What Shopify Analytics is good at
Worth saying plainly, because most pages like this one pretend otherwise.
Shopify Analytics is accurate and it is free. Sessions, conversion rate, average order value, sales by channel, returning customer rate: it reports all of that correctly, straight from the source of truth for your orders. Google Analytics 4 adds acquisition, so you can see which campaign brought the traffic.
Between them they answer what happened with real authority. Nothing on this page suggests you should turn either of them off. Harvv runs alongside both.
The limit is structural rather than a missing feature. Both tools record events your store fires. Neither records what the shopper's hands did in between, so when the funnel drops from 4.1% to 3.2% they can show you the drop and nothing about its cause. You are left guessing, or watching recordings hoping to catch it.
Six questions neither one can answer
Each of these maps to something Harvv measures on a Shopify store today.
Which element on the product page do shoppers click that does nothing?
Shopify + GA4 Not measured. Neither reports clicks that produce no response.
Harvv Dead clicks and rage clicks per page, with the specific stuck element and the number of sessions affected.
Where in checkout do people stop, and on which step?
Shopify + GA4 Shopify gives you a funnel to the step. It cannot tell you what the shopper did while they sat there.
Harvv Checkout step events joined to behaviour: the hesitation pause before the drop, and what they touched first.
Did they leave to compare prices, or did they just go?
Shopify + GA4 Both record the exit. Neither records the reason.
Harvv Tab-away rate, copied text (a price or a SKU being checked elsewhere), and multi-tab sessions.
Did anyone actually read the size guide, or scroll past it?
Shopify + GA4 Scroll depth at best, and only if you instrumented it yourself.
Harvv Read versus skimmed versus skipped, per page, plus elements people hover with interest and never click.
Is the theme broken on mobile for real shoppers?
Shopify + GA4 Not measured. Lighthouse tests a lab device, not your traffic.
Harvv Core Web Vitals from real sessions, JavaScript errors, layout shifts, and tap targets too small to hit.
How it works on Shopify specifically
- Orders without GA4Harvv records orders itself through the Shopify Web Pixel, so revenue, order count and average order value work whether or not Google Analytics is connected. Connect GA4 and we show both sources side by side rather than quietly choosing one.
- The cart is linked to the sessionThe session id is stamped onto the cart so a shopper who adds one item and buys later is still connected to the visit that persuaded them.
- Checkout measured step by stepEach checkout step is its own event, so a drop is attributable to a step rather than to checkout as a whole.
- Non-web orders excludedPoint of sale, draft orders and marketplace channels never touched your website, so counting them against web sessions would inflate every rate. We leave them out and label ours online store revenue.
- Built for real devicesThe pixel is 28 KB gzipped, loads asynchronously and is written in ES5, because a Shopify audience includes older Android webviews where a modern syntax error would break the script entirely.
What you get on day one
Paste one script tag. There is nothing to configure, no events to define and no tag manager. 67 event types across 10 categories start recording immediately, and the first findings appear as soon as there is enough traffic to be worth reading.
Findings arrive as plain statements with the number of sessions affected and the element responsible, not as a dashboard you have to interpret. You can also ask in plain language, in the dashboard, in Slack, or from your own AI assistant through the connector.
Questions
Does Shopify have built-in analytics?
Yes. Shopify Analytics reports sessions, conversion rate, average order value and sales by channel, and it is accurate for all of that. What it does not report is behaviour: which element a shopper clicked that did nothing, where they hesitated, or whether they left to compare a price. That is the gap Harvv fills, and it runs alongside Shopify Analytics rather than replacing it.
Do I need Google Analytics connected for this to work?
No. Harvv records orders itself through the Shopify Web Pixel, so order count, revenue and average order value work with no GA4 connection at all. If you do connect GA4 we read it as a second, independent source and show you both rather than silently picking one.
Will it slow my store down?
The pixel is 28 KB gzipped and loads asynchronously, so it never blocks rendering. It is written in ES5 because a real Shopify audience includes older Android webviews, and a script that throws a syntax error on those devices is worse than no script at all.
How is this different from a heatmap tool?
A heatmap shows you where clicks landed. It does not tell you which of those clicks did nothing, which ones were repeated in frustration, or which shopper was comparing your price in another tab at the time. Harvv reports the outcome of the interaction rather than its coordinates, and it names the element so a developer can act on it.
Does it capture customer data?
No personal data. The pixel scrubs text from password, email, telephone and number inputs, from anything inside a container marked private, and from request bodies. It records that an interaction happened and what it hit, not who did it or what they typed.
Why do your revenue numbers differ from my Shopify admin?
Because we exclude orders that never touched your website. Point of sale, draft orders, Shopify Collective and marketplace channels all appear in your Shopify admin total but no web session produced them. Counting them against website sessions would inflate every rate we report, so the portal labels our figure "online store revenue" and leaves the rest out.