A cohort is a group of users who experience the product value at about the same time.

• Look at users for a month. The first month is Month Zero. All the users that experience the product in Month Zero is your cohort.
• Track the behaviour of each key retention metric for that cohort of users.

A PM can look at data per cohort to see what numbers improved over time. This can highlight the value within your product that attracts and keeps your users.

The Activation Curve shows the number of users activated over a specific period of time.

Many early-stage companies have bad retention at the start. They can get customers in but haven't figured out the mechanics to keep them around. When the metrics start to improve over time, you generally have the concept of product-market fit.

• Have enough tooling to track signups on a cohort basis, such as Google Analytics, Google Analytics 360, Segment, Mixpanel, or Amplitude.
• Know your proxy metric. What action is the user performing? (consuming content, a download, a daily active user?). Your proxy metric will be similar to your main goal metric, such as monthly active subscribers. 

Focus on new users and the initial experiences they have with your product. 

To create a great first impression:

• Run experiments on user onboarding.
• Personalise your sales process.
• Incorporate feedback loops.

Target better users. Look at the channels users come through (targeted and organic). Measure each channel to find your ideal customer.

Kill features that don't provide enough core value. Use Firebase experiments to see how a rollout or a build is impacting retention metrics.

Change your model or your pricing.