DC Thomson

Lesley-Anne Kelly, Head of data journalism

October 2, 2024

Creating powerful local content:
How DC Thomson use data journalism to track the vitality of city centres

This case is an excellent example of how newsrooms can produce data to underpin valuable local journalism. Scottish city centres have been hard hit by the pandemic, high fuel costs and online shopping – something which the data journalism team at DC Thomson wanted to cover in a comprehensive way. The work involves painstaking and continuous on-the-ground mapping of local high streets to produce structured data. This is then published as interactive high street trackers alongside engaging stories about the people and businesses involved. The initiative won the team the Best Data Visualisation award at the WAN-IFRA Digital Media Awards Europe 2024.

Presented by Lesley-Anne Kelly, Head of Data Journalism, DC Thomson, Scotland

Heading 4

By Cecilia Campbell

Article Summary

Lesley-Anne Kelly, DC Thomson’s first data journalist, spearheaded a project to quantify the decline of city high streets in Scotland, driven by post-pandemic challenges and shifting shopping habits. Recognising a gap in available data—especially at a granular level—Kelly and her team undertook the ambitious task of creating a comprehensive database of high street retail units across multiple cities, beginning with Aberdeen and Dundee.

The resulting High Street Tracker not only maps the vacancy/occupancy of retail spaces but also categorises them. The initiative aimed to foster positive change in local communities by providing actionable insights and give reporters a tool to add context to their everyday reporting. Utilizing tools like Google Maps and community engagement, the team built a living database that now tracks around 2,000 retail units.

Launched in June 2023, the Tracker features interactive maps and analytics that help local reporters generate stories and offer context to their articles. The project has garnered significant engagement, prompting local authorities to take action, such as committing funds for high street improvements. With plans to expand further, Kelly emphasizes that this is just the beginning of leveraging data to drive impactful journalism and community revitalization.

DC Thomson is a family-owned publishing house headquartered in Dundee in Scotland, with a presence in London and across the globe.

The group has four regional newspapers. Two of the titles, The Press and Journal and The Courier are two of the top three selling regional daily newspapers in the UK. DC Thomson also publishes a Sunday national paper for Scotland.

The media gourp also includes magazines, radio, a thriving genealogy community and a technology solutions business.

Background: A desire to quantify the problem of high street decline

Seven years ago Lesley-Anne Kelly became DC Thomson’s first data journalist. She came from a background working in the public sector in Scotland in a role that involved a lot of data analysis, while also working as a communications officer for her union branch. “So in a way I fell into journalism by accident. Looking at it now – I believe data journalism is really important for getting to the root of what is happening in our local communities.” After four years, Lesley-Anne became the Head of Data Journalism, and now leads a team which includes two fellow data journalists.

The post pandemic state of the region’s city high streets was a story that was crying out to be told. “Our high streets are suffering – between cost-of-living issues, the lasting effects of the pandemic, and the prevalence of online shopping, vacant units are a constant in our city centres. We wanted to be able to quantify the issue and to see whether it was getting better or worse.”

But there was a major blocker.

There wasn't any data available at either the granularity (unit by unit) or the regularity (up to the minute) the team really needed. There was some national data looking at city level vacancy rates yearly or twice a year, but nothing at the street by street, unit by unit level that the team needed.

The solution was to build the data set themselves, from scratch.

Creating data that drives change

Importantly, when setting out to create their high street data, for the team the project was about more than just tracking vacancy rates. “We wanted to look at the makeup of our high streets. They are the heart of our communities. What businesses are thriving on our high streets and what works really well? What doesn't? Our high streets are such a blueprint of our local communities.”

But it went even deeper than that – Lesley-Anne and her team live in the cities they report on after all. “We wanted to create data that drives change. That was really at the heart of this project. By providing this data and reporting on it – could we impact positive change in our communities?”

Another purpose of the project was to create a tool for the newsroom reporters. The team looked through the group news sites and realised there were a lot of stories written and published relating to high streets on a regular basis, and what’s more the reader engagement with the articles was high. “We saw there was an opportunity to give reporters extra context for their stories, around what else has happened on that high street, over the past months or year. We saw that our data and tool could become the go-to-source for this topic in our region,” said Lesley-Anne.

What? A living database of high street units and high street tracker pages

The idea for build of the data was first pitched in December 2022 and work began seriously on it around February / March 2023. The first edition of the high street tracker was published in June 2023.

What the team has created is a living database of retail units on the high streets, as well as the shopping centres. The cities of Aberdeen and Dundee were covered first, and at the end of 2023 Inverness and Perth were mapped as well, “which means we now track half of Scotland’s cities.” The database currently includes some 2000 retail units.

On the news websites there are now dedicated High Street tracker pages, with visualisations of the database, including:
• 3D maps of high streets and shopping centers (above), colour coded for occupied or vacant. Readers can hover over the map and get information like when did a vacant one closed, how long have they been vacant for and similar.

• Charts (below) that look at which streets are the most vacant and which streets are doing really well in comparison to the average as well as things like what type of business thrives, shops compared to eateries for example.

The reporter tool includes all the public facing data for reporters but it also adds some additional information which reporters can tap into both for context to their stories and also use for story leads. “For example, one story was about the longest standing vacant unit in Dundee, which is an off-licence very near our newsroom. It’s been vacant for almost 15 years now and when it hit 5,000 days vacant our head of business did a story on that. DC Thomson now has comparable data across the four cities on a street-by-street breakdown as well as averages as to how each city compares. “We also take snapshots of the data at the end of each month so we can see the trends on street level as to how things have been changing and whether they're improving or not.”

How? Getting everyone involved!

1. Talk to local patch reporters. The project started out in Dundee and Aberdeen, as that’s where Lesley-Anne and her colleague live. However local newsroom journalists were very much engaged from the outset. “We had opinions on which streets we should track but we needed these local patch reporters as well to feed into that. We had meetings with the local reporters to discuss what the key areas were that would be kind of indicator areas for each city centre – the wisdom of that group of people is really integral to the project.”

2. Take a virtual stroll down the streets. The team used Google Maps Street View, and made lists of the retail units.

3. Take an actual stroll down the streets. The team then walked down each street to verify and to create the final version of the lists/maps for each street/shopping centre. Each unit was also categorised and subcategorised, e g “Food & drink” and “Pub”.

Maps were hand drawn using a free tool called Geojson.io and the entire project runs off Google Sheets, which are updated whenever new data is available.
The data comes in in various ways:
• Reporters who come across information during the course of their reporting
• Readers, who fill in information in the form at the bottom of high street tracker webpages
• Business owners, also often through the form
• The data journalism team go out and perform quarterly manual censuses of the high streets

(Also, check out the list of tools used at the end of the article.)

Launch and leveraging the data

Each city gets a Launch article which is a deep dive into the data and with interviews with local business owners. The Launch articles use data visualisation and scrollytelling, which works particularly well for mobile consumption. “We spoke to local business owners to include some of the people who are actually impacted by this data and we used data visualization and scrollytelling to really bring the data to life.” Each quarter when the team does the census, another such article is produced. “This keeps the project fresh in our readers’ minds and keep them reminded that we are the authoritative place to come for information about the high street.”

Each city has a Tracker article looking at the high streets and shopping centre. The Tracker articles, which include interactive maps, are written in an SEO friendly way, and as evergreen content sit on the site and gather clicks every day, which means they gain SEO authority in the eyes of Google.

The tracker articles are intended to present the data in a very straightforward factual way, but the team wanted to build out the data and create more newsier stories. One example was the result of a reader survey, where one question was “What would make you get out and use the high street more?” Many readers said they’d use the high street more if there were more independent shops. “We checked the data, and it turns out the high streets actually have a high proportion of independent and local traders, so that became a bit of a myth busting story.” Another example of building out the data was an investigation around the ownership of the vacant units, not an entirely easy undertaking due to the complexity of Scottish property ownership records.

Bringing the data to life: Town hall meeting

The team understood that the data also presented a real opportunity to engage with the community in person. “We organised a High Street Summit at the Dundee offices and invited everyone who had responded to the survey to put their opinions to our local city council leader. We also had local business owners, local experts in town planning and a keynote speaker that was from the UK High Street Task Force – he had never been to Dundee before but he had a lot of knowledge about high street regeneration. He was able to come up with fresh eyes and talk about Dundee's issues in a very interesting way.” One of the outcomes of the town hall was a 40 page white paper with all the analysis, which was sent out to subscribers.

The data journalism team and beyond

Lesley-Anne Kelly leads the project, and her data journalism team of three journalists are the core team for the project, but the project has also tapped into resources across the publishing group.
• Audience teams are involved in how best to sell the project in terms of getting it in newsletters and how to sell it in social media.
• Insight specialists help build the reader survey and advise on how to engage with readers in this direct way.
• SEO specialists are on hand to make sure the Tracker articles have solid SEO to really leverage the news publisher’s position as an authority.
• Most journalists across the newsrooms have been involved at one time or another.

The results: Exclusivity, more context, more data points, boosted search authority

The database was a considerable investment from the publisher’s point of view, not least in terms of journalists' time. But the benefits and gains are many, some easier to measure than others:

• Exclusive insights. By building their own database, DC Thomson newsrooms have access to exclusive data which in turn can generate exclusive journalism – journalism which is close to the heart of the local communities served.
• Impact journalism & strengthening the newspaper brands. Through the reader survey, the town hall and the general campaign, the local council has actually committed extra funds to clean up the Dundee high street. “We get to shout about the fact that we’ve created this campaign, we’ve made the council take notice and act, we reported on the town hall of course and show our readers that we care about them and our community.”
• Boosting local reporting. Reporters across the local newsrooms use the tool to find story leads and add context to their articles – thanks to the database they may have 10–15 extra data points related to a particular story.
• Boosting Search Authority. As mentioned above, any time a reporter covers a high street related story, they link back to the Tracker article, which has high Search Authority in Google’s rankings thanks to daily engagement and clicks. This sort of “rubs off” on the local high street article, and boosts the search authority of the news site as a whole.
• Awards! DC Thomson won the Best Data Visualisation category of the WAN-IFRA Digital Media Europe awards this year, as well as the Digital Story Telling (regional) category of the Press Gazette Future of Media Awards.

Key success factors & Challenges: Communication mainly!

Lesley-Anne Kelly stressed the importance of getting reporters in the local newsrooms involved in the project from the start – getting their support in building out the data, but also making sure they understand how they can leverage the database. “It took a while to get reporters to be aware of the project enough to integrate it into their daily practice, i e when they're reporting on these things, to check the data, put a few lines in and give readers some context. However, we're now in a really good place with that, which is excellent.”

Making it easy for readers and business owners to contribute is key. “We've attempted to crowdsource some of this data because we can't be out there all the time and we sometimes miss things. At the end of every article there's a form that readers can fill in if they spot something that is not in our maps already. We've found that local business owners have also been using the form which shows they really care about being on our map. But it also goes to the circularity of the project – with updated data there may be new stories for local reporters to pick up on.”

Resources was a challenge – particularly in the set-up phase. The data team works daily with various projects with the newsroom, so it was hard to find time to build up the first version of the database. “It was months of work to get that set up, however once we had the blueprint for the first two cities built, it’s now relatively easy to take that blueprint and drop it into other areas and build from there.”

It’s only the beginning…

DC Thomson now have High Street Trackers for Dundee, Aberdeen, Inverness and Perth, and in the next few weeks they will be adding a fifth area. Lesley-Anne Kelly says the past 18 months of work is only the beginning. “Whilst we've done dozens of articles of bespoke analysis from the high street tracker we've only scratched the surface. I imagine we'll continue to tap into this data to do more bespoke analysis and more high street related stories going forward. It seems to really resonate with our readers.”

Useful links and contact information

Here are links to presentations, the DC Thomson tracker pages as well as list of tools used by the team:

The webinar presentation can be downloaded here.
Lesley-Anne Kelly can be reached on lkelly@dcthomson.co.uk
Links to the tracker article pages: Dundee and Aberdeen
Link to a Launch article with scrollytell and visualisations.

The tools the team use are:
• Google sheets for the database
• Geojson.io (free) for hand drawing units
• Figma (free) for hand drawing the shopping centers to make the interactive shopping center floor plans.
• QGIS (free) for more advanced spatial analysis
• Flourish (paid) for generating charts and graphs. The charts all run through Google Sheets. “Essentially whenever we need to make a change we click one or two buttons within Google Sheets that then triggers off six or seven other bits of analysis to happen in the back-end and then that fires straight to Flourish and all of the maps and charts automatically update”

You are welcome to contact the WAN-IFRA Innovate Local team, if you have questions or examples of similar cases.
Cecilia Campbell: c.campbell@wan-ifra.org
Niklas Jonason: n.jonason@wan-ifra.org