How To Optimize Your Marketing Strategy With Cohort Analysis

Philippe Pavillet

Mar 30, 2023

Have you ever wondered which acquisition channels attracted the most engaged customers over time? Or which period of the year is the best time to run a marketing campaign? Or which products are good at keeping your customers coming back?  If so, it's time to dive into the world of cohort analysis. This powerful tool allows you to better understand your customers and their behavior, so you can make data-driven marketing decisions and boost your revenue. Cohort analysis may sound complex, but it's a simple concept that can transform the way you approach your marketing efforts. In this article, we'll explore why cohort analysis is important, how it works, and most importantly, how you can use it to supercharge your marketing strategy. 

So, what is cohort analysis? It is a method of analyzing customer behavior over time. One of the most well-known use cases is Google Analytics, which uses it to track how visitors keep coming back to a website. But it also makes a lot of sense to use it to track the behavior of existing customers! Cohort analysis involves grouping customers based on a common characteristic, such as the month they made their first purchase, the marketing channel they came from, or the product they have bought. By tracking these groups (or “cohorts) over time, you can gain insights into how customer behavior changes and identify patterns and trends. 

But, why is it so important? Because it provides a more nuanced view of customer behavior than traditional metrics like conversion rate or customer lifetime value. By tracking how different cohorts behave over time, you can see how your marketing and product strategies impact customer behavior and track retention rates over time.   While it’s possible to create your own cohort analysis in Excel, it may become a long and complex task, and it should be updated constantly. That is why if you plan to use it more than once and/or don’t want to spend a lot of time preparing your models, having a tool can help you be more efficient. 

Fructifi’s cohort analysis is focused on existing customers. You first select how to create the cohort - by date of first order, by acquisition channel, by account owner, and by product, then choose a metric you want to track: client retention, re-ordering patterns, or revenue retention. Finally, you decide the periodicity of the analysis: monthly, quarterly, or annually. However, just having that data will not help you if you don't use it to adapt your marketing and sales strategy. Thus, how do you interpret the graphs? 

Here are 4 examples of analysis you can make at Fructifi (note: all numbers below are fictional and purely provided as examples):  

1) Client retention by date of first order

In this chart, the “Average” line shows that, on average, 62 new customers start buying your product every quarter; and that one quarter after their first purchase (i.e. “Quarter 1”), 78% keep buying from you; a proportion that increases to 90% two quarters after the initial purchase, then stays at 90% in the third quarter and finally increases to 97% in the fourth quarter.

The point is then to compare the performance of each period to this average, as well as to other periods, in order to know when is the best time in the year to conduct customer acquisition campaigns. Because after all, what is the point of attracting a lot of new customers, if they don’t stick around?

Looking at the table above, it is pretty clear that customers who arrived in 2022-Q1, although numerous (133 vs. the average of 62), actually underperformed compared to those who arrived in other periods. Given this, it would probably be a good idea to avoid spending too much marketing budget in the next Q1 - or at the very least, to dig into the reasons why this has happened. Looking at the acquisition channels used during that period is a good way to do this.

2) Revenue retention by acquisition channel

Analyzing retention rates by acquisition channels will help you identify the channels that most effectively work for your business, i.e. those that attract the most loyal customers.

In this screenshot, you can see that customers who arrived via Google Adwords not only spent more in their initial quarter ($374,735) but over time they increased their spending a lot more than customers from other acquisition channels. For example, after 4 quarters, they generated 169% of their initial spend (i.e. 169% x $374,735 = $633,302), while after the same number of quarters, other channels showed either a decrease compared to initial spend (LinkedIn: 89%, Facebook: 87%, Word of mouth: 0%) or an increase that wasn’t as strong (Referral: 152%).

Knowing this type of information, you can optimize the way you allocate your marketing budget across your different channels - instead of randomly assigning or equally dividing it.

3) Order retention by account owner

Here, you can analyze the performance and efficiency of members of your sales team in retaining customers and compare them to their peers. For example, in this case, Clark Kent has performed best in keeping orders coming across every quarter.

This is precious information to have from a sales management perspective. Let’s say you have one sales rep who is great at attracting a lot of new customers, but underperforms when it comes to keeping them.

You could then choose between two different options:

  1. Train this person to do a better job at keeping customers engaged. 
  2. Decide this person is a natural “hunter” and ask him or her to exclusively focus on getting new customers, but not managing them over time. 

4) Client retention by product 

By analyzing a product’s performance over time, you can see which products work in the market and which do not. The goal is to keep the former product in the market and expand their line while getting rid of the latter which doesn’t retain many customers. 

A typical curve for a strong performing product will have a V or W shape: a strong start - due to the communication and excitement around the product launch - followed by a dip (allowing distributors time to sell their initial stock), and then an upturn in orders as they realize that the product has flown off the shelves.

A weak product? You will see a downward or flat slope: initial excitement (if any), followed by low numbers, and then no pick-up.

In the above example, the company has done a great job at making sure all their products keep customers engaged over time, “Product O” being the one that performs best. 

By analyzing client retention by product over time, you can identify how “sticky” each product truly is and adjust your product strategy accordingly. Simply put: invest in developing more products in the sticky category, and kill products that don’t keep customers coming. 

Note that in the examples above, we have analyzed the data on a quarterly basis, but in reality, the periodicity depends on the timeframe of your business goals, client behavior, or the seasonality of the product: you may need to analyze it monthly or yearly. 

In conclusion, cohort analysis is a fantastic tool for marketers to gain insights into customer behavior: interpreting data and taking action based on insights is crucial to optimizing marketing strategies. Regularly tracking and analyzing cohorts can help businesses refine their marketing efforts, increase customer retention, and drive revenue growth!

Inkubationszentrum in Paris 

Wir stellen ein!

© 2018-2022 Fructifi alle Rechte vorbehalten -Rechtliche Hinweise- Allgemeine Geschäftsbedingungen - Datenschutzbestimmungen