Like most websites, we use own and third party cookies for analytical purposes and to show you personalised advertising or based on your browsing habits.
Generally, the information does not directly identify you, but can provide you with a more personalised web experience. Because we respect your right to privacy, you may choose not to allow us to use certain cookies. However, blocking some types of cookies may affect your experience on the site and the services we can offer.
For more information, you can read our cookie policy.
How the Twitter (X) Algorithm Works in 2023
Twitter (X) has been continuously going through changes and updates this year, the algorithm being one of them. Twitter has tapped into that “For You” timeline style, so you can find recommended content similar to your interests, and what you have interacted with on the platform.
The process in how Twitter chooses which content appears in your timeline is complex and depends on multiple different factors and actions.
Liking, retweeting, and sharing still have the power to make tweets go viral, but just like many other social media platforms, the algorithm is the powerhouse of the network. Therefore, it’s important to understand how it works, so you can better rank your content and reach a larger audience.
What is The Twitter (X) Algorithm?
The Twitter algorithm is a series of rules and factors that determine which content appears on each user’s timelines.
The Twitter (X) algorithm controls what content we see and how we see it based on components such as location, virality, personal interests, recency, and whether or not the post contains media.
Knowing the basics of this ever-changing algorithm can help you tweet to your fullest potential, so let’s break this down even further.
How the Twitter Algorithm Works in 2023
The Twitter algorithm was created to personalize each user’s feed, and recommend content that is similar to what each user interacts with on the platform.
Twitter uses a set of core features that collects information from tweets and user history to answer questions like, “What is the probability you will interact with this content in the future?” or, “What are the trending tweets within Communities?”
The network has also revealed its algorithm code on GitHub.
Twitter breaks down this process into three steps:
Candidate Sourcing
In this step, Twitter retrieves recent and relevant Tweets from a variety of recommendation sources, for both In-Network sources (accounts you follow) and Out-of-Network sources (accounts you don’t follow).
Therefore, the goal is that each “For You” page displays a mix of content from both In-Network and Out-of-Network sources. Whereas the Out-of-Network tweets are more complex, identifying million of tweets from accounts you don’t follow, the In-Network source organizes recent tweets from accounts you follow.
Ranking
In this second process, each Tweet is ranked by using a machine-learning model, regardless of if it’s an In-Network or Out-of-Network source.
Twitter hasn’t released exactly how the ranking mechanism works, but according to the network, Twitter awards each tweet a score, which represents the probabilityof an engagement.
Heuristics, Filters, and Product Features
After the ranking process explained above, Heuristics, Filters, and Product Features take over. Aside from the ranking process, Twitter also applies factors such as heuristics, filters, and product features to refine the tweet selection. The factors that Twitter uses to enable diverse content
These also filter out Tweets from users you’ve blocked, NSFW (not safe for work) content, and Tweets you’ve already seen.
They include:
- Visibility Filtering: You will not see Tweets from accounts you block or mute.
- Author Diversity: Avoid too many consecutive Tweets from a single author.
- Content Balance: Balance of In-Network and Out-of-Network Tweets.
- Feedback-based Fatigue: Ranks the tweet lower if it has received negative feedback.
- Social Proof: Ensure someone you follow has engaged with the Tweet or follows the Tweet’s author.
- Conversations: Reply through the thread of the original tweet.
- Edited Tweets: Update your tweets with new ones to replace the old ones.
These three processes make up the “For You” that we see today, however, there are also a series of ranking signals. By understanding these, you can continue to improve your content and stay relevant on the network.
Decoding the Twitter Recommendation Algorithm
The Twitter Recommendation Algorithm has a lot to do with the portion of Twitter regarding explore, search, and ads sections.
Twitter (X) Algorithm Ranking Signals
The new algorithm emphasizes certain types of content so it is important to include these in your Twitter strategy to show up more frequently on user’s feeds.
Relevance
The most important factor on Twitter depends on users’ behavioral history, meaning which content each user engages in. Thus, those tweets that generate a lot of replies, likes, and comments are considered more relevant.
Recency
Another factor that impacts the algorithm is recency. This includes tweets including trending topics, current events, etc.
Variety
Twitter aims to diversify your timeline, including content that sparks open conversations and shows different perspectives. Therefore, creating content that includes multiple points of view can help increase your content visibility.
Multi-Media
Generally, content that includes images and videos attracts more engagement and are favored by the algorithm. With that being said, content that is considered “spammy” will not be prioritized.
There has been a huge push for more videos on Twitter (X) recently. In fact, where you used to see ‘You Might Like’ you will now see ‘Popular Videos.’ This is not too shocking given how popular video content has become on many platforms due to the rise of TikTok.
Twitter Metrics Prioritized By Twitter
- Tweets that have a high number of shares and retweets
- Tweets that gain lots of likes and replies
- Tweets containing media
- Tweets with a CTA to follow the author
- Tweets not reported as spam
- Tweets with one relevant hashtag
- Tweets without an external link
- Tweets from reputable accounts
Tips to Rank Your Twitter Content
Tweet at Optimal Times
One way to reach more users is to post content at the “best times”, meaning when a larger percentage of your followers are active. This can boost your content’s visibility and engagement. But, how can you find these best times?
With Metricool, you can find the best hours to post for your specific audience. Inside the Planning section, view the calendar that includes a gradient scale and percentages of when your followers are more active. Then, schedule your content for these times and let the tool auto-publish content for you.
Consistency and Audience Engagement
The algorithm favors consistency on the network, so it’s essential to stay active and continue to engage with your audience. Aim to post at least once a day, and try to include captions that provoke engagement such as asking questions, providing resources, answering commonly asked questions, etc.
Replies are also being emphasized in the algorithm as of recently. Tweets with more replies are proving to do better on the app. This is in the hope that it sparks some more engagement and urges more users to converse on the app via tweet threads.
Utilize Keywords and Hashtags
Hashtags are very much still a useful tool on Twitter. Including hashtags is important to help your content reach a larger audience, especially in specifc topics. With Metricool, you can track hashtags on Twitter and measure their impact.
Analyze your Content’s Performance
An essential step in your strategy should always include analyzing your content. This will help you understand what is working well, and what you should focus on to improve. Twitter has native analytics that you can utilize, or you can use Metricool to view your account metrics.
The Twitter Algorithm is complex and has proved to be ever-changing so it is important to stay up to date! For a full tutorial on Twitter’s new X, be sure to click the button below: