How World History Encyclopedia Gets 72M Pageviews Per Year: Q/A with CEO Jan van der Crabben

Website Planet had the pleasure to interview Jan van der Crabben, CEO and Founder of the World History Encyclopedia, the world’s most-read history encyclopedia worldwide.

Jan described the struggles, mistakes, and success factors that made him reach 72 million pageviews and 40,000 newsletter subscribers with a website started during his train commutes with zero experience.

What’s your story and what led you to start your website?

Several threads came together and led to the project being started.

First, I grew up around history. My parents were keen travelers and as a child often found myself playing among the ruins of a Greek temple on an Aegean island. In 1992, my father bought the game Sid Meier’s Civilization for me, of which I became a dedicated fan. When I was studying at university, I created a modification for the game, called The Ancient Mediterranean Mod, which became one of the most popular modifications for Civilization III & IV. I launched a companion website for the modification, added some historical information, and at some point realized that more people were visiting the history pages than the game pages.

After university, I joined Creative Assembly as a game designer. I first worked on Empire: Total War, a historical strategy game set in the 18th century. I was responsible for a significant amount of historical research for the game (down to estimating the historical GDP output of different regions of Europe to make the game as accurate as possible). During that research, I noticed that there was no reliable, free, and easy-to-understand history resource on the internet.

There was Wikipedia, but especially back then it contained a fair amount of questionable information, particularly on less popular subjects. Several university websites offered historical information, but they were hard to navigate and –above all– hard to digest into anything useful. Finally, there were websites with paywalls, such as Britannica. That is the gap that WHE was built to fill a few years later.

Around the same time, I was thinking about information represented as a network. My first idea was about language and translation, creating a website that would link meanings to words in different languages, creating a sort of networked multi-language dictionary and thesaurus. I started programming it; even had a name for it (Arbolingo, I still like it) and a domain, too. I abandoned the project, but its codebase became the timeline that was the first feature of WHE.

What linked all of those things came to me through my research on Napoleon: Total War, the next game I worked on after Empire was released. In school, in Germany, I learned about Napoleon and his attempted conquest of Europe. While I lived in the United States, I learned in school about the Louisiana Purchase, where France sold vast swathes of land to the United States. I had also learned about Simón Bolívar, the great hero who freed South America from Spain (or so I was told).

During my research on Napoleon, I quickly figured out that history, too, is a network or web where everything interconnects. Napoleon sold Louisiana to the United States to finance his military campaigns in Europe, and Simón Bolívar seized an opportunity when Napoleon conquered Spain and thereby cut off the Spanish Empire’s head. I was stunned at how these links were not taught in school (at least not to me, despite having had excellent history teachers)!

I merged this revelation with my plans of a networked dictionary for Arbolingo, and I applied the network concept to history: All of history is interconnected, and I wanted to create a website that can represent that, programmatically to avoid having to manually link everything.

The first timeline did just that and it quickly expanded to include definitions, articles, images, and much more. In 2008 I started programming this new website on a small laptop while commuting by train, and in August 2009 the very first version of the website went live, with the title “Ancient History Encyclopedia”.

What went into building the first version of your site?

The website originally began as “Ancient History Encyclopedia” and featured a basic database of timelines. Each timeline entry was linked to a specific subject through tags. The idea behind this linking was that users could navigate from one timeline to another by clicking on a related entry. For instance, if you were looking at a timeline about Greece, and you saw an entry about Alexander the Great, you could click on the link to take you to the Alexander the Great timeline, which would contain the same entry and others. The original code that powered the timeline on the website is still in use today.

Once that was working, I thought we could add more than just timelines, but also definitions of each of the topics, and display the timeline on the side. These definitions would become the articles that now form the core of the website. Then we needed images, of course, which became yet another type of content. As with the original vision, every type of content is linked through tags, so that different pieces of content will show up on pages that are related to it.

This is what the first version of the site looked like:

As you can see, definitions were much shorter than they are now, and there was more emphasis on the timeline. Once the website was live, I contacted the first writers.

My grandmother had recently passed away and I had inherited some money. I decided to invest $1000 of that money to pay for the first 100 submissions to the encyclopedia. You could call it the initial –and only– investment in the organization, as we’ve not had any other investment since.

I then searched the internet for people who wrote high-quality and engaging articles about ancient history and reached out to them. A few people answered and were willing to write for the website, including a lady who only went by “writer873” and whose name I still do not know (her articles have since been replaced), the history author Brian Haughton, and a certain Joshua J. Mark, a history professor at Marist College who ended up writing the vast majority of the initial 100 first articles.

Joshua got paid his $700-odd dollars and bought a stack of Harry Potter Lego for his daughter back in 2009. When I lived in Montreal and visited Josh at his house in upstate New York in 2021, he handed me a pile of boxes of Lego for my son who was now at that age (and very much into both Harry Potter and Lego!), so in a way my initial investment came full circle.

What challenges have you faced during the years, and how have you overcome them?

The biggest challenge when starting was to get the website off the ground. We had no funding (apart from my initial $1000) and no marketing. We just had a website and a few people who loved to write about history. For the first three years or so, traffic only grew very slowly, with a few thousand people visiting the site. That did not deter us, though: We were doing this because we wanted to do it, not because we wanted to make money (there weren’t even any ads on the site at the time).

Being persistent like this paid off: In 2013, Google launched the “Hummingbird” core algorithm update, which rewarded high-quality content. Almost overnight, our traffic jumped from about 5,000 visitors per day to around 30,000 visitors per day. That gave us a big morale boost and we felt invigorated to improve the website more, write more articles, and go more in-depth.

In 2015, another Google update gave us another big boost, bringing our daily visitors to around 70,000. That was the moment when we realized that this could become employment for some of us, and in 2016 we employed the first staff writer, Mark. I was only the fourth employee in 2018… a whole 10 years after I first started the project!

We used this formula for all of our employees: Everyone who works for us has volunteered for us first. We’ve never hired anyone without knowing that they love World History Encyclopedia enough to do it for free. That way, we have the most engaged and committed team I have ever worked with. Everyone knows what to do, how to do it, and mainly, why they are doing it. It’s a project driven by passion, not paychecks.

What tools are in your current tech stack, and how do they help you run your online project?

From the very beginning, we built World History Encyclopedia on a custom-made CMS that I initially programmed in that train on my commute to my day job. This was a big success factor, as the CMS was built so that it did exactly what we needed it to do, and it was also built with SEO in mind from the very beginning. The way the articles are formatted and presented is optimized for SEO, and our editor interface provides many features that help the editorial team optimize our content for SEO purposes.

Another thing that was important from the start was the realization that not every visitor needed to see the live website. We cached our pages very heavily for those users who are not logged in (which is about 99% of them) so that it would load extremely quickly. On top of that, we later added Cloudflare, again with heavy caching including the raw HTML of our articles, so that in most cases, visitors will have the website served entirely from the edge cache on Cloudflare, not from our server. This allows us to run a very high-traffic website with a small server setup.

Finally, we love automation. On the site itself, all cross-linking between articles is automated, including links inside the article texts. We’ve built a system that will not only automatically insert links to relevant articles on new content, but it will also update old articles and insert links to relevant new content. This allows our editors to only worry about the article they are working on, saving them immense amounts of time, while at the same time giving us a significant SEO benefit.

Since you started, what helped you grow your audience, and how long did it take to get constant traffic?

We got a few things right and a few things wrong when we launched our website. None of us had any experience in marketing or even running a business when we started, so we had to learn that on the job. Also, we had no budget, which certainly did not help.

As I mentioned before, we only relied on organic search traffic for many years. Even though that worked out fine in the end, it was a very slow process and we could have done so much better had we marketed our website from the beginning. We could have started by reaching out to scholars and teachers letting them know about the existence of our site. We could have placed advertisements in historical publications. We did none of that, and if I were to start over again, I would not forget about the marketing!

We are still mostly reliant on organic search, which drives the bulk of our traffic. On the one hand, that’s great because it’s free traffic, but on the other hand, it makes us very reliant on Google’s algorithm. Overall, Google’s algorithm changes have been net beneficial for us, but there were several years where an algorithm change caused our traffic to drop and as a result our revenue to go down. These were tough years.

The same is true for social media: It’s very fickle. It was around 2013 when our Facebook page grew significantly and sent us a large amount of traffic. Then a few years later, Facebook changed its algorithm and our website only received a trickle of traffic from Facebook. Algorithms can be extremely beneficial for growing an online presence, but they are dangerous to rely on.

Now we focus very much on building our brand, reaching out to teachers who will send their students to our website, and engaging our community to build a more loyal following. We’ve now got newsletters in three languages with over 40,000 subscribers as well as thousands of members who support us financially every month. All of this has reduced our reliance on Google and social media a little, which is good, but there’s a lot more work to be done.

What monetization techniques have you tried over the years?

As a non-profit website, we use three monetization methods, each with its pros and cons.

Our primary revenue source is programmatic advertising on our website. This can generate a lot of money for a large website, but it does require at least 500,000 page views a month to generate any significant amount of revenue. The downside is that it is very cyclical: In summer, we see fewer students visit our website, which means traffic is down, and so is advertising revenue. In 2023, the economic downturn has caused online publishers to earn about 30% less from advertising, us included. It’s easy to make money with ads, but it’s very cyclical.

The second revenue source is membership subscriptions. These are people who pay us regularly in order to support our non-profit organization. As a thank you, they receive an ad-free experience on the website as well as a members’ magazine with behind-the-scenes information. Membership revenue is highly valuable to us as it is less cyclical; we earn the same amount of money in summer as we earn in winter, for example. Also, even though membership sign-ups decrease in an economic downturn, revenue does not drop as drastically as advertising revenue does. At the same time, building a subscriber base takes a lot more time.

The final revenue source is fundraising. Every once in a while we start individual projects for which we solicit donations from our audience. This is helpful to kick-start something, but it is less useful to sustain the day-to-day operations. We have also found that donations are strongly affected by the economy; in 2023 we collected a much lower sum than in the previous years, for example.

I am extremely grateful for all of the memberships and donations that we receive. This helps us stay afloat even if the economy is going badly.

What resources have helped you the most?

There are a few books that helped me in building an organization with a culture that inspires the people working for it.

The first book is Re-Work by Jason Fried and David Heinemeier Hansson, the founders of Basecamp (formerly 37 Signals). Their mantra is that a company is not something to be sold, but something to be lived. While most start-ups have the goal to grow quickly and get sold for a profit, there is another way: A company that grows slowly, to provide a great service to both its clients and its employees; a company for the humans who work there, not for investors. I took many ideas from this book when I was thinking about how World History Encyclopedia would grow.

The second book is The Infinite Game by Simon Sinek. In this book, Sinek argues that truly successful businesses that inspire both their employees and their customers are not looking to optimize their next quarterly results, but instead are those who think ahead 10, 20, or even more years. These are companies that are built to change the world, and to leave a legacy. This resonated with our team because we always thought of our work as our legacy to the world.

How are you doing today and what does the future look like?

This year has been tough. Since January, we have been losing money and it seems like we’re only now turning that cover and returning to profitability. Even though we are a non-profit, we need to make money; otherwise, we’d go bust like any other company.

This loss was caused by several things coming together: In late 2022, we were aware of strong inflation, so we gave our existing employees a larger raise than usual (we thought we could afford it). Then in 2023, we hired more people and planned to make a small loss in that year, again because we thought we could afford it. However, 2023 turned out to be a bad year for online advertising, so this combined with our increased expenditures meant that things did not at all go as planned.

The entire team has worked very hard this year to reduce expenses, raise brand awareness, and improve our programmatic advertising stack. Some of our partners were instrumental in increasing revenue, and a special shout-out goes to Assertive Yield for helping us earn a lot more money from advertising. We onboarded their technology in May and saw a very significant increase in session and pageview revenue.

At the same time, we’ve focused very much on our members. Our membership magazine is becoming rather popular and we are making a conscious effort to show our members how we operate and who is doing what. We want to build more personal relationships between those who support us and our team; appear less like a faceless corporation and more like a team of dedicated people with a passion for history, which is what we truly are. There’s a lot more to be done, but we’re slowly getting there!