Every major social app, including Facebook, Instagram, LinkedIn, Reddit, and Snapchat, has a activity or News feed or equivalent feature. A user’s subscriptions often include user-generated content such as photos, profile changes, and birthdays, as well as “professional” content such as news stories from media organizations and posts by celebrities. Many activity-feed-based products also rank this content, highlighting updates from the sources or types of content a user engages with most, each time they visit their feed.
The specific metric used to track and improve engagement with your product depends on the use case. Engagement in a activity feed environment, for example, is very different from engagement in an environment where content is consumed via a browse section, or through “stories.” In this document, we will focus primarily on activity feeds, which highlight information based on users’ subscriptions and, in some cases, on AI.
Engagement within a activity feed takes many forms. A user might simply scroll through multiple stories, or they might interact with a post by clicking on a link and reading the story, reacting with a “like” or comment, or perhaps even engaging in a conversation with other users. They might also take a picture of a piece of content, edit or comment on it, and share it with their friends.
WHY ENGAGEMENT MATTERS
Because engagement reflects whether users truly love a product and find value from it, it is the earliest indicator of product-market fit. It is also the most important driver of both retention and sustainable growth. Therefore, a deep understanding of the drivers of engagement is critical to product success.
In this series, we will explore and provide guidance on the various ways you can improve engagement: making content production easier for users, helping them connect with the content that is most relevant to them, showing them the right content in the right order, making sure they can consume content easily on any device and network, and enabling interaction with your product. Each post will also include “key metrics to track,” designed to help you know whether product is truly successful.
PRODUCTION-CONSUMPTION SYSTEMS
Before we begin, we must understand the production-consumption systems that drive engagement with social products. In these systems, users produce content (or inventory) that others consume, and feedback from those consumers in the form of upvotes, comments, reactions, etc. motivates users to produce more content (see Figure 1).
Content created is inventory, how an individual user’s inventory scales with their number of connections and the total content creation in the system. Open follow systems, like Twitter and Instagram tend to have more inventory per user than reciprocal follow systems, like Facebook. In most inventory-rich environment, like Facebook and Reddit, the order in which stories appear in users’ activity feeds is determined by a ranking algorithm.
Engagement is fed by this production-consumption loop and, in most products, is ultimately measured by “time spent” metrics such as number of sessions, number of views, and total time spent with the product. Therefore, increasing one or more of these metrics will generally be a top-line goal for your Engagement team. In the next several posts, we will explore deeply the different parts of the engagement loop.
Follow us on Medium for weekly updates.
This work is a product of Sequoia Capital’s Data Science team. Chandra Narayanan and Hem Wadhar wrote this post. We would like to thank Jamie Cuffe and Jenny Wang for their contributions to this post. Please email data-science@sequoiacap.com with questions, comments and other feedback.
