Analytics builds every number on the page from the orders in a date range you choose.
Choosing the range#
Set your own from and to dates, up to a year apart. Leave them out and you get the trailing 30 days. You can also group the charts by day, week, or month. A wide range grouped by day is noisy; group it by week or month and the shape of the trend shows up.
If a range comes back empty, that means no orders landed in it. Widen the range rather than assuming something is broken.
What is counted#
- Revenue: what your non-cancelled orders in the range are worth, by bucket.
- Orders: how many orders came in per bucket, cancelled included, broken down by status.
- Average order: revenue divided by order count within a bucket. This is the number that tells you whether bundling or a free-shipping threshold is working.
- New and returning customers: split by bucket, based on whether a customer's first-ever order fell in that bucket or an earlier one.
The charts#
Revenue, orders, and average order are all reported per bucket across your chosen range, so you can see them move day by day, week by week, or month by month rather than as one flat total.
The customers chart splits new from returning per bucket. For most small stores the returning share is the number worth watching, because a returning buyer costs nothing to acquire.
Top products ranks the ten best sellers in the range by units sold and by revenue, counting only non-cancelled orders. The two rankings are often different: a cheap item can top units while contributing very little revenue, and that is useful to know before you reorder stock.
These numbers count orders in your store. They are not an accounting record and they are not a tax return.