Analyzing Threads as A Product Data Scientist
Introduction
Meta, the powerhouse behind numerous social media platforms, has recently unleashed its latest creation: Threads, an app designed for sharing text updates and participating in public conversations. It has been gaining enormous popularity, with 100M users in 5 days!!! However, amidst the buzz surrounding its rapid user acquisition, skeptics have emerged, labeling Threads as a mere imitation of Twitter. Notably, industry giant Elon Musk, who acquired Twitter for $44 billion, has threatened legal action against Meta. Will we witness a clash of the tech titans, both verbally and perhaps even physically? The anticipation is palpable.
While the potential clash between Elon Musk and Mark Zuckerberg captivates the imagination, let us turn our attention to a more analytical perspective. In this article, we will explore Threads through the lens of product analytics, seeking to unravel the metrics that determine its success, growth, and user engagement. This is the best way for data scientists to develop product sense and understand a new trend.
What is Threads?
TLDR: Threads is an app similar to Twitter and is compatible with Interoperable Networks in the future.
Threads is an app from Meta and is built by the Instagram team for sharing with text. While platforms like Instagram primarily focus on visual media such as images and videos, Threads fills the gap by providing a dedicated space for users to share ideas, engage in discussions, or simply express random thoughts through plain text. Moreover, Threads seamlessly integrated with Instagram, allowing users to easily sign up using their existing Instagram accounts. This brilliant move has resulted in a surge of sign-ups, as the one-button setup has eliminated barriers and encouraged user adoption. Moreover, it partially solves the cold start problem for a new platform. This means that upon signing up, users already have a foundation of connections and threads to explore.
When examining how Meta introduces Threads, it becomes apparent that the app’s main objectives are centered around:
- Sharing text updates and joining public conversations
- Enable positive, productive conversations
- Compatible with Interoperable Networks
Interoperable Networks is no doubt the most attractive selling point for Threads. This concept entails compatibility with ActivityPub, an open standard protocol for decentralized social networking. ActivityPub enables different social media platforms and applications to communicate with each other and share information in a federated manner. For users, it means we can follow and interact with people from other platforms and vice versa. I believe that is what makes Threads different from Twitter.
However, as the Interoperable Networks is still under development I won’t go further about it at this point. In the next section, I will focus on how Threads achieve the first objective, as well as explore the relevant metrics that can be used to evaluate its success.
How to measure the success
Before diving into the measurement, we should always define the objective of the product and its features.
Although the functionality is indeed very similar to Twitter, I will discuss it in detail for those who may be unfamiliar with it.
Sharing text updates and joining public conversations
Threads addresses a longstanding issue on Instagram: the need to share images every time users want to express themselves. This limitation often restricts conversations or topics that don’t have readily available related photos. Additionally, as a follower, I probably will focus on the image and the video more. With Threads, the focus shifts to text, enabling users to share text updates and engage in public conversations more seamlessly.
Another important difference is how threads are pushed to the feed. On Instagram, unless users explicitly share a post, it can go unnoticed if their friends reply to an account they don’t follow. This can hinder participation in conversations. With Threads, you can see and respond to the replies from people you are following and also the conversation they engage in. What’s more, you can repost a conversation and initiate new discussions from different perspectives.
Therefore, we should design measurements/experiments related to the usage of text threads and replies.
Product-market fit
Analytics at Meta actually have published a Medium post discussing the most important question we need to answer: “Does this product have product-market fit?” The article defines what is product-market fit and why it is essential to a new product/ feature.
In addition to assessing the usage of text-only threads, let’s explore the metrics that evaluate the three key measures outlined in the post:
- Sustainable growth
- Stable retention
- Deep engagement
Sustainable Growth
The first metric to evaluate growth is the number of new users every day. Given that Threads draws users from the existing Instagram user base, it would be also valuable to examine the percentage of Instagram users converting to Threads daily and the number of customers who joined Instagram because of Threads.
Apart from the number of new users, Meta defined other ‘states’ to describe users in the article:
Retained: Users were active on the product both yesterday and today.
Churned: Users were active on the product yesterday but inactive today.
Resurrected: Users were inactive on the product yesterday but active today.
Stale: Users were inactive on the product both yesterday and today.
We can then derive the Daily Active Users(DAU) and Daily Net Growth from these states:
Daily Active Users = Daily New + Daily Resurrected + Daily Retained
Daily Net Growth = Daily New + Daily Resurrected — Daily Churned
Stable Retention
Retention is definitely the most important measure for a new product because it indicates the sustainability of the product. Do they want to stay or do they revert to familiar products after several days?
Having Instagram users from the beginning has its pros and cons. Threads can get new users in a very short time. However, I doubt the retention rate would be high in this situation. Especially since the product is still in an initial version.
The retention rate defined by Meta: The percentage of people that are still active on the product a specific time period after their first action date.
Deep engagement
Engagement level can be assessed from the time users spend on the app per day. With more time spent, the higher the engagement level would be and potentially higher retention.
Metrics for the objective
Returning to our main objective of understanding how users engage with text-only threads on Threads, several metrics can be employed:
- Volume of text-only threads By finding the average number of text-only threads per day, we can see how much more content users post on Threads that they might not share on Instagram.
- Volume of reposts Having the average number of reposts with or without quote per day can help to understand how engaged the users are to join the public conversations.
- Volume of replies engagement As the replies are posted to the feed in Threads, we can look into the average number of replies and the level of engagement (likes, reposts, and replies) received by those replies
Metrics compared to Instagram
Given that Threads acquires users directly from Instagram, it becomes crucial to compare the usage patterns of both platforms. By examining the metrics mentioned earlier, such as the volume of replies and time spent, we can conduct a side-by-side analysis to uncover insights.
One aspect worth exploring is the usage of Instagram users before and after the launch of Threads. This analysis allows us to determine if there are any notable effects on Instagram’s usage following the introduction of Threads. We can assess whether there is a negative impact, such as reduced usage on Instagram, potentially indicating a shift in user behavior towards Threads. Conversely, we can also investigate whether Threads has a positive effect, possibly reigniting engagement among certain Instagram users who were previously less active.
By carefully comparing and contrasting the usage metrics of Instagram and Threads, we can gain valuable insights into the impact of Threads on user behavior and identify any potential synergies or divergences between the two platforms. This comparative analysis will aid in understanding the dynamics of the user base and help evaluate the success of Threads within the broader context of social media engagement.
More Features?
While Threads is having a successful start, it lacks a lot of features such as hashtag and trending topic functionality. Undoubtedly, users have high expectations and a desire for more robust capabilities within the app. However, introducing new features poses challenges for Threads, particularly in terms of conducting A/B testing effectively. The influx of new users contributes to a novelty effect, which can complicate the process of gathering accurate user feedback and discerning genuine feature demands.
The novelty effect, in the context of a new product, refers to the initial excitement, interest, and increased attention that consumers have when they encounter a new product for the first time.
As it is still in the early stages and hasn’t gained significant traction or established a solid user base, it may be beneficial to focus on refining the core functionality and addressing any initial issues before adding new features. For example, please add a web version!
Furthermore, it is worth noting that there is a considerable number of threads within the app itself discussing desired features for the future. Leveraging the power of Natural Language Processing (NLP), it becomes feasible to analyze these threads and identify the most frequently requested features. This data-driven approach enables Meta to gather valuable insights into users’ preferences and prioritize the development of features that align with their expectations.
Conclusion
In summary, Threads, the new app by Meta, has gained significant traction in the social media realm. From a product analytics perspective, we have explored some key metrics.
Sustainable growth, stable retention, and deep engagement metrics provide insights into product-market fit. By monitoring new user numbers, daily active users, and retention rates, we can gauge its user base stability. Additionally, metrics such as time spent on the app, volume of text-only threads, reposts, and reply engagement indicate the depth of user engagement and the fulfillment of its objective to facilitate public conversations.
In summary, Threads shows promise in the competitive social media landscape. As Threads evolves, tracking the metrics and adapting to user feedback will be essential for its long-term success in the dynamic world of social media.
If you found this article insightful, don’t forget to clap and leave a comment to let me know your thoughts. Stay connected by following me for future articles on various topics in Data Science. Thank you for your support!
Reference
Introducing Threads: A New Way to Share With Text | Meta (fb.com)
My Articles
The project I did to land my first data scientist job | by Leo Leung | May, 2023 | Medium