Ecommerce KPIs: the weekly numbers a DTC brand should watch
Eight ecommerce KPIs worth a weekly check, with the formula, where each lives in Shopify or GA4, what a weekly change means, and what to leave for the month.
The ecommerce KPIs worth a weekly check are net sales, orders, sessions by channel, conversion rate, average order value, add to cart rate, checkout completion, and sales by channel or landing page. Shopify is the record for the money (sales, orders, order value). GA4 explains it (which channel, campaign or page moved). Lifetime value, acquisition cost and repeat purchase rate matter too, but a single week is too short for them to change a decision.
Most ecommerce KPI lists are long, with no cadence and no word on which tool holds each number. A weekly KPI has to pass the same tests as any weekly marketing KPI: it can move in seven days, the move is bigger than noise, and someone would change a page, a budget or a product because of it.
The one equation behind every weekly change
Sales come from three numbers multiplied together:
Sales = sessions × conversion rate × average order value
It is an approximation (a session with two orders counts once in Shopify’s conversion rate), but it is the most useful way to read a week. When sales drop, one of the three moved. Sessions down is a traffic problem. Conversion down with traffic flat is a site, offer or stock problem. Order value down means the basket changed. The KPIs below are these three numbers, or ways to find which one moved.
The 8 ecommerce KPIs to check every week
| KPI | Formula | Read it in | Weekly question |
|---|---|---|---|
| Net sales | Gross sales minus discounts and returns | Shopify | Did we make more or less money? |
| Orders | Count of orders placed | Shopify | Did fewer people buy? |
| Sessions by channel | Visits, split by channel | GA4 | Did traffic change, and from where? |
| Conversion rate | Sessions with a purchase ÷ sessions | Shopify (or GA4, computed) | Did the same traffic buy less? |
| Average order value | (Gross sales minus discounts) ÷ orders | Shopify | Did the basket shrink? |
| Add to cart rate | Sessions with a cart addition ÷ sessions | Shopify, GA4 Purchase journey | Did product pages stop persuading? |
| Checkout completion | Purchases ÷ checkouts started | GA4 Checkout journey, Shopify | Is the checkout leaking? |
| Sales by channel and landing page | Purchases and revenue per channel or page | GA4 | Which source or page moved the total? |
1. Net sales
Formula: Shopify defines net sales as gross sales minus discounts and sales reversals. Shipping and taxes sit in total sales, not net sales.
Where: Shopify Analytics and the Finances report. GA4’s closest metric is purchase revenue, “the sum of revenue from purchases made on your website or app, minus any refunds given”. It only subtracts refunds that are sent to GA4 (as a refund event).
A weekly change means: split it with the equation above: sessions, conversion rate or order value.
Ignore: a dip from a returns-heavy week. Shopify records a return “on the date the return was processed”, so returns from a big sale drag down a later week.
2. Orders
Formula: the number of orders placed. GA4’s version is ecommerce purchases, the number of times users completed a purchase (it counts purchase events only).
A weekly change means: orders down with order value flat is a volume problem. Orders flat with sales down is a basket problem (KPI 5).
Ignore: the gap between Shopify orders and GA4 purchases, as long as it stays stable (why it exists). A gap that suddenly widens means a tag broke.
3. Sessions by channel
Formula: visits, grouped by GA4’s default channel rules (Organic Search, Paid Social, Email, Direct and so on).
Where: GA4, Reports, Acquisition, Traffic acquisition. Shopify also splits sessions by referrer and UTM tags, but GA4 puts its default channel groups, campaigns and landing pages in the same reports as purchases and revenue.
A weekly change means: a drop in one channel usually has one cause: an ad set paused, a newsletter skipped, a ranking lost. The traffic drop checklist walks through them in order.
Ignore: Shopify sessions, conversion rate and other session-based rates compared across September 21 to 23, 2026, when Shopify rolled out a new way of counting sessions.
4. Conversion rate
Formula: Shopify’s conversion rate is sessions that completed checkout divided by sessions. GA4 has no single default ecommerce conversion rate: divide ecommerce purchases by sessions yourself, or use session key event rate if purchase is your only key event.
A weekly change means: traffic flat and conversion down points at the site or the offer: a product out of stock, a shipping price change, a broken mobile button, a promo that ended. Read it per channel in GA4.
Ignore: generic “average ecommerce conversion rates”. See how to read your own conversion rate.
5. Average order value
Formula: Shopify’s average order value is (gross sales minus discounts) divided by orders. GA4’s equivalent is average purchase revenue, the average revenue per transaction.
A weekly change means: a discount, a shift toward cheaper products, or an upsell that stopped showing. GA4’s Ecommerce purchases report shows items viewed, added to cart and purchased, and item revenue, per product.
Ignore: one big gift or wholesale order in a small week.
6. Add to cart rate
Formula: Shopify’s added to cart rate is sessions with cart additions divided by sessions. GA4 shows the same idea per user in the Purchase journey report: session start, view product, add to cart, begin checkout, purchase.
A weekly change means: the product page persuaded fewer people. Look at price changes, new photos, missing reviews, or a channel sending visitors who were never going to buy.
Ignore: GA4’s cart-to-view rate as a store-wide KPI. It compares users who added a product with users who viewed that same product: useful per product, noisy as a total.
7. Checkout completion
Formula: purchases divided by checkouts started. GA4 counts checkouts from the begin_checkout event. In Shopify, use checkout conversion rate (sessions that completed checkout divided by sessions that reached checkout).
Where: GA4, Reports, Monetization, Checkout journey, which shows abandonment between begin checkout, add shipping, add payment and purchase.
A weekly change means: something between cart and payment got worse: shipping costs, a failing payment method, a forced login.
Ignore: an empty Add shipping step on Shopify. Google says Shopify collects begin_checkout, add_payment_info and purchase automatically, but you must set up add_shipping_info yourself.
8. Sales by channel and landing page
Formula: purchases (as key events) and total revenue, split by session channel or landing page.
Where: GA4. Traffic acquisition and the Landing page report both show key events and total revenue per row. Google’s Traffic acquisition guide notes that purchase events need both value and currency, or total revenue stays empty.
A weekly change means: “sales are down” becomes “Paid Social sales are down”, which you can act on by Tuesday.
Ignore: the split as the final word on what caused each sale. It depends on attribution settings, and many buyers touch several channels first.
The ecommerce KPIs that belong in the monthly review
| KPI | Why it is not weekly |
|---|---|
| Customer lifetime value | Built from months of repeat orders |
| Customer acquisition cost | Needs a month of spend and new customers |
| Returning customer rate | Shopify divides returning customers by all customers who ordered; it moves slowly |
| Gross and contribution margin | Need costs that live outside analytics tools |
| Return on ad spend by campaign | Small budgets swing week to week (how to calculate it) |
| Refund rate | Returns land weeks after the order |
Skip Realtime (it checks tracking, not performance), bounce rate and session duration. Compare each week with the average of the previous four, not only last week; around Black Friday, compare with the same week last year (Black Friday analytics checklist).
A 15-minute Monday routine
- Shopify, 3 minutes: net sales, orders, average order value, conversion rate, versus the last four weeks.
- Decide which branch moved: sessions, conversion rate or order value.
- GA4, 10 minutes: if sessions moved, open Traffic acquisition. If conversion moved, open Purchase journey and Checkout journey. If order value moved, open Ecommerce purchases and sort by item revenue.
- Write one line: what moved, the likely cause, and what you will change this week.
For a same-hour alarm when orders stop, see ecommerce sales drop alerts in Slack. If GA4 revenue looks wrong, fix the setup first with Google Analytics on Shopify.
Getting the GA4 half without opening GA4
The GA4 half of the routine is what we built Dialytics for, so read this with that in mind. It connects to GA4 only, not to Shopify, and does not read revenue, orders or order value.
Every Monday at 8am (Paris time) Dialytics posts a digest in Slack on what moved in your traffic, channels, engagement and key events, and the likely reason. Every morning it compares sessions, users, new users and the total of your key events with the same weekday over about the last 12 weeks, and posts at most one alert a day when one of them is unusual. purchase is a key event by default, so a collapse in purchases can show up there, unless other key events make up most of the total (linking Google Ads with Ads Personalization on, for example, makes GA4 mark add_to_cart and even session_start as key events). You can mention the bot in the thread to ask which channel or page drove a change, and it answers with the history of previous weeks.
Frequently asked questions
What are the most important ecommerce KPIs?
For a weekly check: net sales, orders, sessions by channel, conversion rate, average order value, add to cart rate, checkout completion and sales by channel or landing page. Sales equal sessions times conversion rate times average order value, so those three explain most weekly moves. Lifetime value, acquisition cost and repeat rate belong in the monthly review.
Should I use Shopify or GA4 for ecommerce KPIs?
Use Shopify for money and GA4 for the cause. Shopify reads every order, so its sales, orders and average order value are the record. GA4 only sees purchases its tag sent from a browser that allowed it, but it knows which channel, campaign and landing page brought each session. Pick one source per KPI and keep it.
What is the difference between average order value in Shopify and average purchase revenue in GA4?
Shopify computes average order value as gross sales minus discounts, divided by orders. GA4's average purchase revenue averages the revenue sent with purchase events, which Google's spec says should exclude shipping and tax. They describe the same idea but rarely match, because GA4 misses orders from browsers that blocked or declined tracking.
How do I know if a weekly drop in ecommerce sales is real?
Compare with the average of the previous four weeks, not only last week, and check the last day again after 48 hours, since GA4 can take that long to process. Then split the drop into sessions, conversion rate and order value. If Shopify orders held while GA4 purchases fell, the tag broke, not the business.