You know what drives me nuts? People who blame “the algorithm” when their content doesn’t take off.
Now I know that sounds like a personal attack, but I don’t mean it that way. Content creators who blame the algorithm are, for the most part, not lazy or even bad content creators. They just don’t understand how content gets seen.
This lack of understanding doesn’t come from stupidity either. Social media and search engines deliberately obfuscate the details of how algorithms rank content to stop people from gaming the system, showing their users garbage content, and causing people to leave their sites.
Your content could be videos, social media, or blog posts. Doesn’t really matter what you do, because one way or another, on the internet, you either answer to an algorithm. Social media platforms use them, as do search engines. Even your marketing emails are vetted by algorithms that decide whether your messages are “spam” or “not spam.”
In this article, I want to talk about how algorithms control what we see. I’m going to talk about what social media and search algorithms are and how they work. Then I’m going to explain how they’re ultimately driven by people’s desires, not the other way around. I’ll then conclude with some practical tips on what you can do so that you don’t feel like the algorithm has it out for you.
Let’s take the power back.
What is “the algorithm” anyway?
Regardless of the kind of content you make, there’s a lot of it on the internet. All of it clamoring for attention. The algorithms use math and science to show people what they’re likely to want to see.
The formal definition of an algorithm is “a finite sequence of rigorous instructions, typically used to solve a class of specific problems or to perform a computation.” The problem that social media platforms and search engines seek to solve is “how do I show people what they want to see, keep them on the platform, and earn more through ad revenue?”
In other words, the algorithms are programmed to keep people’s attention. Here are two concrete examples.
Google uses an algorithm to answer your questions
Google makes most of its money through advertisements on its search engine. In order to keep people coming back to Google for their results, instead of going to Bing, Yahoo, DuckDuckGo, or Reddit, the results need to be relevant. That means whenever you look something up, Google needs to be able to accurately answer your question with a relevant result, ideally within the first two or three results.
This is why Google is so full of how-to guides, lists, tutorials, and news articles. People go to Google when they have questions that they need answered. That’s also why Google so often pulls data straight from Wikipedia and other websites to display on the side or near the top.
If you’re familiar with how Google indexes pages on its site, you’ve probably heard of PageRank or SEO. PageRank is the name of one of the primary algorithms Google was using up until 2019. SEO, or search engine optimization, is how you make your website more suitable to rank well in search engines.
Google likes to promote trustworthy, well-structured, highly functional websites. That’s why Google considers things like: high-quality content, mobile-friendliness, page loading speed, and quality of links. Basically, it uses different types of data to make an educated guess as to how likely the information is to answer your questions.
But, as with many algorithms, it’s not one-size-fits-all. Different searches will yield different results. I’m just describing a general trend.
TikTok uses an algorithm to keep you entertained
On the whole other side of the spectrum, you have TikTok. While TikTok does have plenty of educational content on it, the primary purpose of it is to keep users entertained. Videos are short and fast-paced, and what you see is tailored heavily toward your interests.
TikTok is renowned for its algorithm because it constantly adjusts what it shows you to what you want to see to be entertained. For example, when you first open up the app, you’re likely to see dancing videos and sketch comedy. However, after a couple of hours of scrolling, liking, and bookmarking, you’ll start seeing much more niche content.
My wife, if she got into TikTok, would probably see a lot of baking, home renovation, Taylor Swift, and dogs.
One of our authors, Kenny, would likely see wrestling, Formula 1, gaming, and anime content.
And me, because I’m a profligate workaholic, see a lot of science and business content, but there’s a lot of millennial nostalgia mixed in there too. (It also seems to be convinced I have ADHD, despite having never received a diagnosis or even having broached the subject with a doctor).
The point is, TikTok customizes every single person’s experience based on what they are likely to spend hours and hours scrolling through. In doing so, viewers are likely to see more ads and generate more revenue for the platform.
On Google, the worst sin is to make useless content. It can be funny, but it generally has to be perceived as useful. On TikTok, the worst sin is to make boring content. It can be educational, but you better believe it can’t be dry.
Again, this is a broad generalization. Different users’ habits will trigger different results. But in my experience, you’d have to actively try to get a dull TikTok feed.
Algorithms use neural networks to show you what you’re likely to want.
Now that you know what algorithms actually do, let’s talk about how they do it. For that, we need to understand neural networks. Here’s a very quick explanation of a very complicated concept.
A lot of algorithms run on neural networks. These networks are inspired by the way the human brain works. Neural networks are made of nodes which are connected by “neurons.” Each individual neuron is weighed differently based on whether or not the network as a whole is getting the desired result.
In other words, neural networks adjust their neurons in order to increase the odds of meeting a certain objective. Based on my viewing habits, for example, TikTok notices that I spend more time watching science videos, so it strengthens the neurons associated with “liking science videos” and shows me more of them. This habit continues until the neural network is more optimized to show me things I like.
This works the other way too. TikTok has a bad habit of sending me ADHD videos. I scroll past them pretty quickly, so it weakens the neurons associated with “liking ADHD videos” and shows me less of them. This habit continues until the neural network is more optimized to not show me things I don’t like, thus leaving more room to show me what I do like.
I’ve glossed over a lot, so here’s a quick primer video if you want to go way down this rabbit hole.
For the most part, people decide what algorithms do…not the other way around.
Once you know how the technology works and how it’s applied, things start making more sense. Nearly every company that curates the massive amount of content on the internet is doing so for profit. They have a very strong incentive to make their platforms as attractive as possible. That’s why they spend an enormous amount of time and money figuring out how to show people what they want to see.
That’s the key part: algorithms show people what they want to see.
Some people imagine algorithms as some sort of Terminator-like overlord that forces content upon us whether we want it or not. But that’s just not true. Algorithms are based on people’s desires.
Now that’s not to say that every algorithm accomplishes its objective. There’s a good argument to be made that Facebook is losing its luster since many complain about the platform showing them things they don’t want to see. But when algorithms work, they work by making people want to stay online by showing them things that elicit an emotional response.
You could argue that algorithms are bad for society. Maybe we’re amusing ourselves to death. Maybe it’s a Brave New World style of dystopia. Honestly, I share a lot of those worries and I sometimes even agree with those sentiments.
But regardless of whether algorithmic content curation is a positive or negative for society, the point still remains – social media networks don’t force messages on us like Nineteen Eighty-Four. They show us what we want to see.
So you must ask yourself: am I making what people want to see?
My one exception to the rule – it’s OK to blame the algorithm when it spreads misinformation or hate.
I think it is a mistake to blame the algorithm for not sharing your content. I do not think it is a mistake to blame the algorithm when it spreads misinformation and hate.
The reason I bring this up is because people tend to roll all their grievances with “the algorithm” into a jumbled mass. These are two very separate problems.
One of the awful byproducts of creating computer systems to keep people engaged is that strong emotions increase watch time, engagement metrics, and other things algorithms pay attention to. That is to say, much like Monsters Inc., social media networks and search engines sometimes run on screams.
You can see it in Facebook’s problem with misinformation, Conspiracy Tok, and Twitter’s incessant political bickering. You can sometimes see it when Google accidentally ranks neo-Nazi books as recommendations.
Social media platforms and search engines cannot possibly remove all the misinformation or hate on their systems. But what they can do is keep making efforts to isolate and remove it from widespread distribution by tweaking their algorithms to separate beneficial engagement from harmful engagement on their platforms.
Easier said than done, yes, but given the consequences, I think it’s OK to hold big companies’ feet to the fire on this one.
Here’s what you can do instead of blaming the algorithm.
For you, as a content creator, the fact that algorithms are based on what people want to see opens up massive opportunities. You don’t have to please some soulless computer system. You have to please people instead so that they can tell the computer to put them in touch with you.
Pleasing people is no easy task, but hey, we’ve got a whole discipline that teaches us how to do it. Marketing! It’s been around for a long time and we’ve established a lot of best practices that don’t require you to know computer science.
I’ll leave you with this – here are six things you can do instead of blaming the algorithm. These will help you increase your odds of getting seen without feeling like you’re being handcuffed by a machine. You’re not going to immediately and always succeed with these tips, but if you follow them, then over time, you will likely prevail.
1. Don’t game the system.
From the mid-1990s onward, people have been trying to game algorithms. Black hat SEO operators used to pack keywords in their pages to fool Google into ranking them higher. People used to cram a million hashtags into their social media posts to get seen by more people. As long as there has been a system, someone has tried to game it.
Here’s the thing – system gaming tactics have a lifespan of a couple of years. If it’s easy, cheap, and not in the best interest of users, platforms will shut it down. Then what will you do?
Sure, spamdexing can earn you followers and viewers in the short run. But what will you do when it stops working? Will your audience really care about your content since you took shortcuts to earn their attention?
Maybe – but do you want to take the risk?
2. Improve your product-market fit.
Much better than taking a shortcut is doing the long, difficult work of achieving product-market fit. That’s a favorite marketing term of mine that describes when a product is perfectly made for a specific kind of target market. That can be anything from a book specifically made for certain readers to a YouTube video made for certain viewers.
The only reliable path to achieving product-market fit is trial and error. You have to get your ideas or products in front of people and see how they react. If they don’t love it, you’ve got to work on crafting the perfect product, service, or content by taking their feedback and making something new.
Yes, this can take forever. But taking the long way makes it really hard for someone to copy your effort and steal your competitive advantage.
3. Make content that fits in with the medium you’re on.
With content creation in particular, it helps to think not just about product-market fit, but medium-message fit. Technically, you can drop 8-minute lecture videos in landscape mode on TikTok. But it’s not going to get engagement. You’d be better off putting that on YouTube, where it might do quite well.
If you want to get big on any platform, you have to work within certain patterns. For example, we write about basically anything we want on Weird Marketing Tales, but we also make sure that it’s:
- Related to business or marketing
- Optimized for SEO
- Has a killer headline
That’s because we rely on Google searches to keep people coming to the site. If we don’t play by Google’s rules, we don’t eat. That’s why we find the intersection between things we want to do, things people will actually read, and things that people will find on Google.
4. Build a consistent brand.
Getting seen once is easy. You can upload a dance video to TikTok and see a million views, and then be forgotten the next day like nothing ever happened.
That’s why you need a brand. You need something that makes you or your business truly unique that makes people remember who you are. If you want a few concrete examples, here are some articles you can check out to see how this plays out in reality:
- Daft Punk, or: Building a Massive Brand by Anti-Marketing
- WWE vs. AEW: The Wrestling Brand Smackdown of a Lifetime
- See How Cult Classic Movies Teach Us About Branding
5. Build a community.
Want to build an empire? You don’t just need a few clicks, a few views, and a few sales. You need an active community of people who are really interested in what you’re doing in the long run.
In other words, you need a community. Building a community helps insulate you from the impacts of sudden algorithm changes. If people are actively interested in what you’re making, they will seek you out instead of waiting for your stuff to be served up to them by random chance.
Here’s a guide on how you can do this.
6. Always be launching something.
This is the last tip I’ll leave you with. If you want to be remembered, you need to always be doing something memorable. Product-market fit is how you make people take an interest in what you’re doing. Medium-message fit is making sure you get seen. Launching new products or services is how you get people to take action.
Even if you build up a massive YouTube channel or Instagram following, that won’t last forever. If you want your brand to last for years and not being beholden to algorithms, you need to create different products and services to be remembered by.
For YouTubers, this might mean launching a second channel, a Patreon account, and a Discord server. For educators, this might mean making books and courses. In traditional businesses, this might simply mean making more new products for people to see.
Everything in business has a lifecycle. The creators who last know this and are always evolving over time.
Do this and you won’t have to stress out so much about the algorithm. You will gradually accrue the freedom you’re looking for, which allows you to try unusual things.
The algorithms we interact with on a day-to-day basis do their best to serve us what we want, and do a pretty good job doing so. Don’t let technical slang confuse you. Ultimately, you’re answering to people not a computer.
