HNA’s Philipp David Pries on local news and moving with the [digital] times

In the UK and US the waning fortunes of the regional press is an altogether familiar story. Last August the UK’s Press Gazette reported falling revenues for many regional titles with only a couple of exceptions. A similar story seems to be playing out in the States, where, a recent Pulitzer to a local news outlet notwithstanding, local and regional titles are struggling in the face of competition from larger publishers able to cast their digital nets further afield.

But as John Oliver has said, “the media is a food chain which would fall apart without local newspapers”, and as we discussed recently with Norway’s Amedia, local news is an altogether different proposition in Europe, where it continues to both thrive and embrace the digital technology that has caused it to flounder elsewhere.

In Germany’s Kassel region, Hessische/ Niedersächsische Allgemeine (known more concisely as HNA) is one such publication. In addition to a print circulation of 200,000 readers per day, it enjoys a loyal readership of more than 7 million online visitors a month. We caught up with editor, Philipp David Pries (who also oversees the publication’s analytics activities), to find out a little more.

Regional and local news seems to be doing well in Germany. Why is that the case, do you think?

Largely, I think it’s that people in Germany are very passionate about their regions. In times of globalization and a world in turmoil, they want to know more than ever what’s happening on their doorsteps, and they want to be involved in their proximate region. This extends to us as publishers as well: we know that we need to produce content that’s relevant to our readers, and it works to our advantage at HNA to do so because the result isn’t being replicated in Berlin, Munich or Hamburg.

Does the regional press in Germany approach news in the same way that HNA does, or are you doing something differently?

What makes us stand out is that we always lead with local — online and in print. When you look at other regional newspapers and their websites, you’ll mainly find that the first page is full of big politics, big sports, big television.

I’m always surprised by this because it seems to me that their USP should be local news. A big, big share of our content is local news and local sports.

The future of editorial analytics (in under 15 slides)

What are the issues you’re facing as publishers?

Loyalty is very important to us and we have a very strong base of users who are very loyal to the brand and loyal to the quality, so we’re lucky there. Newspaper activities still constitute the main source of revenue, although digital activities are slowly gaining a bigger share.

The problem is that it is becoming difficult to attract newer or younger audiences to subscribe to a newspaper. In fact, I’d say it’s nearly impossible and I only have to look at myself to see why: I’m in my mid-thirties, I’m on the internet all the time reading things from all kinds of news outlets. It’s normal for me, and it’s normal for many others — at least when it comes to daily newspapers.

Do you notice a drop off in subscriptions as a result of this?

Actually, we have a very small loss of subscribers each year, and this is usually those who move out of the region. It’s extremely rare to hear people say that they don’t like the content anymore and cancel their subscriptions. The challenges we’re facing now are centered around how to transform the business in these digital times and to tailor digital offerings to specific audiences.

That’s the million dollar question at the moment, though, isn’t it? How do you make that transition, given the issues with subscription models and ad revenue? How do you balance that at HNA?

It’s about volume, but above all it’s about delivering the right content to the right people.

Our base is HNA.de with more than seven million monthly visits, and because the business model is based on ads we aim to get the reach as wide as possible. We are convinced that we can achieve this target best with the strong local focus mentioned earlier.

There’s also Kassel-Live, the 24/7 coverage of the Kassel region, and we have a different focus here: it’s more about snippets rather than articles. More photos, videos, social media, nightlife and entertainment tailored to a young audience. It’s a real honeypot for attracting new and younger readers and getting them involved in the HNA cosmos, so it’s very informative and ranks really highly in the Google rankings, especially when it comes to timeless local service content. The theory is that people will land on Kassel-Live perhaps without having the HNA brand in their minds and find that the content is useful, relevant and engaging, and in doing so we might attract them to us in the long run.

How do you succeed with the subscription publishing model?

Are you delving into the world of subscriptions?

From June onwards, the section we call HNA7, a digital regional magazine where we try to offer premium content, will be positioned on a subscription model. We’re looking here at ways to gain new audience insights; to enrich the local content we’re producing through different approaches to digital storytelling, through more data journalism, and of course all of those things take more time. Hopefully we’ll be able to offer the reader a broader approach into local players and activities, beyond what might simply be called ‘the news’.

Naturally, we’re always interested to hear about analytics. What kind of things are you using to measure digital success at HNA?

Well, we’re still not at the point that I’d like to be, but at the moment we’re using a mixture of different kinds of short and long-term analytics — monitoring and surveys aligned to our editorial priorities and strategies.

I wonder if you’re able to drill down into analytics to glean any information which is more sophisticated than simple sales figures?

One of the basic things that we know from analytics is that the majority of our users are returning visitors who like the brand HNA — users who habitually use the site rather than coming from flyby sources like Google. I know of other newspapers where up to two thirds of their traffic comes from social or Google, but to me, those kind of percentages don’t speak that well for loyalty. We also chart user behavior on the page and all digital channels, activities and apps. Our overarching task is to derive actionable insights and to find the hidden connections which drive readers.

Do your analytics inform your editorial choices?

Of course. All of the analytics we use have an important function — providing we know how to make them work for us. We use both day-to-day analysis as well as longer term insights in the editorial process. Both are really important because you have to gain insights into things you might need to adjust. You might have to enrich an article or promote it in a different way or give it a different headline or publish it elsewhere — all minimally invasive editorial decisions, but ones which can drastically optimize the output. And to just name a few additional analytics activities — driving our growth strategy, using segmentation to understand how content worked for our audiences, tagging, creating KPIs that reflect editorial and audience priorities… To put it breifly: data-informed activities which try to narrow the distance to editors/reporters and other departments.

How has that analytics use changed since you’ve been working as a journalist?

Completely. We started with Yahoo Analytics what feels like centuries ago. Those analytics were only able to present page views and visits and that was it, but then I think we were happy to have any data, because even that gave us more than we used to get from the newspaper where, apart from reader scans, we really knew next to nothing about our readers and how people used us.

Things have changed dramatically in those five years since Yahoo Analytics — some industries don’t change that much in 20 years. We’re in a very dynamic phase and a dynamic time right now where a lot of new tools and methods and workflows are developing. I think the money has flowed into these analytics as the last man outside has finally understood that digital won’t disappear. There’s a growing attention and increase in money that’s being put into these analytic tools and the tools are improving correspondingly.

I’m very happy that we’re now able to profit from these tools and workflows, and we see the result of that in better journalism — and in the end that’s what’s important.

Is that the key change do you think — that analytics have become more appropriate for an editor, rather than for the advertiser or the marketer?

The kind of analytics which are available now help editors decide how to invest their time and resources in the best possible way, and this can only be good for journalism. Even 10 years ago it was likely that the only kind of feedback you might receive might be something along the lines of a letter to the editor.

“Page views and likes were an easy fit and it has been difficult up to now to find a consensus about a metric which could replace the page view”

I’m not an expert in advertising because obviously I work on the journalistic side, so I have no deep insights into that world, but up until recently those analytics were driven by the advertising business model: page views and likes were an easy fit and it has been difficult up to now to find a consensus about a metric which could replace the page view. Whether we like it or not, our online revenue is largely driven by that advertising model, so we have to pay attention. Nevertheless, as business and editorial goals partly diverge, editorial analytics have a right to exist. More than ever.

How’s that going to develop at HNA, do you think?

Well, as the data becomes more sophisticated we’ll have more options. There has been a lot of talk about behavioural data concerning analytics to accurately measure user interactions as best as possible, for example. We try to define quality metrics through scored metrics to measure the value of interactions. We try to make more A/B experiments. And it would be interesting to try to combine our analytics insights with profile data — at that point we’d be able to get considerable insights to tie the user with their interests and their location and so on.

So that’s a highly personalized approach to news consumption and production.

Yes. It has happened elsewhere. If you look at Amazon, that’s a great example right there. If I open it I see a lot of garden and hiking things — stuff which reflects my personal interests — but for you it would probably be completely different. Compare that with the newspaper: that still produces the same content for everyone and presents it in the same way for everyone — it’s anachronistic.

Of course, it’s easy to make this observation, but much more difficult to do anything about it. Personalization is a very complex task with a lot of challenges. The Washington Post [now owned by Amazon’s Jeff Bezos, of course] have started to transform their business in quite a radical way based on data. We could also talk about machine learning, about cross-device tracking, about websites taking into consideration how readers interacted with the content before.

“Even as a regional player we should keep in touch with the developments and try to implement and experiment with them as much as we can”

These are the things on the big horizon. Of course, the HNA doesn’t have that kind of manpower and we don’t have these resources at the moment, but even as a regional player we should keep in touch with the developments and try to implement and experiment with them as much as we can. So we need more behavioural analysis, we need more content insights and more metrics which really reflect what the user does when it comes to the right activities.

How do you reconcile a personalized newsfeed and experience with the problem of filter bubbles?

Well, it certainly needs some kind of surprise and priority element in the news feed — a kind of, ‘you might not think you’re interested in this content, but we’ll present it to you nevertheless in a nice way’. If you combine this with news that we know the user will want to read based on their former behavior, I think you’re maybe looking at an ideal combination. This is a question for ten years’ time, I’m sure: we’d need new algorithms and new workflows. Journalism in general is clearly more than business: it needs to fulfil a public, enriching function, and I am convinced all this is fostered and supported by editorial analytics more than ever.


Originally published at contentinsights.com on May 25, 2017.