How we built the BBC’s most shared web page

The BBC Class Calculator achieved six million views in just 24 hours, was the BBC’s second most viewed page of the year and, at the time, the site’s most shared page ever. So, how did a sociology study about class capture the nation’s imagination?

Creators Joe Sharpe, co-founder of Applied Works and Michael Orwell, consultant and the BBC’s former Digital Commissioning Editor for Knowledge & Learning, talk to Applied Work’s Head of Marketing Hannah Foulds about how and why it was made.

Hannah: So, to get started – what was the original research about? Who was doing it, and why?

Mike: Well, it was actually originally meant to be for a BBC documentary series. My team, BBC Lab UK, had been approached by the Current Affairs department, as they were interested in making programmes on British social class – a sort of state of the nation, informed by data. We partnered with sociologists and academics – including Professor Mike Savage from York University and Professor Fiona Devine from the University of Manchester – and designed a survey which was completed by over 160,000 people to gather data and discover new aspects of the British class system.

The academic team spent twelve months analysing the data sets, but by this point the TV commissioning round had passed, and the TV production team had seemingly lost interest, finding the early analysis a bit dry. But about three months later, seven new class clusters began to emerge from the research – we’d uncovered a brand new snapshot of Britain.

We needed to do something with these insights – I wanted to get maximum value for the time and effort the academics and my team had put into the research, so we started looking at articles and blog posts with infographics. But we realised there was an opportunity to do something a little more interesting, and brought my experience of working on the interactive elements from BBC Lab UK into the mix.

We approached Applied Works, and together came up with an idea to place the audience at the centre of the story – enabling them to share their own information and discover their own place in the social class research, based on their answers.

We built a prototype and presented it to BBC News, who were pretty enthusiastic about it and keen to run it, so we had to work quickly to build a final product. The research was no longer boring, but treated like gold dust.

“The research was no longer boring, but treated like gold dust”

Joe: The class calculator really captured the attention of the nation – quite how much caught us a little by surprise! People immediately wanted to find out what class they were in and share that on social media, or at work or with family and friends. It even made it onto Have I Got News for You! There were parodies as well – all the class tropes came out of the woodwork. Apparently a rave was organised for people in the ‘Emergent service workers’ class.

“People immediately wanted to find out what class they were in and share that on social media, or at work or with family and friends”

Hannah: What were the challenges with building a tool to use as part of a national news story? I imagine there were time constraints and a strict editorial process to go through?

Mike: Well, for a start the academic team wasn’t used to getting so much attention for their research, so I had to make sure they weren’t over-burdened by interviews. I was also worried that they would be offended by the parodies, but Professor Mike Savage found it fascinating – he even wanted to write a paper on it! He believed these methods were an important development for modern sociology.

“The academic team wasn’t used to getting so much attention for their research”

Also, BBC News needed the calculator to have a ‘peg’ – it needed to be related to something happening in the wider-world at the time. So we announced the research at The British Sociological Association conference and through the journal ‘Sociology’ – it became their most downloaded edition ever.

Joe: We had to be quite careful with how we created images of the new classes. We wanted the illustrations to be witty and represent each group, without falling into stereotypes. It was a fine line to tread but it seemed to land really well – I think because it didn’t take itself too seriously. Our designer Panja Göbel did a great job with these.

Hannah: What was the outcome? And why do you think it did so well?

Joe: We were the first external team to embed interactive content in the BBC News website. The whole news division ran with the Class Calculator, and made it a big thing. It was featured on the BBC’s homepage for a day and half, achieved six million views in 24 hours, and was the most shared BBC page ever. The share function was a big driver of traffic, with over a quarter of views coming via social platforms. It also won the data-driven applications category at the Global Editors Network Data Journalism Awards.

“It was featured on the BBC’s homepage, achieved six million views in 24 hours, and was the most shared page ever”

Mike: Everybody is included in the class system of Britain – people couldn’t help but go and have a quick look.

Joe: One of the things that made it enjoyable was that the tool visually responded to the inputs as you moved through the quiz – it allowed people to start anticipating the final result before they got there.

Mike: Yes, it wasn’t just a black box that spat out a result at the end. You could see the metrics moving. Also, regarding the UX – the way people were guided through it – Applied Works brought a lot to the party and made it extremely easy to interact with. It wasn’t fiddly… people could easily get a result within minutes.

“It wasn’t just a black box that spat out a result at the end. You could see the metrics moving”

Joe: Yeah, that was all part of the prototyping process. We tested it on people to ensure not only that the algorithm worked, but also that it was fun and easy to use. Equally, there was serious academic research behind the calculator, and it was important to get the balance right. The editorial components helped to draw people deeper into the stories behind the research. I think that’s part of the reason why it had such reach. We even had people reverse-engineering their answers to find the thresholds between the class groups.

Hannah: Joe, Applied Works also designs for organisations like Chatham House and the UN – have you applied any of the learnings from the class calculator?

Joe: Absolutely – placing the user in the heart of the story is such an effective way to add context and make it more relevant, as well as driving engagement. We worked with Mike on a number of similar tools for BBC iWonder, spanning really fascinating subjects, like personality traits, attitudes to money, online dating and ambidextrousness.

“Placing the user in the heart of the story is such an effective way to add context”

For Chatham House, we built a calculator called Tribes of Europe, as a window into a major survey and research paper published in 2017. Each of the six ‘tribes’ represent highly nuanced values and attitudes to the EU, whilst individuals within each group share similar opinions and life experiences and tend to behave similarly at elections. Again, we used illustration to depict the groups, and needed to convey attitudes to quite sensitive topics and often conflicting opinions across the spectrum.

The site launched in nine languages, supported by a lead feature on BBC news, and was widely shared across social media. With nearly 90,000 visitors in the first four weeks of launch, and 97% of visitors completing the test, Tribes of Europe became Chatham House’s most engaged piece of digital content to date. It also won a Digital Communications Award in 2018. As with the Class Calculator, I think a lot of that success was built around putting the user at the heart of the story.

Hannah: With that in mind, what factors do you think make a good data tool?

Mike: For me, I think from an editorial perspective, it’s making it user-centred – telling users something that’s relevant to them, and putting them in the picture.

It’s also important to have one that’s iterative as you use it – the user’s information changes the story, they can influence it and see the visual workings. They should also be able to have a non-linear experience – being able explore the data and choose the comparisons themselves is important.

“Users should also be able to have a non-linear experience – being able explore the data and make the comparisons themselves is important”

It’s also got to be usable, clear and obvious how you can do the things you want to do. UX and visual language is vital to make something absolutely frictionless to use. All that design ethos goes into making a great data tool.

Joe: We’re often tasked with conveying complex information in accessible ways. For me, I feel we’ve done a good job when the user isn’t conscious of, or daunted by the complexity of the information they’re absorbing. There’s a great quote by a French aviator named Antoine de Saint-Exupéry that I read in the autobiography of Yvon Chouinard – the founder of Patagonia – that really chimes with me: “Perfection is finally attained not when there is no longer anything to add but when there is no longer anything to take away, when a body has been stripped down to its nakedness.”

Hannah: What do you think of the quality of data visualisations out there at the moment?

Joe: It’s a mixed bag. We see some exceptionally good ones on a regular basis from certain outlets and individuals, but there’s a lot of mediocrity as well. The best ones are just immediately intuitive and accessible, not just beautiful. As far as possible they should make sense to non-technical audiences and not need ‘how to use’ explainers to be understood. Often it’s a lack of good editorial context or poor UX that can let them down.

Mike: I think there was a huge explosion of data visualisations around the Covid-19 pandemic because it was one of those weird situations where data was like water in a desert… it became a matter of survival. People were desperately trying to find elegant ways to show trends. I think the quality of data tools increased over that time.

Hannah: What can organisations that need more people to act on the data they hold learn from this editorial and interactive approach?

Mike: You need some great statisticians to do the analysis, and to work out what the trends and clusters are. The BBC Class Calculator was a very curated experience. If we’d just made a series of bar graphs, it would have been of little interest. It needs the story packaged up to present the findings. If you have interesting data, you need to get some analysis done and you need to work out where the stories lie within the analysis.

“If we’d just made a series of bar graphs, it would have been of little interest”

Don’t give people a complete sandpit of data, but allow them to explore within the analysis. People should be able to play around and find their own stories within it. Allowing people to form their own conclusions. The story should emerge in front of them.

Hannah: Lastly, and top tips for anyone planning on taking this approach?

Mike: If it’d been left to my editorial team to create the Class Calculator, it wouldn’t have worked – we’d never have been able to get to the bottom of it. I needed Professor Mike Savage and the statisticians’ understanding of the concepts, and Joe and the Applied Works team’s understanding of the visual design and interaction. You need a marriage of editorial expertise, UX expertise and an understanding of the data.

”You need a marriage of editorial expertise, UX expertise and an understanding of the data”

Data is useless unless you get to the analysis stage – that’s when you can start telling stories. As soon as you can show meaning from the data, that’s when it’s powerful.

If you’d like to get in touch with us to discuss a project, or express an interest in joining our team, please click here.

Subscribe to Rows and Columns to receive inspiring data visualisations from around the world and keep up to date with Applied Works’ upcoming workshops and events.