Cohort studies applied to marketing and e-commerce
Tags
Según la Gestalt, “el todo es más que la suma de las partes”. Si aceptamos el mantra de la escuela alemana, estamos abocados a aceptar que la construcción estética contiene una gramática propia, unos elementos básicos que en su conjunción forman un todo con entidad propia, pero disoluble y flexible.
In a world where there is fierce competition in the market for visits, clicks, sessions, subscribes and sales, it is increasingly important to optimise the investment and effort expended to obtain them in order to achieve the best margin for each one.
It’s nothing new, data has always been there, it’s just that it’s increasingly taken into account and considered as a tool to improve processes, ratios and business indicators.
The value of data: vanity vs. results
In addition to “vanity data”, data to feed the ego (likes on Facebook or Instagram followers) which is often public data to which a part of the effort is always dedicated, we find “actionable data”, which are those that have a direct impact on the business (the number of Twitter followers or visits to a page do not usually have an impact on the balance sheet).
When it comes to actionable data, good data management starts with good data selection, a process that must be done jointly by all those involved in the management of the company, often with the help of an information analysis consultant.
Good management
of actionable data starts with good data selection.
Interested?
The SMART principles should be adhered to, which will help to make a good selection of indicators: each indicator should be Specific, Measurable, Achievable, Relevant & Time-bound.
Typical examples include conversion rate, average purchase value, recurrence, recency, or lifetime value. By combining certain values, the information obtained in raw data can be greatly enriched: cross-referencing impressions with users, purchases with time, etc. As quite a visual example of cross-reporting we have the cohort study.
The cohort study, what is it and why is it important?
A cohort is defined as a group of individuals in a population who share a certain characteristic: iPhone or Android users, smokers or non-smokers, patients following treatment A or treatment B.
In marketing and e-commerce, time cohort studies are useful, i.e. classifying users according to when they perform their first action (registration, first purchase, etc.). They provide very interesting information to assess the impact of constant changes in the ecosystem (new functionalities, campaigns launched, discounts, etc.).
Example of a cohort study
As an example to analyse the recency and quality of users recruited over time, we have the following chart, where in the first column we can see the percentage (%) of users that were recruited in each month and in the following columns we can see the percentage of users that repeated purchases in subsequent months:
Significant lines can be seen, marked in yellow:
- Horizontal line: the percentages in that cohort are higher than the rest, it may indicate that the users recruited in March are more profitable than the average user. Actions taken in March that may have led to acquiring higher-quality users should be analysed.
- Diagonal line: in the month of August, for all cohorts, there is a “cell darkening”, i.e. a higher value. This indicates that in that month some action was taken or some circumstance occurred that encouraged users to buy again. Time will tell whether, in addition, the August cohort consists of high-quality users in the long term.
- Vertical line: Coinciding with cohort 7, i.e. the seventh month after the user is recruited, there is an increase in the percentage of purchases. This data is also worth analysing. It may indicate, for example, that there are repeat users who buy spare parts for a device that last 6 months, or that the automatic campaign consisting of sending a discount coupon 6 months after the first purchase is proving successful.
In addition, these cohort studies can be enriched by adding more data such as the average purchase value or the sum of purchases made in that month, providing information that is not easily identified at first glance.
These advanced analytics tools are very useful for defining strategies aimed at achieving secondary objectives, improving user recall, optimising shipping costs, improving product recommendations, etc.
Genetsis Group
Share
Contact
Do you want to implement advanced analytics tools?
Get in touch
