Germany’s Handelsblatt fights AI traffic slump with ‘content warehouse’ and Smart Search
AI is not only changing how journalistic content is produced but also how people access that content. For at least the past year, many news publishers have seen visitor numbers to their websites fall dramatically as AI-related summaries are increasingly replacing traditional search results.
As users change their consumption habits to look for direct answers rather than articles relating to a given topic, publishers need to rethink how their content is distributed and how they can retain control of the access to their content, Janina Reimann, Director of Digital Products, Handelsblatt Media Group, told participants at WAN-IFRA’s Frankfurt AI Forum.
WAN-IFRA is working on a report about how publishers are responding to the impact of AI Search and bot traffic. We would be thrilled to have your perspective. TAKE OUR SURVEY: in English, German, French, Spanish.
“In the past we had the content we created, we had our products, we could build paywalls or advertising. We decided how we want to deliver the content to the customer,” she said.
Building a central content hub
Seeing the changing consumption patterns, about two years ago, Handelsblatt began preparing “for a world where we don’t really know how exactly customers will consume our content,” Reimann said.
That meant building a content warehouse, essentially a centralised content hub where all of Handelsblatt’s content is structured for different outputs. That includes articles, podcasts, video as well as transcriptions of their podcasts and from the hundreds of conferences they have every year.
Handelsblatt uses records of the panels, the discussions, the presentations, and also transcribes them to be included in the content warehouse, she said.
“The aim is to have a powerfully structured database for everything we’re doing in future,” Reimann said.
The platform is intended to be a truly central hub for both internal use by staff as well as externally by paid subscribers.
It is also very important for Handelsblatt’s syndication business, she added.
“In the field of business-to-business, we also are able now not only to sell our articles but also our audio and video content,” Reinmann noted. “This is something we put a lot of effort into in order to really make our content accessible, because this foundation is important for everything we do with AI.”
Handelsblatt aims to provide consumers with exactly what they are looking for and suggesting additional resources, while keeping them within Handelsblatt’s platform and using their products and services. They see this as a powerful engagement tool that will keep consumers on their platform longer, which also helps to decrease churn.
WAN-IFRA Members can download Janina Reimann’s presentation from our Knowledge Hub, by clicking here.
Learning what their consumers value most
As part of the preparation for building their content warehouse, Handelsblatt conducted extensive market research to better understand what their customers really wanted and expected from them. They also wanted to find out why users would come to them rather than just using AI-related search to get an answer.
What they learned, Reimann said, was that first and foremost, their customers wanted quick, direct answers.
“They really want to save time. This was their most important use case,” she said.
Another finding was about the importance of trust.
“We have the trust in the content we deliver as a publisher, and customers don’t have to double check content that they have found on our platform,” Reimann said.
Launching Smart Search, and maintaining trust
Based on this, Handelsblatt launched their first customer-facing AI product, Smart Search.
While Smart Search is similar to the chatbots found on platforms from other news publishers, Handelsblatt put an added emphasis on trust. Knowing how important trust is to their consumers, Handelsblatt won’t allow Smart Search to give answers that are not accurate.
“We really have strict guidelines,” Reimann said. “We’d rather say to the system, ‘Don’t answer a question if you don’t have enough sources.’ ”
What was interesting, she noted, is that while users are sometimes frustrated because they don’t get an answer, they also say this increases the trust in the answers they do get. They know if there is an answer, they don’t have to double check. They know it’s valid, she said.
Another important aspect for Handelsblatt is they didn’t want to build a new standalone product, but rather one that people use to interact with the publisher’s core products.
“What we really focus on is how we can cross-promote our different kinds of content,” Reimann said.
While the content warehouse provides access to Handelsblatt’s content, there is also a linking logic behind it, she added.
“We don’t only give an answer, but we also give hints of further articles, interesting podcasts or events with the possibility, of course, to buy an event ticket for example, so we really focus on this cross-promotion of our content,” she said.
Preparing for change with business goals in mind
Ultimately, Handelsblatt’s goal is to be well-placed to move with consumer expectations as they develop and change.
Reimann notes that, while no one really knows what will happen in the next two to three years, Handelsblatt believes their efforts will prepare them to react to different market developments.
In addition, in terms of Handelsblatt’s own products, she said it’s important for them to invest in conversational interfaces because people are becoming more accustomed to directly getting answers to their questions instead of searching through article lists.
Thirdly, it’s not an option for media companies to ignore these developments. “We can’t stop the user behaviour of moving more and more to the AI systems. We really have to think about how we handle this, and how we can react to it, always based on the business goals that we follow as media companies,” Reimann said.
