Blog · Board and management reporting
Consumer analytics compares recorded purchasing behavior across defined groups of buyers. This guide works purchase frequency, basket value and retention from transaction records, then explains the checks and decisions each measure supports.
Consumer analytics compares recorded buying behavior across groups of customers: how often they order, what they spend per order and whether they return. It gives commercial teams evidence for product, retention and service decisions when the customer groups and measurement windows stay consistent.
A larger basket is one finding. A larger basket combined with lower repeat purchasing is a different business problem, and neither alone establishes customer profitability.
Start with the question the team needs to answer. A product team might compare buyers entering through different product ranges. A retention manager might compare acquisition cohorts. An operations manager might examine purchasing after a delivery problem.
Keep observed behavior separate from explanations about motives. The SBA's market research guidance describes research that combines consumer behavior, economic trends and demographic information. Your own transaction records provide a narrower view: what happened among buyers you can observe. They usually cannot explain why someone chose another seller or decided not to buy.
Use three complementary measures:
Purchase frequency = Qualifying orders / Distinct active customers
Average basket value = Net sales / Qualifying orders
Next-period retention = Baseline customers buying in the next period / Baseline customers
Define qualifying orders before calculating. This example includes completed, positive-value orders, excludes canceled orders and counts an order once regardless of its number of lines. Sales exclude sales tax and are after discounts and credits. For the sales adjustment principle, IRS Publication 334 explains that returns and allowances reduce gross receipts in arriving at net receipts.
Request order identifier, order date, customer identifier, product code, quantity, net sales and order status. Bring refund records with the original order identifier. Use a stable internal customer key in the working file; names and contact details are unnecessary for these comparisons.
Assign each customer to a segment using a documented rule. If the rule is first-purchase product range, keep that assignment fixed for the retention analysis even when the customer later buys another range. Otherwise successful cross-selling can move customers out of the very group whose retention you are trying to measure.
Finance owns the net sales definition and reconciles it to the sales register. The commercial or consumer insights lead owns segment definitions. The retention owner decides which follow-up action is practical. An analyst or reporting owner checks customer matching, date coverage and the calculation population.
Distinguish an order-line export from an order summary. Summing line sales gives order sales, but counting those lines gives item rows rather than orders. Aggregate to one order before counting frequency, then to one customer before counting retention.
A fictional US household-products company groups its June 2026 buyers by their first-purchase range: Starter or Family. Every customer below bought in June, and the company observes all of July before measuring next-month retention. There are no returns in these illustrative records.
The table is the customer summary after completed orders have been deduplicated. A zero July order count means a confirmed absence of qualifying orders in a complete file, not a missing record.
| Customer | Entry range | June orders | June net sales (USD) | July orders | July net sales (USD) |
|---|---|---|---|---|---|
| C101 | Starter | 1 | 40 | 1 | 45 |
| C102 | Starter | 2 | 90 | 0 | 0 |
| C103 | Starter | 1 | 50 | 1 | 60 |
| C201 | Family | 2 | 160 | 2 | 180 |
| C202 | Family | 3 | 270 | 1 | 95 |
| C203 | Family | 1 | 90 | 1 | 100 |
| Total | 10 | 700 | 6 | 480 |
For Starter, June frequency is (1 + 2 + 1) / 3 = 1.33 orders per customer. Its basket value is (40 + 90 + 50) / 4 = $45.00. Two of its three baseline customers buy in July, so next-month retention is 66.7%.
Apply the same steps to Family and the whole baseline:
| June segment | Active customers | June orders | June net sales (USD) | Orders/customer | Basket value (USD/order) | Retained in July | July retention |
|---|---|---|---|---|---|---|---|
| Starter | 3 | 4 | 180 | 1.33 | 45.00 | 2 | 66.7% |
| Family | 3 | 6 | 520 | 2.00 | 86.67 | 3 | 100.0% |
| Total | 6 | 10 | 700 | 1.67 | 70.00 | 5 | 83.3% |
Orders per customer and dollar averages are rounded to two decimals; percentages are rounded to one. Calculations use unrounded values. For more detail on the denominators, see average orders per customer and average order value.
Customer counts reconcile: Starter's three plus Family's three equals the six distinct June buyers. Order counts reconcile: four plus six equals the ten completed June orders. Sales reconcile: $180 plus $520 equals $700.
The overall basket value is $700 / 10 = $70.00. Do not average the displayed segment basket values; segments have different order counts. The underlying identity is net sales = order count × unrounded basket value.
Retention also reconciles. Two retained Starter customers plus three retained Family customers equals five. Divide those five by the original six, not by July's order count. The cohort retention reading guide explains how to extend that same fixed population across more periods.
Within-June repeat purchasing answers a separate question. C102, C201 and C202 have more than one June order, so the within-month repeat buyer share is 3 / 6 = 50.0%. July retention is 5 / 6 = 83.3%. Both figures are correct because their events and windows differ.
Family has higher June frequency, higher basket value and more returning customers in this small example. That supports investigating its product fit and replenishment pattern. It does not justify declaring the segment universally better or assuming the result will persist in a larger population.
The retention owner should inspect Starter's missing repeat purchase before changing messaging. Was the product meant to last longer? Was the customer still inside the normal replenishment cycle? Was delivery delayed? Transaction analysis identifies the group to investigate; customer research can test the explanation.
Finance should add contribution costs before approving a larger acquisition budget. Family's higher sales could come with heavier products, more shipping support or deeper introductory discounts. Consumer goods teams can connect these questions to CPG operating comparisons.
Bring a transaction export with stable customer keys, order dates, net sales and your documented segment assignments. Covirage maps the columns, and its tools compute and check the comparisons; the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring a sample of your own data.
Connect the comparison to customer profitability before deciding where to invest.
Transaction analysis measures what identified buyers did in the records you hold. Market research also investigates potential customers, attitudes, competitors and wider demand. Purchase records can reveal a retention gap, while interviews or surveys may help explain the reasons behind it.
Purchase frequency is qualifying orders divided by distinct active customers in a defined period. Count each order once, even if it contains several product lines. State whether fully returned orders qualify, and apply that rule consistently across every segment being compared.
They depend on different populations and windows. A within-month repeat purchase rate counts buyers with multiple orders that month. Next-month retention follows a fixed group of earlier buyers and asks how many return. Label both the denominator and observation period.
Larger baskets identify higher sales per order, but they do not establish profitability. Product cost, discounts, returns, shipping subsidies and servicing costs can change the ranking. Add those costs on a consistent basis before deciding where to invest commercial effort.