As promised, Twitter has released its algorithm code on GitHub, which of course has sent marketers into a frenzy to analyse how they can get the best results.
As explained via Twitter:
“On GitHub, you’ll find two new repositories (main repo, ml repo) containing the source code for many parts of Twitter, including our recommendations algorithm, which controls the Tweets you see on the For You timeline. For this release, we aimed for the highest possible degree of transparency, while excluding any code that would compromise user safety and privacy or the ability to protect our platform from bad actors, including undermining our efforts at combating child sexual exploitation and manipulation.”
If you aren’t sure exactly what you’re looking at by reading the code (you’re not alone), don’t fear. Plenty of pros have taken to Twitter itself to summarise what we have learnt.
A Tweet from Peter Yang outlines the weighting behind certain actions that see your Tweets prioritised:
- Likes 30x
- Retweets 20x
- Twitter Blue 2-4x
- Trusted circle 3x
- Images/videos 2x
- Replies 1x
And also, what you shouldn’t do to not have your Tweets suppressed:
- URL only
- No text
So, as expected, likes and retweets are the main engagement points you want if you are looking to get expanded reach. Other the other hand, URL-only or no-text posts should be a big no-no in your strategy to avoid getting shushed.
Overall, Twitter’s release of its tweet ranking algorithm data provides high-level insights into how the platform’s algorithms work. Twitter has also provided a simpler explanation of the system to help users better understand how their timeline is curated every time they open the app.
Want to take your social media game to the next level? Find out more about our Social Media Marketing Intensive Program which is designed to rapidly upskill you in social media marketing.