26 April 2018, Elizabeth Cochrane
“It’s been an excellent collaboration. I think we’ve learnt a lot from each other and really valued the different perspectives and culture that Golant Media Ventures have brought to the whole project.” Leo Sharrock, Director of Data Platforms
A user-centred approach is often the only way to unpick a knotty problem. We have been supporting The Audience Agency (TAA) to solve the mystery of why their users weren’t accessing TAA services – and all the invaluable data available through them – in the way they might.
Data is crucial to understanding audiences, and to expanding them. TAA are funded by Arts Council England (ACE) to collect data across institutions in order to provide a consistent view on how people engage with arts and culture. The quid pro quo for each organisation providing that data is that they can access information about their own audiences – and audiences across their particular sector – for more effective planning, programming and marketing. TAA became aware, however, that people weren’t engaging with these services as extensively as they could be. It looked as though those users might just hate data.
Golant Media Ventures (GMV) was brought in to investigate, turning a fresh pair of eyes on the situation and aiming to develop a prototype tool which might help TAA’s users. Having supported TAA to access funding offered by Innovate UK, we set about getting to the root of the issue, using a mixture of quantitative and qualitative research within the framework of a service design approach.
‘I love data, but I just don’t have enough time in my life to do what I want with it.’
An e-survey was conducted to explore what TAA’s service users themselves might perceive the problem to be and GMV’s Emma Cornes followed this up with a context study, ensuring that any innovations would reflect the real relationship users have with data in their working lives. Almost 600 people replied to the survey, providing a robust set of results which were highly revealing. For example, despite the many specialised tools available, the most-used programme, by an astonishing degree (92%), is still Microsoft Excel. Although the sample was very diverse, a remarkable consistency of themes emerged – and they did not reveal a large set of ‘data haters’.
Emma’s observations of participants engaging with data within their own working environments backed this up. TAA’s users know why data is important, it seems, but because they lack confidence in handling it and talking about it, they believe they lack the time to engage with it. Remarkably, this lack of confidence and fluency applies even to the largest of the organisations receiving strategic funding from ACE, the National Portfolio Organisations (NPOs). It’s another illustration of the well-worn truism that data is only useful to the extent that people recognise it as the answer to their questions.
‘Often when I’m using a search engine, I don’t know what to put in. I want something, but I’m not quite sure what I want yet.’
Over half of the research participants made it clear that they would access and use data more if the interface was more user-friendly. One of the clear solutions was to change the language used, so that data no longer feels like the province of specialists. For example, while ‘data’ can be an off-putting term for some, the synonym ‘information’ is not – and makes more sense to many. Crucially, respondents said they’d like to be able to ask a question in their own way and get back data results framed as the answer to that question, reducing the amount of interpretation required.
These findings – which are rich enough for GMV and TAAS to produce a stand-alone report – really shifted TAA’s understanding of what the problem was for their users. Leo Sharrock, TAA’s Director of Data Platforms, says that taking a service design approach “made us re-evaluate what we were doing, who for, and what the needs were compared to what we thought the needs were”. This meant that the product prototyping sessions were quite different from what had been envisaged at the start of the process. Rather than designing tools to serve up data in a more user-friendly way, the goal had become helping users to work out how they might want to pose their questions.
‘It could start by giving you some prompts, is it this kind of thing or that kind of thing? Is it more like this, this or this? Sometimes an empty box is difficult to fill.’
GMV’s Digital Transformation specialist John Denton led a series of sessions with participants from across TAA’s various user groups, aiming to prototype possible design solutions to the problems identified. First they created specific user-personas which participants embellished to create a range of people (imagined as marketing managers, fundraisers, developers and so on) who might be envisaged interacting with TAA’s systems, from the type who are totally data savvy to those who only use data once a year. For each, they imagined the user journeys. John then modelled various lo-fi prototypes, including drop-down menus, search-bar style features and FAQs, to see whether these made the enquiries more appealing to users, and then developed these progressively in line with feedback from participants.
These sessions were a rewarding experience for participants, who were generally delighted by the opportunity to discuss data in a wider forum with their peers from across the arts and cultural sector, especially now that “GDPR has brought data to the forefront of all our consciousnesses,” as Jen Bartle of Rich Mix says. Jen is a self-professed data-lover who was simply “overwhelmed by the complexities of data in an organisation that has multiple facets”. She says it’s crucial for everyone to become more data-confident, because “we all need to own our responsibilities for how we manage and work with the data, [so as to] maximise the opportunity that it presents”.
‘This would be amazing – it would save me having to pull so much data. You’ve saved three days of work.’
The collaborative design process resulted in a prototype with certain qualities identified as crucial. The ideal product will allow the user to pose questions in natural language, but also provide prompts and categories to help with posing the question, and then deliver a response in the same form as the question. Ultimately, the technical solution being developed focuses on the tasks and requirements of the users rather than being shaped by the data itself.
GMV’s development team have been refining the prototyped solutions to give TAA a better idea of what they might be able to offer their users and how this may impact on their data infrastructure to support it. TAA’s COO Cimeon Ellerton is very optimistic about the outcome. He says that moving away from dashboards and more towards natural language processing “will allow us to take our next leap in terms of service and UX (user-experience) design for our data tools ... to lead the user to the information they need, and not expect them to be able to find it themselves”.
There are already take-home benefits from this research for TAA, who are taking some of the learnings from the project and applying them to their services and communications now. They have collated a list of questions commonly asked by their service-users, to which they can provide the answers in a straightforward way. In addition, says Leo Sharrock, they’re “removing all assumptions and unnecessary jargon from the way that we talk about our tools, and explaining much more what we’re trying to do” – taking a user-friendly approach to a more fundamental level than before.
Participants in the project will also receive feedback on what they’ve helped to produce, having been so instrumental in designing the solution to TAA’s conundrum. Leo Sharrock says: “By going through the prototyping, we’ve discovered the challenge from users is about helping them to know what questions they could or should ask from the data, and then having the data serve up something easily understandable, with the possibility to delve into it further at a granular data [level.]” The service design approach also means that TAA now know how different their users’ experience of extracting data has been from what they imagined, and will have impact beyond the immediate outcomes of this project. “I think it will help us to develop and position our products much more responsively to the needs of the sector,” says Leo Sharrock.
By actually talking to and observing their users in action, TAA discovered that most people who seem to hate data actually just need support in accessing and using it. It turns out there really is no substitute for engaging closely with your users, and involving them wherever possible in the co-creation of tools intended for their benefit.
Meanwhile, we are broadening out the findings from the prototyping sessions to help TAA and other cultural industry specialists make themselves more future-focussed. Voice-enabled services like Microsoft’s Cortana or Amazon’s Alexa or customer-service chatbots are changing expectations around digital assistance, and while some sectors are already strongly engaged with these emerging technologies, there is much that can be done to help arts, cultural organisations and creative businesses to take advantage of these developments.