Social Media

How Do Social Media Algorithms Work?

Jul 3, 2023 | By Chris Jacks

Social media algorithms are the backbone of social networks and organize the incredible volume of content according to the needs of users.

That’s great for users, but brands and influencers face challenges in trying to understand algorithms and gain a strategic advantage. To make things more difficult, each platform has its own ever-changing algorithm quirks and ranking signals that mean the difference between visibility and fading into obscurity.

Knowing the idiosyncrasies of the algorithms for each platform, then optimizing content for them, is the key to seeing real results from social media marketing.

What Are Social Media Algorithms?

An algorithm is a mathematical set of rules that specifies how a set of data behaves. On social media platforms, algorithms organize and categorize search results and ads to make them discoverable for users.

About 60% of the world’s population uses social media, amounting to about 4.9 billion users in 2023.[1] That’s a lot of users searching for relevant content – and even more content shared to platforms. Algorithms bring order to chaos.

The inner workings of each platform’s algorithms are proprietary, but marketers have some advantages in understanding the ranking factors and general best practices for keeping content visible on social media.

Platform Algorithms

Algorithms vary, but there are some evident ranking signals on Facebook, YouTube, Pinterest, LinkedIn, Twitter, Instagram, and TikTok.

Facebook

Facebook’s goal with the platform is customer engagement. Truly designed for the social aspects, the platform prioritizes posts from friends and family or locally relevant information over business posts.

The ranking signals for Facebook are:

  • Connections: This is content from the people and pages a user follows and interacts with.
  • Content type: Users get content based on the content they already view. For example, users who watch videos get more suggested videos.
  • Engagement level: Popular posts that already get a lot of engagement are more likely to be boosted by the algorithm, especially among users with previous interactions.
  • Content quality: The general category of ranking signals is described as meaningful, informative, accurate, and authentic by the platform.

Facebook can be challenging because the algorithm has been updated many times over the years. Organic posts in particular are less likely to be seen without solid strategy.

Organic reach on social media is in a decline in general, however. Facebook’s average reach for an organic post is down over 5% with an engagement rate of just 0.25% — less for accounts with over 100k followers.[2]

Paid Facebook content is ranked separately but still prioritizes relevance, engagement, and response. It’s important for marketers to build connections and increase engagement by replying to comments or questions, use Facebook Stories, and create live videos.

Another challenge with the Facebook algorithm is the imperfect fact-checking system. External content needs to be credible and not misleading or “clickbait.” Still, sometimes authoritative content is flagged as misinformation.

YouTube

YouTube is a popular option for influencers and some brands, but it has a stunning volume of content each day. Here are the ranking signals:

  • Video performance: Popular videos get more visibility in the feed, which is measured by likes, dislikes, click-through rate, and view duration.
  • Watch history: YouTube recommends content that’s similar to what users have watched before.
  • Context: Topically related videos or videos that are watched together tend to come up in the suggested videos for users.

YouTube is more about presenting new content or accounts to users than showing them content from accounts they subscribe to. For example, users who watch videos of fitness tutorials are more likely to see suggestions from other fitness professionals, not videos from the fitness trainers they already follow.

Pinterest

Pinterest works a little differently from the rest of the social media world. There are only two known ranking signals:

  • Website quality and ownership: Pinterest determines the quality of a website based on the popularity of the Pins that link to it and prioritizes the content from the website itself.
  • Engagement levels: For both individual Pins and the Pinner’s account, engagement levels are a key ranking signal.

Pinterest also uses a guided search method with data collected from past content interactions to encourage new links. For example, users who viewed Pins about wedding themes are going to see more Pins about wedding planning and inspiration.

This algorithm can be more difficult to decipher, but the benefit of being interest-based is that it shows the user things they’re already searching for. The content distribution is controlled as well, so it means content is more likely to be engaging for users.

LinkedIn

LinkedIn is a more niche social media platform, but it holds plenty of value for some influencers and brands. The known LinkedIn ranking signals include:

  • Post quality: The algorithm sorts content when its posted to flag it as spam, low quality, or high quality. Naturally, the high-quality content gets priority.
  • Early engagement: Early engagement is a secondary quality test that the algorithm uses to determine if content should be more visible to larger audiences.
  • Connections: The algorithm shows new content to users with previous connections, but pages, groups, and hashtags are used to determine which new users should see it based on their interest in a topic.

Twitter

When Twitter first emerged in 2006, it ranked posts according to the timeline. It was more important to get the day and time right rather than the content, but the new algorithm change refocused the ranking signals:

  • User interactions: Accounts that users interact with frequently are more likely to show in users’ feeds and recommendations.
  • Recency: The recency of posts affects what shows up in the trending topics or “What’s Happening” feed.
  • Location: Location determines what Trends show up for a user.
  • Current popularity: The engagement and activity surrounding a topic or trend, particularly with users in the same network, make content more discoverable.

Instagram

Instagram is unique in that it’s owned by Meta – the parent company of Facebook – yet aiming to compete with TikTok. It has a blend of the two platforms with ranking signals like:

  • Relationships: Users are more likely to see content from other users or brands they follow and interact with. This means brands and influencers should encourage and respond to followers as much as possible.
  • Interests: Like Facebook, Instagram aligns recommended content with the interests the users’ already show, such as animals, art, and fitness.
  • Relevance: Relevance is based on factors like topical or trending content.
  • Popularity: The level of engagement and speed of interaction with a post or account signal popularity, which is important for getting it on the Explore page.

Many of these signals align with best practices, but Instagram is known for frequent algorithm updates. Adam Mosseri, the head of Instagram, is fairly transparent in sharing the platform’s core focus, however, including the priority given to videos.

1hul0euFFUYCQ GPETkPkLbR0Q1jkySQq6hyYDIwbVHKvODeL43Jk2aoT6pWigtBVd9Sm8PdLfkPq4RUu4wQyRZPo6jyOIJXEqXmQ1Zvj1A igQO6B6rROaqWcVx7q

Source: Twitter

TikTok

TikTok is a new frontier for a lot of marketers, as the platform has only risen to popularity in recent years. TikTok’s known ranking signals include:

  • Previous interactions: Content is more visible based on signals like accounts followed or hidden or content that users have engaged with previously.
  • Discover tab behavior: This factor analyzes content characteristics like captions, sounds, effects, and trending topics.
  • Location and language: Users see more content from their own country or their own language.
  • Trends: Capitalizing on trends in effects, sounds, or topics makes content more discoverable as it aligns with what users are searching for.
  • Native features: Native features like sounds, effects, and text overlay are common in TikTok content and make it more discoverable.

Unlike other platforms, follower count doesn’t matter on TikTok. The algorithm is designed to bring new content to the masses, rather than showing users content from people they already like and follow.

XEfzxLz0rinc1hft8UL3nP9 Q1eKlcQ1DBYVQwArphJ20qGEqhwPS

Source: TikTok

Best Practices for Social Media Algorithms

Though each platform has its own ranking signals, there are some general best practices that apply to all social media platforms:

Post Relevant, High-Quality Content

Relevant, high-quality content is more discoverable across all platforms, regardless of algorithms. The intent of social media algorithms is to show users the content they want to see, and that’s rarely content that’s poor quality or unrelated to the audience interests.

Avoid Clickbait

Clickbait has soured a lot of the social media user base in the early days, leading platforms to train their algorithms to avoid content that’s misleading or spammy. Avoid headlines, captions, or hashtags that are spammy or designed to mislead the user into clicking.

Follow Trends

Trending topics keep people engaged, so social media algorithms are looking to provide more of that content. While brands and influencers shouldn’t try to keep up with every single trend, anything that is relevant to the brand messaging is worth capitalizing on.

Know the Best Post Times

Algorithms often consider recency and early engagement, which means it’s important to know when the audience is most likely to be online. Every platform and audience differs, but some audience research and experimentation can reveal the ideal times for the majority of the relevant users.

Experiment

There’s a lot of science to algorithms, but it takes some experimentation to get it right. Brands and influencers need to try new ideas, content, times, and approaches to see what works and refine social media strategy over time.

Algorithms Are Part of Social Media Marketing

Sometimes illuminating, sometimes frustrating, and always changing, social media algorithms are integral to success with social media marketing. Brands and influencers should learn the ins and outs of each social platform’s algorithms and experiment to find the right formula.

Sources:

[1] https://www.smartinsights.com/social-media-marketing/social-media-strategy/new-global-social-media-research/

[2] https://www.forbes.com/sites/forbesagencycouncil/2023/04/27/are-the-days-of-organic-social-media-reach-over/#:~:text=On%20Facebook%2C%20which%20remains%20the,also%20experiencing%20the%20same%20issue

author_photo
ABOUT THE AUTHOR

Chris Jacks is an influencer marketing professional with over a decade of experience in the digital marketing sphere. As the Director of Growth Strategy, Chris oversees and drives strategic initiatives to fuel business expansion. With a keen eye for market trends and opportunities, Chris develops comprehensive growth plans and aligns business objectives across cross-functional teams. With a strong focus on crafting impactful, ROI-driven influencer campaigns across multiple sectors, Chris utilizes his expertise to enhance market positioning and maximize results.

linkedin_chris_jacks
Brands we’ve worked with
target
adidas
honda
coke
wb
mtv
oreo
ebay
ricola
mcdonalds
microsoft
nfl
Have an upcoming objective?

Our award winning strategy team is on standby.

Let's connect arrow