Cohorts

A cohort as a collection of customers grouped based on specific criteria.

Updated
May 22, 2023
Collections

Understanding Cohorts

Insight into customer behavior and the buying journey is one of the best tools in your ecommerce toolbox.

Identifying a group of customers, also known as a cohort, is one of the best ways to research and understand who shops at your online store and how they interact with you over time. 

At Tydo, we look at the definition of a cohort as a collection of customers grouped based on specific criteria. The most common way to group customers (and the way we do it at Tydo) is based on time—usually the month a customer is acquired or when they make their first purchase.

By tracking a cohort of people and their common characteristics, you’ll start to see patterns emerge: what keeps customers returning (or churning), how to optimize lifetime value (LTV), and why or how they spend the way they do.

How to calculate Cohorts

No calculation here

Data source: Shopify

Example

In September 2022, 1,500 new customers visited your online apparel store and made their first purchase with you. These 1,500 first-time customers make up your September 2022 cohort.

Takeaway

A cohort is like a graduating class. If you entered freshman year in September 2021, your cohort would be your classmates—also known as the class of 2025.

Using cohort metrics and tracking cohort-related data helps you keep a pulse on your business. 

It gives you an inside look into whether or not your retention strategies are working and where you can make improvements, such as building community or increasing engagement to increase lifetime value (LTV).

You can answer key questions about your business through a cohort analysis. 

  • What happens after a customer makes that first purchase? Do they return, or do they churn?
  • Is there a specific cohort buying more frequently or spending more? Why?
  • Are certain products or inventory selling better than others?
  • Which groups of customers drop off more frequently? Is there a way to reduce churn?
  • What retention strategies can you implement or optimize to increase customer lifetime value (LTV) or average order value (AOV)?

Read more about Cohorts

How does a cohort analysis help you?

Reduce your churn rate 

Tracking churn trends over time is powerful in helping you identify areas of weakness in your ecommerce business. Acquiring new customers is important, but retaining them over time is even more essential (and cost-efficient).

Running a cohort analysis allows you to understand customer performance (why are your customers doing what they’re doing?).  

Aashish Ramamurthy, the host of the DTC Drive Podcast, recommends tracking these metrics frequently. “Brands need to start looking at granular cohort analysis, be it weekly, biweekly, [or] monthly across each metric they want to keep top of mind.”

Different cohort analyses will focus on how long those different groups of customers engage with your brand and why they might churn. You can even use tags to see the difference between customers with subscriptions and non-subscribers. 

Once you pinpoint these patterns, you can test new strategies to reduce or eliminate churn, such as directly sending a follow-up to customers to learn more about why they’ve moved on.

Find your best-performing products

A cohort analysis offers insight into your customer’s behavior and highlights how different products impact customer loyalty.

You can filter your cohort analysis by product to see which products bring in the most customers and increase avg. customer spend. 

You can answer key questions: What’s the AOV of cohorts who’ve bought a specific product? Which products drive the most repurchases? 

It’s possible that one of your products is better suited for getting more people in the door (acquisition) and another product is better suited for keeping people around (retention). It could also be time to introduce a new product. 

Leverage your holiday data

The holiday shopping season is a crucial time for any ecommerce business.

Looking back at recent metrics helps you better understand and plan for upcoming seasons. Start by asking yourself: How was my holiday season last year? Are those holiday shoppers returning customers now? What can I do to improve customer engagement this time? 

A cohort study of your Black Friday Cyber Monday customers will show you if those customers have continued patronizing your business or have churned over time. Are they repurchasing certain items or ordering at specific intervals? Maybe, you can find the right time to remarket to your holiday customers as well. 

Building a retention playbook with Cohorts

Every retention playbook should include a cohort analysis.

A lot of what you need to know about your customers and retention can be found in a cohort analysis. You can build your retention playbook by answering these questions with a cohort analysis: 

  • Which customers are most likely to return and repurchase?
  • Which customers come back frequently throughout the year?
  • Is there a cohort that’s doing better than others?
  • Are certain SKUs selling better, and can they be part of a bundle to increase average order value (AOV)?
  • Are certain influencers driving traffic to your website, and are those visitors becoming regular customers?
  • What does the AOV look like for returning customers vs. first-time customers?
  • Where are your customers coming from? 

Using a cohort analysis, you can discover which products, demographics, and seasonal patterns lead to repeat purchases and lifetime value. You can discover the correlation there. 

Then, you can move your marketing dollars to the customer segments that are most likely to make repeat purchases. Do as much as you can to eliminate tensions that lead customers to drop off and churn.

Cohorts and LTV

A cohort analysis also enables you to get a good picture of the lifetime value (LTV) of your customers and try new strategies for boosting connection, engagement, and potential spending over time.

Are your new customers sticking around? How does an average customer from a year ago compare to a recent customer? What’s an average customer worth to you after 3 months, 6 months, etc?