In this new series, Did You Know?, I invite you to decipher commonly used words in digital analytics vocabulary. No overly long definitions, no complex demonstrations, but just real, concrete explanations to teach concept and application. The first term? Cohort. A cohort is a visitor segmentation on a specific event at a specific time.

The segmentation tool is essential in order to define targets and create action plans. Contrary to appearances, there are several strategies for use depending on the objective required, both retention and acquisition.

ninja interogationOK, so I can use segmentation to isolate a behaviour, then apply a standard action each time it is repeated?

Yes, but you can go further…
Segmenting visits on abandoned baskets, for example, will initiate a marketing campaign. But in addition to this corrective action, I can also try a preventive action. I will study this segment to find out if there are any strong correlations (source, route, action, profile, etc.). In this case, I can act in advance to eliminate or mitigate any possible cause for the abandonment.

For example, you can see when segmenting the visits that the majority of abandoned baskets occur after checking the delivery charges.

ninja exclamationBy segmenting the visits, do I get the full picture, enabling me to make a reliable diagnosis and therefore resolve most problems?

Only in part…
Visitor segmentation* is complementary: it makes it possible to create standard profiles which you can act upon (to find out more, download Stéphanie Legrand’s white paper). If I segment the criteria that interest me, I can therefore act on this population.

For example, visitors who have checked delivery charges and then abandoned their basket mainly come from Southern Europe, via a promotional email campaign, and were already customers.

ninja interogationCan I find out about the current behaviour of visitors who have already taken part in a special event?

Yes, the cohort** is obtained by using visitor segmentation.

This is interesting in order to keep a log of the population exposed to a specific event at a specific time, then to study its subsequent behaviour, by comparing it to the non-exposed population.
The cohort highlights valuable insights and makes it possible, among other things, to assess the final impact of certain campaigns from the past which have increased the ROI. Cohorts can also be used to track customer loyalty development campaigns, retention, calls-to-action and incentives.

For example, I am going to study customers who have abandoned a basket during the sales and follow them over time, through other events.

ninja interogationI want to be proactive as well as reactive. Can I anticipate and launch a campaign in advance?

Yes, exactly. To answer this need, we will use the persona***, which relates more to acquisition or risk prevention.

In medicine, you study a cohort (population of sick people) to define the at-risk population (= persona) among whom you can raise awareness and towards whom you can target prevention efforts.

After having created my cohort, I run all the analyses and make a note of the main characteristics from which a profile will be created.

For example, Italian or Spanish visitors, those who use Firefox, have already ordered, watched a video, left a comment, checked delivery charges, or else (when available) a socio-demographic profile, or a CRM profile.

Together these elements form a standard profile for my cohort.
So I create a visitor segment based on these characteristics: This is the persona.
By acting on visitors with this profile, I can reduce the risk of abandonment.

Compared to cohorts, creating a persona allows me to take one step further: I presume that people having the same characteristics will most probably behave similarly in the future. So by acting on the visitors with this profile I will be able to reduce the risk of abandonment in the future.

ninja interogationI’m worried that it’s too complex and, above all, very time-consuming?


No, the segment manager app is very intuitive! You simply need to select “visitors” and “period”, then just select the elements by dragging and dropping.
The time spent will be productive, as it is dedicated to analysis.




Visitor segmentation on a specific event at a specific time is a cohort.
** This cohort has distinctive characteristics.
*** Visitor segmentation on these characteristics is a persona.

You tell us…

We’re interested in your comments and experiences!


Knowledge Manager After On-the-job and Off-the-job training in purchasing and management at Carrefour, and sales training at Procter & Gamble, JM evolved in the mass retail sector in top management positions for large hypermarket, central purchasing and logistics groups, with an expatriate experience in Africa as a Central Director. In late 1995, JM created an Internet start-up company and after three years (late 1998) he joined Alain Llorens and the AT Internet team where he took up his position in sales, and was also at the heart of the pioneering adventure in Web analytics. At 55 and after almost 13 years seniority in the company, JM has been Knowledge Manager since 2009.

Comments are closed.