With input from Jostein Larsen Østring, Amedia’s Vice President for Development


Owner of more than 70 titles, Amedia is Norway’s largest publisher of local newspapers and reaches a daily readership of some 1.8 million people.


In 2016, Amedia began a concerted effort to analyse the data coming from their newsrooms to better understand their audience and the content that audience was paying attention to. In the first instance, they spent around two months working with chosen newsrooms, looking at different topics the publication covered in order to enhance the experience for their readers and better understand how to balance free articles and those behind the paywall. The analysis was expanded to other areas, including gender data. One of the main learnings Amedia has taken away from these projects is the importance of relevance when it comes to creating content for its readers and subscribers. A number of factors influence what stories readers are interested in, such as their age or geographical location, meaning Amedia newsrooms need to produce a variety of content that is relevant to these different audience segments in order to keep them engaged. In the case of gender, Amedia looked at articles from 19 newspapers published across a 21-month period and discovered that publications which featured more stories containing women sources had more women readers. This correlation was identified as part of a wide-ranging investigation into gender balance across 660,000 stories from 64 newspapers published over the same 21-month window. It revealed that, on average, only 34% of names mentioned in a story belonged to women.


The ultimate goal is to better serve all segments of the population who choose to read Amedia’s newspapers. Improved gender representation also has the potential to attract more subscribers and increase revenue.

  1. Externally: To use the data to attract women readers, deliver more clicks in the short term and encourage more women to read and subscribe to Amedia’s titles in the long term.
  2. Internally:
    • To encourage newsrooms to use the data to understand which stories are more relevant to women, and act to increase engagement among women subscribers
    • To encourage culture change in Amedia newsrooms, making it possible to implement the desired change and have a real impact


Amedia used natural language processing techniques to automate the process of identifying and counting the number of women and men mentioned in its content. They first identified so-called entities referenced in stories, such as people, businesses, or organisations, in order to differentiate between, say, a business with ‘Nina’ in its name and an actual person named ‘Nina’. The second step involved matching recognised names with Statistics Norway’s public database of men and women names. With this data in place, they were able to run analyses on content published by Amedia newspapers.


While Amedia’s analysis of 660,000 stories revealed that, on average, only 34% of names mentioned belonged to women, a closer look at the data revealed significant variationin aseemingly consistent picture.Forone, gender balance differed across the 64 newspapers that were part of the analysis. The top performer had a 42% share of women’s names in its stories, while the newspaper at the other end of the list had a 28% share. In general, the content of smaller publications tended to be more gender balanced than that of larger, regional ones. Given that the latter more often work on regional and sometimes national topics, this could indicate that the authority figures they interview are more often men, which could be as much of a contributing factor as the choices Amedia’s journalists make. The differences in gender balance became even more apparent when Amedia mapped gender across different story topics.

In ‘Education’, which encompassed some 30,000 stories, 47% of all names belonged to women, compared to only 21% for the 37,000 stories under the topic ‘Disasters, emergencies and accidents’. The category dubbed ‘Society’, which covers topics such as communities, families, and welfare, was the only one in which women’s names were in the clear majority. Again, these differences in gender representation may well be a result of authority figures skewing towards being women or men in different sectors. However, they could also stem from journalists’ own gender biases influencing who they choose to interview.

Lastly, Amedia discovered evidence to support their belief that better gender representation makes business sense. After analysing stories from 19 newspapers published across the same 21-month window, they found that publications which featured more stories containing women sources had higher readership among women.


Amedia used their data to identify stories and subject areas which are more relevant for women readers. This helped them understand what they need to do in order to produce content that is more relevant to women and increase the number of women subscribers.


The biggest challenge is to ensure that the data changes everyday behaviour and is used by editors and journalists. The response amongst staff differs dramatically. Some embrace the new information while others ignore it. Most people in the newsroom do understand the value of the data and are using it to think about how they can better reach their audiences, but others remain focused on short-term deadlines and their specific areas of interest – they choose not to prioritise this initiative. The key to success is to drive internal culture change and convince everybody within the organisation that this is a priority. That will take time, but within a year or two Amedia hopes that this work will lead to significantly more women sources appearing in stories and more engaged women readers.


  • It takes time to get to a position where data is informing the choices being made by people working within an organisation.
  • Findings are being introduced to news editors who will have a responsibility to translate the findings into action.
  • Gender balance differs across publications and topics. Smaller titles are generally more gender balanced than larger, regional ones. Women are better represented in topics such as health, education, and society, but are featured much more rarely in categories like emergencies and sports.
  • The differences in gender representation may well be due to authority figures skewing towards being women or men in different sectors. However, they could also stem from journalists’ own gender biases influencing who they choose to interview. In short, the answer is not clear cut.
  • All newsrooms are different and have different priorities. The project’s impact will really start to be seen next year, when a significant number of newsrooms have had time to use this data to change gender representation in their stories.
  • Newsrooms are spending a lot more time on gender discussions, and data analysis is driving an increased focus on gender balance.
  • Work to reach shared gender equality goals is also good for business, leading to more readers and subscribers.
Close Bitnami banner