Feed Ranking For Content Apps Apxor

How your home feed can aid in improving your user activation

Acquiring new users is an expensive affair, and if your app’s retention is a leaky bucket, then you are just wasting your money on users who might churn soon. 

A user who has activated quickly on your app has more chances to be retained than one who hasn’t yet activated. Hence, understanding your app’s user activation point has a deep impact on improving retention. And it is critical for the app to activate its users as quickly as possible.

User Activation Explained 

In simple terms, activation happens when a user finds value in your app, understands how to use it and your app matches the user’s expectations with which he/she has installed the app. However, there might be different paths that lead to the point where your users have found value in your app. So which one amongst this should you choose as your app’s activation? The best activation for your app is the one which works positively for your KPIs. Ideal activation path is the one which leads to the best retention (indicating the effectiveness of the path) and also which is easily discoverable, so that majority of users follow this path (indicating the efficiency of the path). 

Optimal Activation for Content Apps

In content apps, Home feed is the first interaction with new users. There can be other assisting features like Search, Recommendations, Subscribed Feed etc., which bring the relevant feed to a user. But these paths might demand more interactions, and choosing these paths for activation can reduce the activation efficiency considerably. 

“It is critical for content apps to activate users on their home feed itself, as it ensures both the effectiveness and efficiency of the activation.”

Best content apps activate their users on their home feed. However, at this stage there won’t be any information about the new user to begin with. Hence, arriving to the right mix of home feed which can lead to quick activation is the biggest challenge for content apps. 

 

Step by Step Method to Identify the Best Activation Point

1. Understanding Activation with Correlations and Identifying Pseudo Activation

The first few sessions of a new user are critical in understanding which functionalities are having positive correlation with retention. For a content app, the number of stories read by new user should have a positive correlation with retention. If that is not the case, then it is a red flag for the app. Some apps might find that reading just 1 story has a positive correlation with retention, while others might find a higher number. Any other correlations you might find other than reading stories, will not be the best activation for your content app. It is imperative to consider only the correlations with reading stories.

Retention

Looking at the above graph, let’s define that your app has identified a pseudo activation which is – Users who read at least 6 articles on Day 0 are coming back on Day 1.

2. Measure the Effectiveness and Efficiency of the Pseudo Activation

Let’s assume that the activity of reading 6 stories is showing 73% Day 1 retention. Then it is good activation as it shows the effectiveness of the activation. However, the bigger question now is ‘How efficient is this activation?’

If you look at the data, only 35% of the new users are reading at least 6 articles and the remaining 65% are dropping off before reading 6 stories. How do you activate more users in this scenario? The secret to success here is Optimising your home feed.

3. Optimising Home Feed – Finding the Right Mix for Your Home Feed

Analysing user paths leading to these 6 articles will give a perspective of how they are finding them. Are they finding them straight away on your home feed without scrolling down much? If so, your first-time user’s feed is well designed. But if users are taking longer paths to find what they want to read, then you will lose users before activation.

Consider the first few articles that provided the best FTUE and made users come back the next day. Analyse these articles for the commonality across multiple dimensions.  That common factor could be:

     Content Category – Most commented / Most liked / Trending / Breaking / Viral / Politics / Bollywood etc.,  

     Position where they are displayed – Banner position / Second page etc.,

     How they are displayed – Size of cards / Image associated with content / Title font etc.,

     any other factor depending on the app.

As shown in the image below, the insight is that at least two articles which are trending (having comments / shares / likes) are being read as initial articles by these retained users. So, now can you identify that the actual activation point for new users as “Reading at least two trending articles on day 0” or “Reading at least two Breaking news articles” instead of our pseudo activation of reading 6 stories.

Attribute Retention4. Arriving at Optimal Activation:

Once you identify the consumption patterns of first time users that are impactful, next logical step would be to iterate on home feed and arrive at a better activation. People who read 2 trending articles or 2 Breaking news article are more likely to come to app again. Once you have ensured the effectiveness of the activation, ensure that they are more discoverable (Position, Display etc.,). This in turn shall ensure good number of users are getting activated. There are other ways to enhance discoverability of apps that have complicated UI, let’s discuss that it in another post.

Now that you have found your actual activation point, you should design your “New User Home Feed” by giving more weightage to working stories to activate more users within the drop off threshold. Activating users without any deviations inside the app will boost activation efficiency and impact your short term KPIs like Week 1 retention tremendously.

 

If your retention is a leaky bucket, your app cannot survive after it hits network saturation. It is important to fix the leaks in retention. Reach out to us if you are interested in finding out the effectiveness and efficiency of your apps activation. 

Data Scientist @ APXOR

Post a Comment