Prediction
Should we predict or define?
Name
Thomas Baekdal
Excerpt
“Any time you hear about publishers using AI to produce journalism as if it’s factory output, that’s when you know they’ll have a bad year ahead.”
Prediction ID
54686f6d6173-25
 

When it comes to looking at the year ahead, we can approach it in two ways: We can either predict what we think will happen, based on the trends, or we can define what outcome we want to achieve.

I’m going to do both, because they’re not the same.

Let’s talk predictions first. There are many things that, based on what we already see, seem likely to happen in 2025.

First, the declining level of traffic from social media will likely continue to decline. Also, the focus many publishers have started to put on TikTok is unlikely to lead to any real monetary improvement — nor is it likely to change the problem of getting young people to subscribe.

Similarly, channels like Threads or Bluesky will grow, but not at the scale we used to think about social media. As such, they are more or less irrelevant when it comes to the actual profitability of publishers.

This isn’t just because of changes to those channels, but more because of how the public has changed. The public simply doesn’t use social media the way they used to, and news, in particular, is a really bad fit for that changed behavior.

Google Search is also likely to continue the declines we’re already seeing, with the total volume of search traffic going down. Add to that the introduction of AI into search, which will have a polarizing effect on who gets the traffic. In other words, fewer sites will get a larger share of the total traffic, leading to the problem of “the rich getting richer.” So, if you’re a smaller publisher, search is becoming a real problem.

It’s simple math. With old search, you’d get a full page of search results. Obviously, the top results would get more of the traffic, but you would still see traffic from results further down the list. With AI, people get one answer, so all the attention is now focused on just that one source.

Another issue with search is how Google has changed its algorithm to only rank publishers based on their primary focus. This is a significant change, because it means you’ll only rank high with articles that Google considers to be your primary focus, while everything else you do will be down-ranked. From a user perspective, this is great, because the highest-ranked sites are also those with that expertise. But for newspapers, this is a disaster because all those other things you do — like entertainment, car reviews, etc. — will no longer be boosted by the ranking of your political reporting.

Speaking of AI, 2025 will obviously be filled with it. In fact, it”ll likely be overflowing with it. This, however, will take us in two opposing directions.

One direction is how AI is transforming the way we work, the way we define journalism, and how we interact and create relevance for our audiences. In 2025, we’ll see a lot of interesting efforts in this area. It will take much more than a year to figure it all out (more like 10 or 15), but 2025 will be another year of AI experiments.

But in the other direction are all the AI missteps. These occur when publishers start using AI in ways that undermine their value. In 2024, we saw a ton of that. Publishers started using AI to produce articles that nobody should ever read, because they were very poor quality and had nothing to do with real journalism. We’ve seen companies turning to AI voices, which made their audio sound generic and disconnected. And it’s not about the quality of the voices — it’s the purpose of it.

In 2025, we’re going to see many, many more such missteps, and it will likely ruin several publishers, because they’ll only realize the damage they are causing to themselves too late. Basically, any time you hear about publishers using AI to produce journalism as if it’s factory output, that’s when you know they’ll have a bad year ahead.

The next trend is audience, and many of the arrows are pointing in the wrong direction. Trust is declining; news fatigue, avoidance, and burnout are rising, and polarization is getting worse, etc. On top of this, we have another Trump presidency, which will cause another Trump bump — although at considerably lower levels than previously — but will also further worsen all the trends I just mentioned.

What’s important to remember, though, is that people need news. It’s just that we aren’t doing it in a way that actually drives relevance. This leads us to the second question: What should we focus on in 2025?

Simply put, we need to focus on changing the way we interact with our audiences to reverse the trends. And doing this is actually quite easy.

The U.S. is kind of weird because the way the public interacts politically is almost the opposite of how people interact non-politically. Take something like climate change. Politically, it’s a topic that is tearing people apart. But as soon as people realize how much money they can save, suddenly it no longer matters what the politicians are saying.

It’s the same with companies. Right now, every corporate executive knows that in order to stay competitive, they need to focus on the next generation of products, many of which need to be sustainability-focused to be competitive. Because if they don’t, China is going to roll right over them. China is fast becoming the world leader in the next generation of climate-focused products, whether that’s cars, consumer goods, or industrial equipment.

So every executive knows that they need to innovate as well. It doesn’t matter what the political situation is like, nor what political party people prefer.

This disconnect, especially in the U.S., has a huge influence on the trend above, and it explains why we in the press face the problems that we do. But we can fix this by focusing on what people actually need, rather than on what they are politically obsessed with.

A perfect example of this is what happened to The Washington Post when they announced that they didn’t want to endorse a presidential candidate. Immediately, it lost 250,000 subscribers. It’s important to understand that this only happened because their focus was so heavily skewed towards the political drama, as opposed to being focused on what would actually be relevant, useful, and valuable to people in their day-to-day lives.

In 2025, we’re going to see a lot more of this. But the publishers who will come out ahead are those who understand that this is an unhealthy focus that undermines the value of their journalism, and instead shift their attention to what people need from them outside of their political obsessions. So this is the most important focus for the year ahead.

Thomas Baekdal is a media analyst and founder of Baekdal.

When it comes to looking at the year ahead, we can approach it in two ways: We can either predict what we think will happen, based on the trends, or we can define what outcome we want to achieve.

I’m going to do both, because they’re not the same.

Let’s talk predictions first. There are many things that, based on what we already see, seem likely to happen in 2025.

First, the declining level of traffic from social media will likely continue to decline. Also, the focus many publishers have started to put on TikTok is unlikely to lead to any real monetary improvement — nor is it likely to change the problem of getting young people to subscribe.

Similarly, channels like Threads or Bluesky will grow, but not at the scale we used to think about social media. As such, they are more or less irrelevant when it comes to the actual profitability of publishers.

This isn’t just because of changes to those channels, but more because of how the public has changed. The public simply doesn’t use social media the way they used to, and news, in particular, is a really bad fit for that changed behavior.

Google Search is also likely to continue the declines we’re already seeing, with the total volume of search traffic going down. Add to that the introduction of AI into search, which will have a polarizing effect on who gets the traffic. In other words, fewer sites will get a larger share of the total traffic, leading to the problem of “the rich getting richer.” So, if you’re a smaller publisher, search is becoming a real problem.

It’s simple math. With old search, you’d get a full page of search results. Obviously, the top results would get more of the traffic, but you would still see traffic from results further down the list. With AI, people get one answer, so all the attention is now focused on just that one source.

Another issue with search is how Google has changed its algorithm to only rank publishers based on their primary focus. This is a significant change, because it means you’ll only rank high with articles that Google considers to be your primary focus, while everything else you do will be down-ranked. From a user perspective, this is great, because the highest-ranked sites are also those with that expertise. But for newspapers, this is a disaster because all those other things you do — like entertainment, car reviews, etc. — will no longer be boosted by the ranking of your political reporting.

Speaking of AI, 2025 will obviously be filled with it. In fact, it”ll likely be overflowing with it. This, however, will take us in two opposing directions.

One direction is how AI is transforming the way we work, the way we define journalism, and how we interact and create relevance for our audiences. In 2025, we’ll see a lot of interesting efforts in this area. It will take much more than a year to figure it all out (more like 10 or 15), but 2025 will be another year of AI experiments.

But in the other direction are all the AI missteps. These occur when publishers start using AI in ways that undermine their value. In 2024, we saw a ton of that. Publishers started using AI to produce articles that nobody should ever read, because they were very poor quality and had nothing to do with real journalism. We’ve seen companies turning to AI voices, which made their audio sound generic and disconnected. And it’s not about the quality of the voices — it’s the purpose of it.

In 2025, we’re going to see many, many more such missteps, and it will likely ruin several publishers, because they’ll only realize the damage they are causing to themselves too late. Basically, any time you hear about publishers using AI to produce journalism as if it’s factory output, that’s when you know they’ll have a bad year ahead.

The next trend is audience, and many of the arrows are pointing in the wrong direction. Trust is declining; news fatigue, avoidance, and burnout are rising, and polarization is getting worse, etc. On top of this, we have another Trump presidency, which will cause another Trump bump — although at considerably lower levels than previously — but will also further worsen all the trends I just mentioned.

What’s important to remember, though, is that people need news. It’s just that we aren’t doing it in a way that actually drives relevance. This leads us to the second question: What should we focus on in 2025?

Simply put, we need to focus on changing the way we interact with our audiences to reverse the trends. And doing this is actually quite easy.

The U.S. is kind of weird because the way the public interacts politically is almost the opposite of how people interact non-politically. Take something like climate change. Politically, it’s a topic that is tearing people apart. But as soon as people realize how much money they can save, suddenly it no longer matters what the politicians are saying.

It’s the same with companies. Right now, every corporate executive knows that in order to stay competitive, they need to focus on the next generation of products, many of which need to be sustainability-focused to be competitive. Because if they don’t, China is going to roll right over them. China is fast becoming the world leader in the next generation of climate-focused products, whether that’s cars, consumer goods, or industrial equipment.

So every executive knows that they need to innovate as well. It doesn’t matter what the political situation is like, nor what political party people prefer.

This disconnect, especially in the U.S., has a huge influence on the trend above, and it explains why we in the press face the problems that we do. But we can fix this by focusing on what people actually need, rather than on what they are politically obsessed with.

A perfect example of this is what happened to The Washington Post when they announced that they didn’t want to endorse a presidential candidate. Immediately, it lost 250,000 subscribers. It’s important to understand that this only happened because their focus was so heavily skewed towards the political drama, as opposed to being focused on what would actually be relevant, useful, and valuable to people in their day-to-day lives.

In 2025, we’re going to see a lot more of this. But the publishers who will come out ahead are those who understand that this is an unhealthy focus that undermines the value of their journalism, and instead shift their attention to what people need from them outside of their political obsessions. So this is the most important focus for the year ahead.

Thomas Baekdal is a media analyst and founder of Baekdal.