Illustration of colorful books on a shelf against a dark background.
A stylized image of a robot sibyl

Table of Contents

The ancient Romans had a secret weapon. Hidden away in the Temple of Jupiter on the Capitoline Hill was a collection of oracular pronouncements written in Greek hexametric verse known as the Sibylline Books. In a world that seemed to be governed by unseen and capricious forces, they offered much-needed certainty. The Romans took comfort in the fact that, when things seemed to be going wrong, the Books would show them what they had to do in order to make things better, whether that was establishing a festival in honor of the goddess Flora or burying Greeks and Gauls alive under the forum

For modern content creators, it can feel like we’re in the same boat as the Romans. Algorithms rule our digital lives, and they can feel as fickle and inscrutable as the residents of Mt. Olympus. A lovingly crafted piece that took months of work can go unnoticed, while a lazy list that required a tenth of the effort gets millions of views. Faced with this reality, it’s understandable why many content creators are eager to find ways to tilt the scales in their favor. And just as the Romans hoped that their problems could be solved with the correct rituals, modern content creators often assume that all they need to do is rigidly follow a list of prescribed rules and the algorithm will grace them with its favor. The problem with this approach is that algorithms are not monolithic and unchanging entities; on the contrary, they’re always evolving.

The new oracles: algorithms as arbiters of online content

There’s no denying the fact that algorithms have become the unseen masters of much of the Internet. With each passing minute, a staggering amount of content is uploaded to each of the platforms where we live our digital lives. Without some form of sorting, users would quickly become buried under an eternal avalanche of new material. After all, most of us are on these platforms because we’re looking for something in particular, whether it’s ways to make paella, build advice for Baldur’s Gate III, or political commentary that reinforces our prejudices. But at the same time, we’re always on the lookout for new material even if we expect it to follow the same well-trodden paths that we’re familiar with.

In theory, algorithms address this problem by helping platforms give people the content they want to see.  But, as they say, the devil is in the details. For starters, the algorithm needs to know what kind of content you want to see. But how does it go about figuring that out?

The shifting sands: how YouTube's algorithm transformed

While algorithms shape content discovery across numerous platforms, one of the most prominent examples that impacts countless creators is YouTube's system. By tracing its evolution over the years, we can gain insights into the changing factors that govern what videos get promoted and recommended to viewers.

It’s also one where the algorithm plays an enormously significant role in deciding what people actually watch—in 2018, YouTube’s Chief Product Officer claimed that 70 percent of what people watched on the platform was governed by the algorithm.

Lots of views = high quality?

According to a piece by Hootsuite’s Stacey McLachlan and Paige Cooper, in the early days of YouTube (c. 2005-2011), the algorithm gravitated toward content that received lots of clicks or views. But once creators figured this out, they started trying to game the system with clickbait.

Experimenting with new metrics

Around 2012, YouTube’s approach shifted to encompass the time spent watching each video as well as time spent on YouTube itself. This transformed the platform from being one dominated by TikTok-like shorts to one that hosted longer-form content. And as Mark Bergen (author of Like, Comment, Subscribe: Inside YouTube’s Chaotic Rise to World Domination) said in an interview with The Verge, this shift also gave rise to new genres of content such as beauty blogging and video game Let’s Plays. But once again, people tried gaming the system. We’ve also seen cases where four minutes of content has been awkwardly stretched into a ten-minute video!

“Be sure smash that like button and leave a comment”

Between 2015-2016, McLachlan and Cooper say that YouTube started emphasizing direct engagement such as likes and comments. And from 2016 onward, there has been renewed focus on combatting misinformation as well as efforts to demonetize content that’s deemed unfriendly to advertisers (that’s opened an entirely different can of worms that we’ll be discussing in a later post, so stay tuned!).

According to McLachlan and Cooper, YouTube’s modern algorithm incorporates aspects of its previous incarnations. It considers the length of time people spend watching a video, but it also looks at direct engagement such as likes/dislikes and comments. It also tracks how people respond to a video when it’s recommended to them. If they ignore it or click “not interested,” it’s less likely to be recommended. It also uses contextual information such as a video’s metadata and your own search history to try to decide whether something is likely to be useful to you.

Prioritizing people: creating content for your audience

But one point that YouTube has consistently emphasized throughout is that creators should focus on making content for their audience rather than the algorithm. Some creators seem to focus on the algorithm to the exclusion of everything else. But as we’ve seen, this kind of thinking often leads to negative results. Relying on dubious tactics like clickbait to draw people in will make it hard for you to build a solid audience that will stick with you over time.

At Newstex, we firmly believe that creators need to focus on making good content first and foremost. Instead of trying to check off some imagined list of criteria that will please the algorithm, make sure you’re providing value to your audience by giving them information that’s truly relevant to their interests and needs. It should also be trustworthy and make use of reputable sources. Solid organization is a must, too. For more on this subject, check out my article "How to create people-first content: producing content for humans not machines."  

The bottom line is that creators can’t rely on a modern-day version of the Sibylline Books to tell us how to propitiate the algorithm. Instead, it’s better to focus on creating high-quality content that’s brings value to your audience. If people feel that your content is useful to them, they will come back for more. The algorithm will pick up on that and start sending more people your way.