An enhanced version of the analysis of how product features impact retention | by Paul Levchuk

A easy and highly effective strategy to segmenting your product options into Core, Energy, and Informal.

Within the previous post, I confirmed you an easy approach how one can measure product options retention.

After making use of the evaluation we obtained the desk with retention per characteristic like this (sorted by lowering [Average % returned users]):

• On one hand, we obtained useful details about which product options influence product retention essentially the most.
• Alternatively, we don’t have at hand figures about what number of customers used these options so we are able to’t be assured that these figures are dependable.

Let’s add [# users] and contemplate this desk another time.

Now we are able to simply spot a problem: for instance, the primary two product options (feature27, feature34) with the very best [% returned users] have fairly a modest quantity by way of [# users].

Truly, this drawback is most typical in plenty of analyses that I’ve seen. Usually analyst brings a fairly attention-grabbing high quality measure however it’s not backed up by amount measure. Because of this, a few of our choices could be at the very least suboptimal and at most simply fallacious.

So how can we repair this difficulty?

Let’s mix each metrics (qualitative and quantitative) into one chart. Essentially the most appropriate technique to do this can be a scatter plot:

• let’s placed on the X axis the metric [% users], it’s our amount metric that measures the reputation of a product characteristic.
• let’s placed on the Y axis the metric [% returned users], it is our high quality metric that measures the worth of a product characteristic.

The ensuing chart may appear like this:

It appears that evidently to this point it’s relatively onerous to make any significant conclusions from the chart.

What can we do to enhance the chart readability?

Let’s apply the 50/80 percentile rule from the earlier put up.

Truly, after making use of 2 thresholds for [% users] and [% returned users] we’ll get 9 clusters.

Clustered product options scatter plot may appear like this:

By including percentile thresholds to the chart we are able to now distinguish such product characteristic clusters:

• Core: [% users] > 80 pctl, [% returned users] > 80 pctl
• Power1: [% users] > 80 pctl, [% returned users] in [50, 80] pctl
• Power2: [% users] in [50, 80] pctl, [% returned users] > 80 pctl
• Casual1: [% users] in [50, 80] pctl, [% returned users] in [50, 80] pctl
• Casual2: [% users] in [50, 80] pctl, [% returned users] < 50 pctl
• Casual3: [% users] < 50 pctl, [% returned users] in [50, 80] pctl
• Set-up: [% users] > 80 pctl, [% returned users] < 50 pctl
• Area of interest: [% users] < 50 pctl, [% returned users] > 80 pctl
• Others: [% users] < 50 pctl, [% returned users] < 50 pctl

Let’s focus on a bit of bit every of clusters.

Core options are the true core of your product. These options are utilized by plenty of customers, and what’s extra vital customers return again to proceed utilizing these options. As a rule, there may very well be a really small variety of such options (2–3 options).

Energy options are workhorses of your product. These options mixed with core options ship about 80% of the common worth that your product creates. A few of the energy options (Power1) are as standard as core options however deliver much less worth to customers. Different energy options (Power2) deliver as a lot worth as core options however are much less standard. As a rule, there may very well be 3–5 options in every energy cluster.

Informal options are options which are used occasionally. Additionally they deliver some worth to customers however for essentially the most half, they’re supporting options.

Set-up options are a novel subset of options which are designed to arrange a product for the next handy utilization. Loads of customers use them, however as common, it occurs one time, on the onboarding section.

Area of interest options are a really particular subset of options that would deliver an infinite quantity of worth however this worth is perceived by a restricted variety of customers.

Now we’re prepared to match the outcomes of this balanced strategy to the outcomes from the previous post:

As we are able to see on the high of the record there are some Area of interest options.

For positive we are able to attempt to enhance their adoption and transfer them from Area of interest to Power2 and even Core cluster. For a few of them, it’s doable, for others — it’s not. However the principle level right here is not to easily assume that any characteristic with excessive retention is a core characteristic.

Additionally, please notice that some options can transfer from cluster to cluster over time. There may very well be totally different causes for this: new person acquisition efforts, UX modifications in options, person base maturing, and so forth.

Lastly, let’s group options into clusters and calculate cluster centroids:

There are a number of vital insights right here:

• Core + Energy clusters account for under ~20% of all product options.
• Others cluster accounts for 27% of all options and on the similar time, it serves solely 8.7% of customers.
• Area of interest options are utilized by simply 11.3% of customers and on the similar time have the very best retention (even greater than for the Core cluster).

Within the next post, I’ll speak about one other perspective on the characteristic retention definition.

P.S. There’s a better way to cluster product features primarily based on the MCC coefficient.