How a Slovenian publisher used AI to boost traffic by 20% and save its print magazine

At the Asian Media Leaders Summit 2026 in Singapore, media executives debated the risks and rewards of artificial intelligence. While large publishers have spent the past two years building massive custom language models, smaller newsrooms are finding different ways to compete. For Dnevnik, a Slovenian publisher with fewer than 100 journalists, the adoption of off-the-shelf AI tools did not just speed up digital news. It actually provided the human bandwidth necessary to launch a highly successful premium print magazine.

Competing without custom models

Massive custom AI models often feel like a luxury reserved exclusively for the giants of the publishing world. Dnevnik editor and journalist Damjan Franz Lu explained to the summit audience that his newsroom simply did not have the resources to build proprietary systems. Instead, they needed a tool built for their reality, their workflow, and crucially, their language. The publisher integrated Newslabs.io to manage their daily digital output through a centralised dashboard.

The integration eliminated the need for disjointed general language models and provided flawless transcription that perfectly supports Slovenian and regional languages. According to Franz Lu, the tool reduced the time spent on extreme cases like transcription and fact-checking by 90 per cent. Standard daily news research and outlining saw a 50 per cent time reduction, while deep structuring for long-form reportage became 33 per cent faster.

The implementation delivered immediate metric successes for the digital operation. Dnevnik saw a 20 per cent growth in unique users and an even larger increase in pageviews. Importantly, this digital growth was achieved with zero increase in newsroom headcount. The dashboard also provided macro-level intelligence, tracking language-neutral trends across global media to identify high-pickup stories in real time.

Reinvesting time into print journalism

The most significant outcome of automating the heavy lifting of daily digital news was not digital growth, but rather the preservation of legacy formats. By unburdening the core team, Dnevnik freed up the human bandwidth needed to launch a highly successful premium print magazine. As Franz Lu noted during his presentation in Singapore, AI did not kill print, but instead saved it.

This strategy highlights a crucial gap in how the wider industry is adopting automation. A recent Knight Center report found that newsrooms primarily use AI to speed up familiar tasks like summarisation and transcription. Report author Djordje Padejski argues that efficiency should create room for wider reporting and new community relationships, rather than simply generating a larger volume of content. Dnevnik offers a practical example of this reinvestment model in action.

The newsroom also leveraged AI to support its weekend output, turning quiet shifts into high-traffic days. During an empty Sunday shift, a reporter secured a phone interview regarding a mountain rescue, which was instantly transcribed and smoothed via AI. The breaking news piece was published in 30 minutes and became the most read article of the day. A similar Sunday success involved an AI-suggested evergreen topic on the adult rediscovery of literary classics, which was outlined, sourced, and published on the same day.

Future workflows and platform expansion

The small players can now compete in quality and quantity with the giants, Franz Lu told the summit attendees. To maintain this momentum, Dnevnik is looking toward planned functionalities that will further refine their editorial workflow. These include better control over research scope, the ability to use proprietary files as sources, and source verification checks before writing begins.

The publisher also plans to integrate automated photo insertion and Google Trends tracking for people, events, and topics. Audio capabilities will expand to include live transcription, merging multiple audio files, and direct transcription from YouTube videos. Furthermore, the newsroom intends to expand its social media automation to include new platforms like Facebook, adding to its existing distribution on X, Instagram, Reddit, YouTube, TikTok, and LinkedIn.

These upcoming features align with the broader industry push to apply AI to specific journalistic problems rather than building technology for its own sake. As NewsLabs’ CEO likes to point out, we’ve built a scalpel for the newsroom while most still try to create wins by relying on a tank (like Claude, ChatGPT, or Gemini). By maintaining a focus on practical utility, Dnevnik ensures its journalists spend less time on repetitive tasks and more time on high-value reporting.

The Dnevnik case study demonstrates that the true value of newsroom AI lies not in replacing journalists, but in reallocating their time toward premium products. As more publishers hit the ceiling of what efficiency alone can deliver, the focus must shift toward strategic reinvestment. Media executives must now decide whether to use automation simply to cut costs, or to fund the kind of ambitious journalism that builds lasting audience loyalty.