Sorry, we could not find the combination you entered »
Please enter your email and we will send you an email where you can pick a new password.
Reset password:


By Thomas Baekdal - February 2018

Fascinating Traffic Experiments by Publishers

Last week, I came across two very interesting traffic experiments, which are worth talking about. One was about how a publisher tried to increase the volume of posts on YouTube, and the other was about a publisher who decided to see what happened if they stopped using Facebook.

Both are very interesting, so let's look at what happened with each and what the result was.

We will start with this:

Fstoppers tried going for volume on YouTube

Fstoppers is a site for photographers that I have admired for many years, and that I have used as an example of a brilliant editorial strategy.

In 2015, I wrote "Drop The Content Strategy, Create A Care Strategy Instead", where Fstoppers was mentioned as a site that did this really well. In June 2017, I mentioned them in "What is the Best Monetization / Subscription Model?" and, last August, they were also mentioned in "Can Your Readers Trust You With Their Time?"

You can probably spot a trend here. Fstoppers is a publisher who is really good at creating very high-value paid-for content that is worth people's time. And the way they earn most of their money is through very high-end tutorial series that you have to buy.

Like this one (price: $299):


The problem with high-value sites like this, however, is that the high-value doesn't necessarily create a lot of traffic, and, as a publisher, you start to get antsy about growth.

This happened to Fstoppers as well. They already have a successful business, but they started wondering if they could just post more to grow more.

The result was that, a month ago, Fstoppers posted a video called "The Future of Fstoppers" where they announced this new publishing strategy.

For the next 30 days we are going to be releasing a new video every single day. If this month is a success, we will be focusing on producing more free content in 2018.

In other words, they would try to replace high-value with frequency. The videos would still be very good (they were not posting viral content), but it wouldn't be as in-depth, or as planned as their usual content.

So... what happened? Did this work?

Well, the month of January is now over and they have published another video where they go into what happened.


This video is worth watching by itself, but let me summarize it.

The first thing they discovered was that the higher frequency boosted the engagement throughout their site and on YouTube. had the highest level of traffic, the highest time on site and the most engagement ... ever!

This is great, but when they looked into their analytics, they found that most of this boost wasn't actually coming from these new, extra videos per day, but from other content posted by their community.

So, these extra videos seemed to have boosted the activity of the site, but they didn't actually contribute to it by themselves.

This is fascinating because it gives us an 'unknown'. We don't really know if this actually worked or not, because the measurable impact was almost non-existent, while the total effect was the best ever. But it also hints to the 'trend of presence'.

I have talked about this before. Presence is often a very big part of the effect that you can have, in that, if you can be present in people's minds, you often experience a kind of spillover effect on your business as a whole.

This is not just true for content, but also everything else ... like advertising. We know that creating an ad campaign where you show up in front of people continually over time is far more effective than just having one good ad.

So, was what we saw here part of the 'trend of presence'?

As a media analyst, I don't know from the data that I have. It may simply have been coincidental in that maybe some of the community content was so good that it worked all by itself. And, if this is the case, these 30 videos they posted were basically a fail.

But this is something worth exploring further.

When we look at the financials, things start to look a lot worse.

First, consider how much extra work they have to put into creating 30 somewhat high-quality videos per month. Not just in shooting them, but also the setup, the planning, the editing, and so forth.

And this is where everything stopped working, because when Fstoppers looked at subscriber growth and actual revenue, things looked a lot worse.

In terms of new subscribers on YouTube, they only managed to double their daily subscriber growth (from about 300 per day to 600 per day). But when we look at it overall, it's hard to see any real difference.

Here is a graph for their YouTube channel of their performance over the past two years.

Mind you, doing this experiment for just a month might not be long enough to build up enough momentum, but, as you can see, there wasn't a spectacular effect.

More to the point, when Fstoppers looked at their ad revenue from YouTube, things got even more depressing. Because the total extra revenue from this entire experiment was ... $400.

Not per video, or per day... no. $400 extra revenue, in total.

And that simply isn't worth the effort. There are far better ways for Fstoppers to earn $400 more per month. For one thing, their high-quality tutorial videos are priced at $299 each. If they had instead spent their time getting just two more people to buy a tutorial video, they would have earned more money.

(BTW: They also revealed that, on YouTube, they now only earn $1 per 1,000 views).

As they said:

[Ad income from] YouTube videos are almost worthless. We might lose money with YouTube videos

This is a general trend that I see all the time. YouTube advertising by itself is terrible, and the main reason there are still so many successful YouTubers is because of their other income streams, like brand partnerships, Patreon, merchandize, affiliate income, Twitch subscriptions, etc.

But regardless of the result, I find this to be truly fascinating.

The reason I don't call this a total failure is because of the 'trend of presence'. I don't really have a problem with YouTube being a money-losing venture, if that creates other opportunities elsewhere.

In Fstoppers' case, using YouTube to create daily presence that can then be used to sell $299 tutorials (or for other YouTubers to create brand partnerships) is a worthwhile strategy.

But what I like about this the most was that they tried. I love these types of experiments, because we learn so much from them.

It does also tell us a lot about the state of advertising and the revenue potential for that as a whole. We constantly hear about Google and Facebook taking over more and more of the total ad revenue in the industry (the 'duopoly' as media execs call them), and their total revenue is growing.

But, the way they are growing their revenue is by scaling up so much that the value per ad view is almost zero. In Fstoppers' case one ad view is now only worth $0.001.

That isn't a problem for YouTube, but it is a problem for any creator.

So, the market for 'exposure-based advertising' where all you really sell are views is imploding because of its scale. It works for the largest platforms, but no single publisher will be able to make it work in the future.

If you are a publisher and you want to earn money from advertising in the future, you have to sell something other than exposure. Because the views by themselves are not worth anything.

Another experiment: What if we didn't use Facebook?

Fstoppers, however, wasn't the only publisher to experiment in January. Another publisher who did was a local TV station from Denmark, TV MIDTVEST, who decided to see what would happen if they stopped using Facebook.

They wrote about this experiment here, but since it is in Danish, I will summarize it.

First of all, let's talk about the bad news.

During this test, TV MIDTVEST experienced quite a drop in overall traffic.

This is quite a lot, and depending on how you are monetized, this could have quite a significant effect on overall ad revenue. (In this case, though, TV MIDTVEST is predominantly funded by public license fees).

Another thing that they lost was the community engagement, the activity and feedback from readers that would have made local news a topic of discussion.

For big publishers I wouldn't really worry about this, because they have too wide an audience to really use that anyway, but for a small local news publisher, this is a really big problem.

One of the key trends that we see around local news is just how important being part of a local community really is, and it's pretty clear that Facebook is a big part of that.

So, these were the bad elements. There was quite a significant drop in traffic and they lost the community feedback.

But there were also some really fascinating good elements.

For starters, the value of their traffic went up quite a bit. People spent 42% more time reading their articles, and read 12% more per session.

But that's not the only thing. There was also a really interesting change in what people were reading.

Before, with Facebook, the traffic pattern was usually very polarized around a single article, but without Facebook all the articles got a more equal share of views.

As they explain: When they usually posted 3-4 articles, one of them would get all the traffic, while the rest would not get the same level of attention. But without Facebook, all articles had an evenly distributed share of the traffic.

This is a pretty significant change in user behavior, which tells us a lot about the type of behavior people have when they use Facebook.

I have talked about this many times before. Facebook is a great channel for low-intent micro-moments, where you are just reacting to things. But this also creates an audience that has very low loyalty and isn't really reading the news as a whole.

This is exactly the effect that TV MIDTVEST found. By eliminating Facebook as a traffic source, the audience that remained read the news in a very different and more valuable way.

Another thing they found was that the traffic patterns evened out and became much more stable. With Facebook, you never really know what works or not because the algorithms have such a big influence on reach, and the audience you get is just random people. So when they removed that, they ended up with a more loyal audience that came to them in a more stable way.

As Nadia Nikolajeva, TV MIDTVEST's Head of Digital, wrote (translated):

For me, this emphasizes that Facebook is not the safe tool for getting attention that it once was. This is not surprising, due to the algorithm adjustments in recent years, but it is time that we, in the media, act on it.

Whether this is a good thing or a bad thing depends on how we talk about growth. For your regular audience and the experience that they have, this is certainly a good thing. Because it focuses the newsroom on what your real audience care about. But, in terms of reaching a new audience, things start to get a bit complicated.

They also discovered that their fear that people wouldn't be able to find them without Facebook was unfounded. People don't have a problem finding the information they need online.

They also discovered that without the constant distraction of posting on Facebook, they had more resources to think about new things.

And finally, during this test, there were no angry readers' letters sent to them. So not being on Facebook removed the animosity that people otherwise had to deal with every day.

All very interesting.

So what now?

Well, TV MIDTVEST is back on Facebook, which is probably a good decision. As I mentioned earlier, the ability for a local news site to be connected with the local community is very important.

But this whole experiment has changed their view on Facebook and what they are going to use it for. Specifically, they have stopped thinking of Facebook as a distribution channel and will now use it more for the community elements.

As Nadia writes (translated):

We have a communication challenge in telling readers and viewers that they do not get the full experience on Facebook. At the same time, we will share fewer, more focused pieces of content. And then the main focus will be on other platforms - mostly our own.

As a media analyst, I find this whole thing extremely fascinating, and Nadia did a wonderful job with it.

Of course, I have about a million questions about it. For instance, was there a difference in the 'owned' audience, or was the effect purely because they no longer had people coming in from Facebook? Was there a difference in terms of age? Time of day? ...and many other things.

But, again, what I love about this whole thing is that they tried it, just like Fstoppers tried changing how they used YouTube.

We need more publishers to conduct these experiments, because we will never learn anything if everyone just chases down the same path.

As Nadia writes:

It emphasizes that the media must develop good solutions themselves, and not only follow trends on Facebook.


Facebook is good at what it does, but we need to be good at what we do.


The Baekdal Plus Newsletter is the best way to be notified about the latest media reports, but it also comes with extra insights.

Get the newsletter

Thomas Baekdal

Founder, media analyst, author, and publisher. Follow on Twitter

"Thomas Baekdal is one of Scandinavia's most sought-after experts in the digitization of media companies. He has made ​​himself known for his analysis of how digitization has changed the way we consume media."
Swedish business magazine, Resumé


—   analytics   —


Creating a propensity model for publishers


How my focus on analytics has changed as an independent publisher


How can publishers measure trust and other editorial metrics?


A guide to analytics for independent journalists


Why producing less news leads to a boost in subscriptions


GDPR: How publishers can track things without tracking people