> For the complete documentation index, see [llms.txt](https://docs.bloomanalytics.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.bloomanalytics.io/customer-metrics/cohort-analysis.md).
# Cohort Analysis
Bloom's Cohort Analysis groups Shopify customers by first purchase date to reveal retention, repeat purchase trends, and lifetime value over time.
### What is Cohort Analysis?
Cohort Analysis in Bloom groups your Shopify customers into cohorts based on when they first purchased, then tracks how each cohort behaves over time. A cohort is a group of customers created based on shared criteria, and Bloom supports both industry-standard and custom cohorts. It is for merchants who want to understand customer lifetime value, retention, and repeat purchase behavior, so they can decide how much to spend on acquisition, measure campaign effectiveness, and identify opportunities for strategy changes and market expansion.
### How it works
Bloom builds cohorts from your Shopify order data and lets you view each cohort through one of four analysis metrics. Each metric answers a different question about customer behavior, and you can switch between them from a dropdown.
**Accumulated Sales per Customer**
Tracks the total revenue generated on average by a customer within a cohort over the monitored period. Useful for gauging the revenue each customer contributes over time.
**Customers**
Counts the individuals from each cohort who have made repeat purchases over time. Focused on retention and new customer gains, so you can see how well you are maintaining and growing your customer base.
**Cohort Sales**
Shows how cumulative sales volume of each cohort increases with time. Reveals purchasing trends and behavior across cohorts, useful for shaping strategies to target them.
**Cohort Transactions**
Details the number of transactions attributed to each cohort during the monitored period. Sheds light on engagement levels and purchase frequency.
**How customer value over time is calculated**
For a single cohort, Bloom shows the average value of customers in that cohort at each period after their first order. The formula is:
(Total first order sales + total additional sales from the cohort in the same month) / total customers in the cohort.
For example, the May 2023 cohort had an average first order value of $104, and three months later that average had grown to $211.
### Data available
The Cohort Analysis report renders as a table. Each row is one cohort, and columns show how that cohort behaves over time.
#### Single cohort report columns
| Column | Description |
| ---------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Cohort | The first cell in each row. Identifies the cohort under analysis (for example, customers whose first order was in May 2023). |
| New Customers | The total number of customers in the cohort who made their first purchase in the cohort period. |
| First Order | The average value a new customer brings to the cohort. Calculated by dividing total sales from the first order by the total number of customers in the cohort. |
| Evolution columns (numbered) | The columns titled with numbers represent repeat purchases made by cohort customers during the monitored period. Used to estimate customer value over time. |
#### Multi-cohort comparison
Read down individual columns of the table to compare cohorts against each other and see how behavior shifts.
| Comparison | What it shows |
| ------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| First order values across cohorts | Insight into first-time buying behavior. For example, the report may show consistent first order values across all cohorts. |
| Repeat sales over time across cohorts | Trends in retention and engagement. For example, the May 2023 cohort may show higher repeat sales in following months than later cohorts, indicating a decreasing repeat sales trend. |
### How to use it
* Navigate to the Cohort Analysis section of the app.
* Set your date range in the \*\*Date Range\*\* field. The default is the past six months. Choose a predefined range or specify a custom one. Use shorter durations for businesses with rapidly changing customer dynamics or for evaluating new cohorts, and longer durations for businesses with stable customer behavior.
* Choose a cohort grouping (weekly, monthly, quarterly, or yearly). Use weekly or monthly groupings for emerging trends and dynamic customer bases, and quarterly or yearly for stable long-term patterns.
* Choose the metric to analyze from the dropdown: Accumulated Sales per Customer (default), Customers, Cohort Sales, or Cohort Transactions.
* Read the report by row to analyze a single cohort over time, and by column to compare cohorts against each other.
* Apply the insights to your workflow:
\- \*\*Identify valuable customers\*\* by looking for characteristics common to repeat buyers (marketing channel, geographic location, loyalty program participation), then engage them through cohort comparisons.
\- \*\*Understand customer behavior\*\* by monitoring average customer value over time. Compare with earlier cohorts to identify consistent patterns or variances, and adjust the analysis period for deeper insights.
