What is cohort analysis and why are we choosing to cover this topic? The answer is simple: because it is a fairly easy and rigorously applied procedure, which few grasp apply wisely. The idea is quite simple; build, test, measure, iterate, test, measure, iterate, test, measure. Doing it every day, aren’t you?

The Wikipedia definition of “cohort analysis” goes as such: “A cohort study or panel study is a form of longitudinal study (a type of observational study) used in medicine, social science, actuarial science, and ecology. A cohort is a group of people who share a common characteristic or experience within a defined period (e.g., are born, leave school, lose their job, are exposed to a drug or a vaccine, etc.). Thus a group of people who were born on a day or in a particular period, say 1948, form a birth cohort.”

Do images speak more than words? Here are a couple of examples of studies based on cohort analysis:

- smoking

- social media

And we are asking again, you are doing that every day no — group your users, followers, customers based on a common characteristic, interest, desire?

Then, why is this valuable at all and why should startups consider creating such a study? One reason why the cohort analysis is valuable is because it helps to separate growth metrics from engagement metrics. Strategically, measuring growth is more than just creating a list that tracks increases in numbers of customers, products sold, or even number of loses / wins on the competitors’ side. These are the results of growth. Instead,  growth metrics must be designed to focus efforts on initiatives that will drive growth.

When it comes to engagement rises and downs, the way in which we can track the profile of a well-engaged user is by looking at a series of simple factors: number of page views by a single user, the number of visiting sessions during a week / month, relatively long vs relatively short sessions and so on. (examples based on the user engagement of a website)

The two (growth and engagement) are important because growth can easily mask engagement problems. It is, thus, not simple enough to track your brands online, see the amount of times they are mentioned on Twitter and Facebook, but more importantly understand why and what’s the feeling they give in designated cases. For instance, if your Facebook page has over 2,000 likes, the engagement will show up pretty strong, giving you a feeling of secure growth and social media importance.

In reality, nevertheless, it may be that people stop being engaged after a couple of weeks on the service. Why? For instance, your service does not bring anything new to the plate, the novelty wore off, too many posts to actively watch from too many actively posting Facebook brands. However, the lack of activity of the users that have been a fan for some time is being hidden by upcoming new ones. And when you receive up to 10 more likes a day, there are enough “wanna be” fans being added to the service that the lack of engagement does not show up immediately. Or, for some marketers, it doe s not seem …that important.

We started thinking of how to actively asses our activities and we’re coming up with a secure plan after reading a very useful study directed by Robert J Moore, the CEO and co-founder of RJMetrics on Twitter’s activity and user engagement. The results are fascinating and the entire article can be fully read here. Another great resource is Dave McClure’s presentation for Seedcamp 2009. As a summary, we mention:

• Twitter’s user growth is no longer accelerating.
• Over 14% of users don’t have a single follower, and over 75% of users have 10 or fewer followers.
• 38% of users have never sent a single tweet, and over 75% of users have sent fewer than 10 tweets.
• 1 in 4 registered users tweets in any given month.
• Once a user has tweeted once, there is a 65% chance that they will tweet again.
• If someone is still tweeting in their second week as a user, it is extremely likely that they will remain on Twitter as a long-term user.

Unlike Robert, we’re on the case with our friends at Ubervu. We are encouraging everyone to adopt the startup metrics methodology and within that methodology, making sure they are looking at cohorts of users, not just all of your users in the entire aggregate. There is a reason Caesar’s quote still applies to the present day; “Divide and conquer.”

Looking forward to your views on the subject and if you have any other suggestions on how to better tackle this topic, be free to drop a line. We’re sure curios to hear your thoughts on the matter.