How Dazed built a better DIY analytics tool

In search of better audience analytics, some publishers are opting to build their own tools. But building custom software doesn’t have to be expensive. Just ask Dazed in the U.K.

Dazed Group has been trying to get a better understanding of Web analytics to inform its content. Recently though, the British fashion and culture outlet became frustrated with Google Analytics and its limitations. Dazed couldn’t group content together or subtract different variables — like galleries or other content types — to help editors spot patterns in user behavior. Dazed wanted something better. But rather than buying software in, it tried its hand at digital DIY.

Dazed staffers came up with the idea of using Fusion Tables, Google’s free charting tool, to create a tool. By plugging its content management system data, Google Analytics data and a social media data together in Fusion Tables, they reckoned they could get a better picture of user behavior. After a sprinkle of computer script and a few days tweaking, they arrived at a low-cost, highly flexible analytics tool, pictured below.

unnamed
Dazed’s analytics tool in Google Fusion Tables. Multiple columns show social analytics and more.

Dazed’s head of digital Matt Jones said, “We’ve got much deeper insights into our content now because we have a lot more flexibility over how we filter our data. That’s to be benefit of our editors and our advertisers. The real beauty of this though is its simplicity. Fusion Tables really easy to use, meaning everyone can get the hang of it and we can build an analytics culture. We aren’t forced to blindly publish stories any more.”

As a result of this deeper insight, it’s now investing in more long reads like this one on mental health in fashion. Dazed said it’s been equally useful in cutting down the variety of content it creates to focus on the formats that really work well. “These improved filters helped us analyze a series of articles we ran last year which didn’t work. The data said that it wasn’t working for our audience, but we gave it one more try. The article performed exactly as our model predicted it would. We’ve got to a point where we’ve now got a much better idea as to what’s going to work before we commission it,” said Jones.

Dazed isn’t the first to have built an audience analytics platform in-house. The Guardian has one, as does NPR. But building software can be a perilous endeavor. That’s a subject Dennis Publishing’s CTO Paul Lomax highlighted at Digiday’s recent Publishing Summit in Europe.

In a talk titled “How to avoid screwing up technology,” Lomax identified the common pitfalls in publishers’ technology strategies. For most publishers — those with a limited budget — his advice is: Write code as a last resort. Rent it instead. If tempted to build, don’t underestimate the resources needed to do so. When renting software, avoid temptation to commit to unwieldy enterprise software.”

Dazed is also mindful of the technology it buys, rents and builds. The company is currently building out a custom CMS which will help it roll out new content verticals much more efficiently. It’s hoping to integrate that with its data analytics tool.

But with a relatively small development team of five, it has to pick its development priorities very carefully. That’s why its analytics tool lacks a little finess. On the plus side though, having small teams gives it a level of agility it might not otherwise have.

“Our development team sit in close proximity to the editorial team. That integration means they can be having conversations [with our teams] on a daily basis. They’re closely attuned to the needs of the audience, the editors, and Dazed as a business overall, rather than being hidden away as a tech team might be in some organisations” said Jones.

https://digiday.com/?p=106408
Digiday Top Stories