Purchase Frequency
Purchase frequency is orders per unique customer in a window — the multiplier that decides whether your store is a one-shot business or a habit business. Here's the formula, benchmarks by category, and how to grow it.
Purchase Frequency
Average number of orders placed per unique customer over a defined window — usually 12 months.
Purchase frequency (PF) tells you how often the same customer comes back. Divide total orders by unique customers in the same window and you have it. A PF of 1.0 means everyone bought once; a PF of 3.4 means your average customer placed roughly three and a half orders in the period.
It matters because it's the multiplier in the lifetime value formula. Average order value and margin set the size of a single transaction, but purchase frequency is what compounds. Move it from 1.6 to 2.2 and you've grown revenue per customer by nearly 40% without acquiring anyone new — which is why subscription, replenishment flows, and lifecycle email all earn their budget here.
Most teams under-measure this metric by mixing windows. A 12-month PF and a lifetime PF tell different stories, and reporting them interchangeably leads to overstated LTV. Pick one window — rolling 12 months is the standard for e-commerce — and hold it constant across cohorts.
Purchase frequency sits downstream of repeat purchase rate. RPR tells you what share of customers bought again at all; PF tells you how many times. A store can have a healthy RPR of 35% but a flat PF if the repeaters only buy twice and disappear — which is the classic signal that you have a trial problem solved but a habit problem unresolved.
Purchase Frequency = Total Orders / Unique Customers
Total Orders
Total Orders
Count of completed, non-refunded orders placed in the window
Unique Customers
Unique Customers
Count of distinct customers who placed at least one order in the same window
A beauty store on Shopify reviewing the last 12 months
Total Orders (12 mo): 48,200
Unique Customers (12 mo): 21,500
→ PF = 2.24 orders per customer
Above the apparel benchmark and on par with skincare replenishment averages. Combined with a €38 AOV, this puts annual revenue per customer at roughly €85 — strong enough to support a €25-30 blended CAC ceiling.
Purchase frequency varies wildly by category. Consumables (supplements, coffee, pet food, skincare) naturally push past 3.0 because consumption forces re-purchase. Considered goods (furniture, electronics, premium apparel) sit closer to 1.2-1.5 because the replacement cycle is years, not weeks. Benchmark yourself against your category, not against the e-commerce average.
12-month purchase frequency benchmarks by DTC category
| Category | Bottom quartile | Median | Top quartile |
|---|---|---|---|
| Supplements & vitamins | 1.8 | 2.9 | 4.2 |
| Skincare & beauty | 1.6 | 2.3 | 3.4 |
| Coffee & specialty food | 2.1 | 3.2 | 4.8 |
| Pet food & supplies | 2.4 | 3.6 | 5.1 |
| Apparel & accessories | 1.2 | 1.7 | 2.6 |
| Home & furniture | 1.0 | 1.2 | 1.5 |
| Consumer electronics | 1.0 | 1.3 | 1.8 |
The fastest levers for lifting PF aren't acquisition channels — they're post-purchase mechanics. Subscribe-and-save converts intermittent buyers into scheduled ones. Replenishment reminders timed to consumption cycle catch the moment of need. Lifecycle email and SMS flows (winback at day 60, cross-sell at day 30) close the gap between orders. Loyalty programmes work, but only when the next-tier reward is reachable in 2-3 more orders, not 10.
Frequently asked questions
It depends entirely on category. Consumables (coffee, supplements, pet food) should target 3.0+ over 12 months. Apparel and accessories typically sit at 1.5-2.0. Considered goods like furniture or electronics rarely exceed 1.3. Benchmark against your vertical, not the cross-industry average.
Repeat purchase rate is the percentage of customers who bought more than once — a binary signal. Purchase frequency is how many times the average customer bought. RPR tells you if customers come back at all; PF tells you how often. Both feed LTV, and RPR usually precedes PF in lifecycle reporting.
The standard LTV calculation is AOV × Purchase Frequency × Gross Margin × Customer Lifespan. PF is the multiplier that turns a single transaction into a customer relationship. Lifting PF from 1.6 to 2.2 raises LTV by ~38% with no change to AOV or margin, which is why retention investments out-earn acquisition spend at scale.
Rolling 12 months is the e-commerce standard because it absorbs seasonality. Use 90-day PF for fast-cycle consumables where annual numbers hide trends. Avoid lifetime PF unless your business is over five years old — younger stores have too many customers with incomplete histories, which suppresses the number artificially.
Three levers move it fastest: subscribe-and-save for consumables, replenishment reminder flows timed to your product's consumption cycle, and post-purchase cross-sell sequences (typically days 7, 30, and 60). Loyalty programmes help if the next reward tier is reachable in 2-3 orders. Discounting alone trains customers to wait, so use it sparingly.
No. Count only completed, non-refunded orders. Including cancellations inflates PF without reflecting real revenue. Most analytics platforms let you filter by financial status — set the filter to 'paid' and exclude orders fully refunded within the window.
Significantly. Customers acquired through organic search and referral typically show 30-50% higher PF than paid social cohorts, because intent at acquisition predicts retention. Segment your PF by first-touch channel — if paid social customers buy once and vanish while organic customers buy four times, your blended LTV hides a profitability problem.
Lifecycle email and SMS flows can lift PF 10-15% within one quarter once they're live and tuned. Subscription programmes take 2-3 quarters to show up in blended PF because adoption builds slowly. Loyalty and referral programmes are 6-12 month investments. Don't expect a single tactic to double the number.
No, though the terms get mixed up. Repeat order rate is the share of total orders that came from existing customers — an order-level metric. Purchase frequency is orders per customer — a customer-level metric. A store can have a 60% repeat order rate but a PF of only 1.8 if a small power-buyer segment skews the order share.
Don't extrapolate from 30-day data — it overstates PF dramatically. Use a 90-day-to-12-month conversion ratio from a mature cohort (typically 90-day PF × 2.2-2.8 = projected 12-month PF for consumables). For considered goods with longer cycles, anchor forecasts to your second-order rate at 180 days rather than raw frequency.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.