There are jobs to be done, but where are the personas?
Sam Hall, Transformation Lead, Office for National Statistics
For a few years at ONS we have been talking about personas, specifically the people that use our website. Personas are ‘synthesized’ people that reflect the user types that use a web service. We develop each persona based on interviews, data and analysis of users’ behaviours and attitudes. Segmenting our users like this helps us to understand their needs, and hopefully build something that will help them achieve their goals.
Lots of work goes into creating these personas and they are reviewed and refreshed to make sure they are still valid as user needs change and products develop. There’s a really great blog post from Alison, one of our user researchers, explaining how this refresh was done for the ONS website personas at the end of 2016.
For the GSS Discovery work we have made the decision not to develop personas. There are lots of reasons for this but the main one is that there would simply be too many of them – after all we’re talking about all GSS statistical data! We could use our limited time and resource creating GSS personas and still find that we’re unable to clearly identify their needs as soon as they ‘move’ between departments, channels and tasks. The task is simply too big to complete meaningfully for this project.
Our next step is to creating a way to bring together all trade data into a “family exemplar”. This will allow us to see all of trade data available, spot gaps and overlaps, and ensure that users won’t have to search through a range of sites and sources to get what they need. To do this, instead of personas, we’ll be using jobs-to-be-done.
Users don’t want access to data because of the type of person they are (a journalist or an academic), they want it because they have a job to do. That’s a simplistic statement to make, but it holds true, and doesn’t need a persona to help us better understand a user’s motivation. There’s a famous quote by Theodore Levitt, an American Economist and editor of Harvard Business Review, who said:
“People don’t want to buy a quarter-inch drill, they want a quarter-inch hole.”
If you keep that in mind then jobs-to-be-done makes much more sense. A job is a task that the user needs to accomplish in a given set of circumstances. We will focus on the job and the circumstances that drive users to undertake it, not the characteristics of the user.
To help us understand more about the jobs that need to doing, we can use a device like the one below which is a ‘Value Proposition Canvas’. It’s actually a really simple tool. Working in the right-hand circle we can add in the jobs users need to do, the benefit they gain by doing them and the pain they might feel while they try to do them. The left-hand square is filled with the things we can do to make the job easier, to add value and to essentially develop the right product or service.
We can populate this canvas for all and any type of user, and intend to complete one with our internal GSS stakeholders, and one with our external users. Together we should have a pretty comprehensive list of jobs-to-be-done and can begin to start developing our product.
Next week we will be bringing together people from across the GSS who are involved in producing and using trade data: our Trade Consortium. As part of our workshop we will produce 2 of these canvases, one for internal user jobs and one for external user jobs. As we work on the trade exemplar these jobs will be invaluable, but what they will also do is form a basis of jobs-to-be-done which we can share with the wider GSS.
For those of you who have wanted to produce personas but have lacked the resources to invest in developing them, it could be that jobs-to-be-done might be a good place to start.