Twitter’s new timeline algorithm recently launched and has been rolled out over the last few months. This new update has many users and companies wondering how this affects them. The new algorithm is now enabled by default across the entire social network and rollouts started as early as March.
With daily active users growth up by 14% in Q1 2017, it seems this may be an update that will continue Twitter’s growth and user engagement.
In this post, we’ll clear up some of the differences in the timeline feed from now and from before the algorithm update, as well as how this might impact you.
Before the Update
To start, before this update, and as far back as 10 years ago when Twitter launched, your timelin was made up of tweets from everyone you followed in reverse-chronological order. When you logged in to your account, your feed was made up of the most recent tweets from those you followed. This means that there was no ranking of importance between tweets in your timeline, it was just shown in order since your last visit.
In 2014, Twitter began making some tweaks to the timeline feed. At this point, they began to implement features that would recommend tweets, accounts, and topics to you. This was the first time that content form those you didn’t follow was introduced to your account.
Soon after, they launched “while you were away,” which was a recap of the top tweets you didn’t get a chance to see while you were offline, from accounts you followed.
In 2016, the famous #RIPTwitter hashtag became viral when users negatively reacted to Twitter’s new algorithm that reordered their timelines based. However, Twitter obviously didn’t die. Their timeline feature actually increased how often users tweeted and retweeted.
The New Update
Fast forward to 2017 and now, with the tweet ranking algorithm, the tweets that enter your stream will be different. In a nutshell, this update will allow twitter to predict tweets that you’re more likely to interact with and will keep those on the top. Your feed will no longer contain tweets in reverse-chronological order, sprinkled with content from other accounts you don’t follow, but will contain content ranked in order of how the algorithm predicts best fits your profile.
According to Twitter, “Tweets you are likely to care about most will show up first in your timeline.” Essentially, their system determines what they think you are most “likely to care about,” and it becomes what you’re most likely to see when you log in.
How is Twitter able to determine which tweets are relevant to you?
There a lot of factors that make up the ranking of the tweets. Here’s how Twitter explained it in their blog:
"In order to predict whether a particular Tweet would be engaging to you, our models consider characteristics (or features) of:
- The Tweet itself: its recency, presence of media cards (image or video), total interactions (e.g. number of Retweets or likes)
- The Tweet’s author: your past interactions with this author, the strength of your connection to them, the origin of your relationship
- You: Tweets you found engaging in the past, how often and how heavily you use Twitter”
By using machine learning and artificial intelligence alongside their new algorithm update, Twitter expects that the relevancy of the tweets will have a positive impact on engagement. This evolution can also lead to a growth of their platform. Mark Cuban agrees, as he recently commented: "I started buying Twitter recently because I think they finally got their act together with artificial intelligence,” a comment which had an impact of almost 4% in their stock increase.
What the Future May Hold
Twitter says that it’s not going to stop here. They have a lot of plans for deep learning and artificial intelligence and how that may benefit their platform:
"Using deep learning as the central modeling component in timeline ranking already gives great results in a production setting. However, the other main reason why Twitter made this change is to open the door to further improvements. In the field of machine learning, deep learning and the development of AI-related work these last few years has led to an unprecedented (and ongoing) burgeoning of new ideas and algorithms. We believe it is critical to let our ML-powered products potentially benefit from the full range of what is out there. We can do so by using an extensible platform that natively supports deep learning.
In the long run, this should help us better understand every individual Tweet and interactions on Twitter in order to be more relevant for everyone, in real-time."