Average purchase revenue per active user
What one visitor to a page was worth in purchase revenue.
What it means
Average purchase revenue per active user is purchase revenue divided by active users, in your store currency. On the Pages tables both halves are counted per row: the revenue credited to that page, over the distinct users who engaged with it. Everyone who visited and didn't buy sits in the denominator, so it lands far below your average order value and is meant to. One thing shapes every row: purchase revenue is credited to the page where the purchase happened, which is normally the checkout, so most page rows show nothing at all.
Show the math
Formula and a worked example
Purchase revenue is the money from tracked purchases credited to that page. Active users is the count of distinct users who engaged with the page — browsers on devices, so one person shopping on a phone and a laptop counts twice.
Worked example. The checkout page shows 1,500 active users and €45,000 of purchase revenue. 45,000 ÷ 1,500 = €30. One user who reached that page was worth €30 on average, buyers and abandoners together.
The same store's product pages show €0.00 on nearly every row. That isn't a verdict on those pages — it's where the purchase was recorded. A product page that sends people to a €30 checkout is doing its job and will never show it in this column.
It answers the question
For the pages that carry revenue, what is one user who reached them worth? That figure is the ceiling on what getting someone there can sensibly cost you.
Why it matters
It puts a euro value on reaching a page, which is what turns a funnel argument into a budget one. If a user who reaches your checkout is worth €30, then every improvement upstream has a price you can check it against.
It matters just as much for the rows that read zero. Knowing the zero is a reporting fact rather than a failing page stops you deleting or rewriting pages that quietly do the work and hand the sale to another URL.
What good looks like
There's no standard to cite, and a euro figure per user couldn't have one — it moves with your prices, your category and which page you're reading. Judge the checkout row against its own trend over the last 90 days, and judge product and collection rows only against themselves, never against each other. For a store-level version of the same calculation, read Revenue Per User on Google Analytics Overview, where the denominator is every active user on the site rather than the users of one page.
How to improve it
| Lever | What you do | Expect | How long | Watch out for |
|---|---|---|---|---|
| Fast Stop sending users to pages they never buy from | Cut the placements and keywords whose landing pages collect users and no revenue anywhere downstream | Value per user up on the rows that remain | 1–2 weeks | Active users and sessions both fall, and some of that traffic was early-stage interest that would have returned to buy later. You won't see which. |
| Fast Raise the basket rather than the traffic | Set free delivery a little above the current average order and cross-sell at the cart | Same users, more revenue behind them | 1–2 weeks | Baskets under the new threshold get abandoned, and bundle discounts thin the margin on the orders that do complete. |
| Slow Take steps out between the page and the order | Remove forced account creation, cut form fields, show delivery cost earlier | More of the users who reach checkout finish | 4–6 weeks | Development time, and the fields you remove were collecting data — phone numbers, marketing consent — that other parts of the business rely on. |
| Slow Give the pages that earn nothing a route to the ones that do | Clear next steps from content and collection pages into the products that convert | More users reaching the pages where revenue is recorded | 1 quarter | Every extra link is another way off the buying path, so watch Conv. Rate on those templates rather than assuming more movement is better. |
Every lever costs something somewhere. The last column is the one to read twice.
Read it with
Value per user and the number of users pull against each other. Watching both is the only way to see whether reach was bought at the cost of quality.
More people, each worth more
Both halves moved the right way, which almost always traces back to one channel finding an audience that matches what you sell. Identify it before the effect fades.
Smaller and better qualified
Fewer users, each worth more. Usually a deliberate tightening of targeting. The risk is delayed — broad traffic often buys later, so a clean-looking cut can show up as a soft month next quarter.
Reach bought with dilution
Plenty more users, each worth less. Fine while the arithmetic holds. The moment value per user drops under what you pay to get one, every extra visitor costs you money.
Fewer and worth less
Both falling together on a revenue-carrying page looks identical to a broken purchase event. Rule the tracking out first — it's the cheapest check and the most common cause.
The same calculation at two altitudes. This one is per page; Revenue Per User on Google Analytics Overview is the whole site. When the site figure falls and the checkout row holds, the loss happened before checkout — fewer people are getting that far, rather than getting there and spending less.
One is a rate, the other is the count underneath it. A rate on a row with 40 active users will swing wildly week to week and mean nothing; the same rate on 4,000 is a signal. Read the count first and ignore the rate on thin rows.
Common misreads
Average order value divides by orders. This divides by everyone who engaged with the page, most of whom bought nothing, so it sits well below the price of an order and always will.
Purchase revenue is recorded on the page where the purchase happened. Nearly every product and collection row reads zero for that reason alone, and reading it as failure means judging your catalogue on a column that was never going to describe it.
You can't add rates, and the users behind them overlap — the same person appears on several page rows. For a site total, read Total Revenue instead.
Also called
Revenue per active user · ARPU · shown as Revenue Per User on Google Analytics Overview
See yoursValue per active user for every page, next to the active users and engagement time behind each row.
Open Google Analytics Pages →