This short article will answer the question: “How do social media algorithms work?”. Understanding modern social media algorithms is key in creating and performing with your brand, business or influencer page. Within this article we will try to give you a thourough understanding of how the algorithms work. This information should be generally applicable to most social media platforms.
The basics of Social Media algorithms
Yes! The algorithms. The great “mystery” of social media platforms. The algorithm on all social media platforms has one job: generate watch time & engagement.
Why? Because longer user attention retention gives the platform more time and chances to advertise to the user.
In reality this means that since most platforms like Youtube, TikTok and Instagram have abandoned the chronological feed. The algorithm is trying to show users the content they like! With a huge amount of interests, hobbies and personalities it is impossible to say what content people will like. However, to understand how the social media algorithms work. All you need to know is how the platforms measure what content your viewers like.
In order of importance these are the most important metrics; Watch time, Engagement, Views, and Impressions. From here on we will call these attributions a posts points. For the sake of simplicity we will have all of these points “weigh” the same. However, platforms may tailor their algorithm to favor business or goals or media formats. In reality we often find that watch time has a greater impact then for example impressions. Let me tell you why these points are important.
The Spotlight Metaphor
When we understand the metrics that the social media algorithm takes track of it is now important to see how it uses these attributed points.
In order to explain this we like to use a “spotlight” metaphor. With a little bit of imagination this will easily represent how the algorithm picks posts to show users.
To start, envision a large amount of people looked at from above. This makes them look like a large amount of dots on a flat surface. (Let’s call this step one)
When an online post is first posted, the algorithm will use a combination of retargeting and interests to showcase the content to a group of people most likely to represent the intended viewer. Usually this group consists of a part of your current following, people that have reacted/watched to your content recently and a few people that consist of a look-a-like audience to your audience members.
At the beginning stages your spotlight is small, as it is competing against other pieces of content that are also being tested by the algorithm in the same audience. (Step 2) At this point the algorithm is checking for the effectiveness of the content by weighing the following:
- Relevancy: Social media algorithms prioritize content that is relevant to the user’s interests and preferences. For example, if a user frequently interacts with posts related to fitness, they are more likely to see posts related to that topic on their feed.
- Engagement: The more engagement a post receives, such as likes, comments, shares, and saves, the higher it will rank on users’ feeds. This is because engagement is a strong indicator of relevance and interest.
- Recency: Social media algorithms also prioritize fresh and recent content, as opposed to older posts that may no longer be relevant or engaging.
- Source: The credibility and authority of the account or page that posted the content also plays a role in social media algorithms. Verified accounts, pages with a large following, and pages with high engagement rates are more likely to have their content shown to users.
- Advertiser preferences: Social media platforms also prioritize content from advertisers, based on their targeting preferences and budget.
(these are not to be taken into account for organic post performance, however as a business might provide interesting opportunities.)
When your post starts collecting these attributed points the proverbial spotlight on your content will increase. As the algorithm has determined it to be a likely that your post will match with more people relevant to the current group of interactions. The larger the amount of people that interact or view the post, the larger your spotlight will become.
This exponential growth of views can lead to a post going “viral” which posts end up going viral or performing well is always unknown however, sticking to optimizing the core principles mentioned above will highly increase the likelihood of virality and boost overall posting performance.
You might be tempted to artificially boost these metrics by means of paid likes, views or alternative. My colleague Borre has recently created a insightfull video explaining the dangers of fake engagement. You can find it here: (https://newage.media/video-never-buy-fake-followers/) We never recommend fake or purchased engagement or views as it will harm your profile in the long run.
Why do the Algorithms work this Way?
The first social media algorithm was introduced by Facebook in 2009. At the time, Facebook was struggling to keep up with the massive amount of content being shared on its platform. The introduction of the algorithm helped Facebook prioritize content and show users only the most relevant and engaging posts. Other social media platforms followed suit, and today, algorithms are an integral part of social media platforms such as Instagram, Twitter, and YouTube.
Machine learning is a key aspect of social media algorithms. Machine learning algorithms are used to analyze and categorize the vast amount of data generated by social media users. These algorithms use a combination of user behavior, engagement metrics, and content analysis to determine what content is most relevant to a specific user. Machine learning is also used to identify and remove spam and fake accounts, which can negatively impact engagement rates.
User behavior plays a crucial role in social media algorithms. Every time a user interacts with content on a social media platform, it sends signals to the algorithm about what content is most relevant and engaging. This feedback loop helps the algorithm learn and adapt to user behavior, ultimately leading to a more personalized experience for each user.
While social media algorithms are designed to prioritize content that is most relevant and engaging, they are also influenced by advertisers. Social media platforms generate revenue by selling ad space to businesses, and the algorithms are designed to prioritize content from advertisers based on their targeting preferences and budget. This means that even if a post is highly engaging, it may not appear on a user’s feed if it is not sponsored by an advertiser. We have seen a multitude of platforms move further away from the more traditional chronological feed and decreasing the impact of posting recency.
You have probably had a 7 year old youtube video recommended to you out of the blue, making the video gain enormous traction disregarding the “bad” recency metric.
Understanding Machine Learning = Understanding the algortihms
All social media algorithms are always evolving and learning form user data. Furthermore, the algorithm might get biased instructions to suit the platforms goals. For example the current push towards vertical format video content. In order to better understand social media algorithms we will take a quick look at the basics of machine learning.
Machine learning is a type of artificial intelligence that enables computers to learn and improve from experience without being explicitly programmed. In other words, machine learning algorithms analyze data to identify patterns and relationships, which they use to make predictions or decisions.
There are several types of machine learning algorithms, including supervised learning, unsupervised learning, and reinforcement learning. Supervised learning involves training the algorithm on labeled data, while unsupervised learning involves training on unlabeled data to identify patterns on its own. Reinforcement learning involves a trial-and-error approach, where the algorithm receives rewards for correct decisions and penalties for incorrect ones.
Machine learning is used in a wide range of applications, including image recognition, speech recognition, and natural language processing. It is also used in recommendation systems, such as those used by Netflix and Amazon to suggest movies and products based on a user’s previous behavior.
While machine learning has many benefits, it also presents challenges. One of the biggest challenges is the need for large amounts of high-quality data to train the algorithm. Additionally, there is always the risk of bias in the data or the algorithm itself, which can lead to unintended consequences.
Overall, machine learning is a powerful tool that has the potential to transform the way we live and work. Its ability to learn from experience and improve over time makes it a valuable tool for a wide variety of applications.
Conclusion
We hope this article helped shine some light on the inner workings of most social media alortihms. Would you like to know more, or explore our services? Take a look at New Age Media and it’s services! Thank you for reading.
