Online store: five numbers to track every week

An online store dashboard shows dozens of charts. Most of them are accurate; few of them help you decide. What helps a manager or an ecommerce lead is a handful of numbers reread every week at the same time, compared with the previous week, each with a clear question attached: what do I change if this number moves?

This guide suggests five. It is written for teams of 20 to 500 people who sell online, on Shopify or elsewhere, and want to spend less time looking at charts and more time acting on them.

Why five numbers, and not fifty?

Because a weekly review has to fit into fifteen minutes. Beyond five indicators, people comment instead of deciding. The five chosen here cover the whole chain of a sale: people arrive, some of them buy, for a certain amount, some come back, and a margin is left.

A number on its own says nothing. Falling conversion with sharply rising visits may simply point to a campaign that brings in an audience more curious than ready to buy. Reading them together is what makes the decision.

1. Visits, by source

Total visits flatter or worry without explaining anything. What matters is the split: organic search, advertising, email, social media, direct. A source that drops from one week to the next points to a precise cause (a paused campaign, an email that did not go out, a page that lost its ranking).

Watch out for bot traffic. Shopify’s own help warns that the conversion rate might appear lower than expected because of bot traffic and suggests filtering for human traffic. A spike in visits with no sales deserves that filter before any conclusion.

2. Conversion rate

This is the share of visits that end in a purchase. Shopify calculates it as sessions that completed checkout divided by all sessions. Its visitor behaviour report also details the steps in between: sessions with a cart addition, then sessions that reached checkout.

Those steps show where the sale gets lost. Many cart additions and few payments: look at costs revealed late and at the checkout, topics of our article on abandoned cart reminders. Few cart additions: the product page is not reassuring enough, and what buyers look for before paying gives the checklist.

3. Average order value

Average order value, as Shopify defines it, is gross sales minus discounts, divided by the number of orders. It moves when the offer changes: a free delivery threshold, a bundle, a complementary product suggested at the right time, or, the other way round, a discount that is too generous.

Track it alongside the conversion rate. A promotion that lifts conversion and lowers average order value has not necessarily earned more: the fifth number settles it.

4. Share of returning customers

This is the share of the period’s customers who had already ordered before. Shopify’s help gives an indicative range:

The same page explains why this number matters: a higher rate reduces acquisition costs. A returning customer does not have to be bought again with advertising. If the share drops, look at shipping quality, after-sales service and the emails that follow the first order; our email campaign guide covers those messages.

5. Margin after acquisition

Revenue does not pay salaries; margin does. Shopify defines gross profit as net sales minus the cost of goods sold, and points out that it is only accurate if that cost is entered in the store. That is the first job if it is not.

Then subtract what the week cost in advertising and fees. A Google Shopping campaign that generates revenue at zero margin keeps the team busy without earning anything; what makes a Google Shopping campaign rise or sink should be read in that light.

How to keep the review going every week

The hardest part is not choosing the numbers, it is rereading them every Monday. Three habits help:

• The same window: Monday to Sunday, compared with the previous week, and with the same week last year when the business is seasonal.
• The same place: a single table, sent automatically, rather than five screens to open. Our article on automated marketing reporting shows how to have it sent on its own.
• A written decision: one line a week, even “nothing to change”.

The same rhythm applies to product data. On an online plush toy store that we are putting back on its feet, we reviewed 653 products and defined 26 fields before enriching 481 product pages, all of them reviewed: a clean catalogue makes per-product numbers readable, whereas a messy one blurs them.

What these numbers do not tell you

They describe; they do not explain. Falling conversion can come from a stock shortage, a competitor slashing prices, a mobile bug or unusual weather. The dashboard tells you where to look; the cause is found by looking at the site, customer feedback and campaigns. The platform matters too: choosing between Shopify or Medusa also decides which reports you get without development, while each app you add can affect speed, and therefore conversion. Finally, a clean feed is what makes sales possible on Google Shopping and marketplaces.

If you are rebuilding your store or want to put these numbers back in order, our ecommerce service starts with a review of your catalogue, your traffic and your reports.

Frequently asked questions

Sources

1. Shopify Help, analytics report fields (new tab)help.shopify.com