customisations that mitigate or take advantage of those differences.
For example how do we help people with poor memories, reward the people who don’t solve many tickets, because they are detail oriented and spend too much time in analysis or knowledge capture
Let’s take a look at some of the areas that we need to examine is we undertake a role based experience design assessment:
What do they do all day, not just what tools do they use, but how much time do they spend waiting, capturing information from users, searching for solutions, going down blind alleys, capturing data etc. What’s their real experience like, how frustrated are they, how engaged are they. How do they feel during the day, when do they get tired, do they suffer from eye strain, how well do they juggle all the different applications they use. In this analysis it important to not just work with averages which hide most of the interesting information. We want to see the distribution curves, we want to see the best, the worst, the long tails.
We invest a huge amount of money in capturing enterprise knowledge that we surface through systems, but what knowledge do the people doing the job capture. Do they have their own notebooks, their own post it notes of quick reference information. Are they frustrated that they have no way of capturing information that they frequently forget or have to endlessly re-enter.
For too long metrics have been used as a tool for management to force conformance to standard levels of performance. As work becomes more complex these standard levels of performance have less meaning. Maybe metrics should be designed to help the people do a better job, rather than as a stick to ‘beat’ them with. In the same way a a fitbit can be used to modify a persons behaviour to encourage more exercise, well designed metrics can help people understand their work performance better and improve it. They can be used as a way to help managers and employees discuss their performance rather than as a way to judge it.
Individuals working on similar tasks for the same customer, or the same service for multiple customers are part of a team. How do we facilitate the enhanced performance of that team. Help the team to share information, help people in the team to support each other, help the team to discuss performance improvements, to share their concerns and problems
We need to take a more holistic approach to the tools that people use, considering all of the applications a person uses, the screen real-estate that those applications consume, whether a big screen or multiple screens are needed. We need to consider how people interface with these applications do they use a keyboard all the time, or a mix of voice, touch and keyboard. Even if a keyboard is the most effective would touch add some novelty/variety to a task, or allow them to change position, stretch their arms.
How many context switches do people have to undertake, moving between applications, moving between users, moving between different types of problems. Context switches are cognitively expensive, so how do we minimise them by for example by providing enough screen space to maintain enhanced peripheral awareness.
How do we design the telephone, headset, keyboard, desktop, screens, applications, virtual desktops etc. to work as an integrated system, rather than individual components that the user has to figure out how to integrate.
How do we extend the users personal environment, to a team environment, we might use team wall mounted screens, a team dashboard, team messaging, a team ‘stock ticker’.
How do we design the environment to optimise the performance of the team, desk space, distracting noise, therapeutic sound, lighting, heat, fresh air, views, plants.
How do we allow the people to connect with a sense of purpose, see the progress they are making, the impact they have on their customers lives. How do we allow the people to develop a sense of mastery, to demonstrate that mastery, to enhance it over time.
How do we provide people and teams with a degree of autonomy in the work they do and how they do it. Giving them a degree of control over how they prioritise work, how they distribute work across the team, balance the need to do what’s important today, with capturing knowledge for the future. Can we give them some control over investments that are made to investigate root causes and eliminate problems. Reduce the degree to which people are driven by some workflow automation system in the sky, over which they have no control or influence, that judges everything they do, without understanding any of it.
How do we capture the frustrations that people live with, can we give them some ability to prioritise these frustrations, how do we give them feedback on how these frustrations will be eliminated.