6 ways metrics have helped transform Birmingham Live (and the other way round, too)
Anna Jeys, executive editor at Birmingham Live, has been at the forefront of Reach PLC’s pilot of a new style of newsroom which aims to be digital first — but more importantly audience first too. In practice, that means everyone spending more time making audience analytics work for their audiences. At the Behind Local News conference, Anna shared her six learnings so far:
When we launched Birmingham Live, it was about much more than a change of name and a change of colour on the website.
As part of this, we also split the newsroom into a number of core functions and on any given day we now have a live team covering live news, a patch team bringing issues from patches to the fore, and a trending team, responding to content which is appealing to readers online at that moment.
One thing we really wanted reporters to work on were passions or patches, areas of content that could be geographical or interest based, but which above all had the potential to engage a loyal audience.
We wanted to encourage reporters to build their own audiences and we were going to give them the tools to do that. Regardless of which team they were sitting in on any day, it’s essential reporters are understanding how readers are reacting to our content.
So we look at page views and unique browsers like everyone, but we are also looking at recirculation from an article — how likely are people to click on a second page — and we are also looking at the number of local readers visiting a story.
But the biggest change for us has been a move away from the just looking at the big numbers towards engagement metrics too, such as time spent per page and the stories loyal readers — the ones visiting every other day- were looking at.
We also wanted reporters to think about their average engaged time — this is a huge metric for us in the newsroom.
Often we need to match the right storytelling techniques to try and increase engaged time on an article.
Putting that metric on the radar of reporters has resulted in us seeing that stories with people at the heart of them retain readers for longer, as well as narrative storytelling.
Digital publishing standards also help improve active engaged time, and focusing on active engaged time also meant we learnt the stories which weren’t resonating with readers, prompting us to try new things with that content.
We are now starting to measure what we call high impact stories. These are the golden stories — the stories that really really engage people on an article for a long time but they also attract local users.
High impact stories are a real mix. It’s a bit of a myth that only long-form or issue-based journalism can be engaging. We’re finding trending and live content be be high impact too.
It’s just about how you tell these stories. We know now having done months a month of looking at high impact stories that court live blogs court are around six times as engaging for us as a standard court story.
Six things we learnt since launching Birmingham Live
- The stats don’t work if they’re just in the hands of a few people. So if it’s only the newsdesk or the key managers and the people that run conference that know about those stats we’re not really going to draw the audience very far. There has to be a knowledge and awareness across the newsroom.
- Reporters need context to apply data to their roles. So previously I think we’ve thrown a lot of numbers at reporters and expected them to know what they are. That doesn’t work. We really need to find the time and space to explain those numbers and give some context to each and every reporter so they can figure out what they really mean for them.
- What does not work in is often much more important for us than what is. So every morning we look at the underperforming stories in conference and we try to interrogate every story that’s not working and figure out why that was. Was it the sell? Was it the picture? Was the social sellnot strong enough? What else can we do to make sure that that doesn’t happen again next time.
- Different metrics can motivate staff in more niche roles or roles that don’t bring in those sort of traditionally really high numbers that we see from court or live coverage. If I’m to talk to our parenting editor about the big numbers like page views and uniques, it’s probably not going to compete with a big court story or live news. But she has built up an audience of her own, with10,000 people engaging with her and her content on Facebook every day. That’s clearly valuable to our audience so we should be using our metrics to highlight that and to pat people on the back.
- We also think it’s important to give reporters a forum to share those discussions. So we have a weekly ideas meeting where each reporter is expected to come and talk to us about what the data told them over the last week. But it’s really useful for how the reporters can connect with each other. There’s lots of questions, people sharing lots of information, and then applying that to new content.
- Data needs to be used not just for performance but to plan coverage iof every big event. Every time we have a large event, something it’s really important to our audience to cover, we forensically look at the data before we start. An example is our coverage of Ramadan currently. We wanted to really know what people wanted to know about it. Our trends analyst worked on it for a number of days to try and find out exactly what — and it’s constantly changing — but exactly what people are searching for. We do that now as a project plan for every single event or high audience job for us and we find that really really effective. We should know what the audience wants to know almost before they start to search it.