Where is the conversion rate in GA4?

In Traffic acquisition it is called Session key event rate: sessions with a key event divided by all sessions. User acquisition shows User key event rate, the same idea per user. Both count any key event, so count purchases separately when you mean buying.

Why does Shopify show a different conversion rate than GA4?

They divide different things. Shopify's rate is the share of its sessions that resulted in an order, while GA4's counts sessions with any key event its tag recorded. Shopify's session measurement rollout ran from September 21 to 23, 2026, so its rate can move while orders stay flat.

Can I see conversion rate by campaign in GA4?

Yes. In Traffic acquisition, click the plus next to the dimension drop-down and add Session campaign. Campaign names come from your UTM tags or auto-tagging, so untagged links will not show up there. Then divide each campaign's purchases by its sessions, as you would for channels.

How to find your best converting channel in GA4

In Traffic acquisition, choose purchase as the key event and export the table. Divide each channel's purchases and revenue by its sessions and drop channels below a minimum. Then check the order in User acquisition and split mixed channels by campaign.

By Joris van Huët, Founder & CEOUpdated 2026-10-018 min read

Run the numbers for your store: the free manual attribution time calculator.

In GA4, open Traffic acquisition, choose purchase in the All events menu and divide each channel's purchases by its sessions. Add revenue per session and a minimum number of sessions. Then check the order in User acquisition. A channel that leads on all of it, at real volume, is usually your best converter.

Eight steps and one spreadsheet. You need the Viewer role at the property level to export a report. The menu paths are from Google's Analytics help pages as they read on 1 October 2026. Keep a calculator nearby: GA4 shows neither purchases per session nor revenue per session by default.

Step by step

Pick a long, closed date range. In the date picker at the top right, choose whole weeks that ended before this week. Google may withhold data in a narrow date range when user or event counts are low. A long range keeps small channels on the table. Menu path: (report) > date picker > custom range > Apply.
Open Traffic acquisition. It groups sessions by Session default channel group, the channel associated with the start of each session. Its default columns include Sessions, Key events, Session key event rate and Total revenue. Menu path: Reports > Acquisition > Traffic acquisition.
Choose purchase as the key event. Open the All events menu on the Key events column and pick purchase, so the count shows purchases only. The rate column keeps its own definition: sessions with any key event, divided by all sessions. Unless purchase is your only key event, step 5 does the purchase rate by hand. Menu path: Traffic acquisition > Key events column > All events > purchase.
Export the table. Click Share this report in the top right, then Download File, then Download CSV. The file holds up to 100,000 rows, far more than your channel list needs. Menu path: Share this report > Download File > Download CSV.
Do two divisions per channel. Purchases divided by sessions gives purchases per session. Total revenue divided by sessions gives revenue per session. GA4's recommended counting method counts a key event every time it is triggered, so a session with two orders counts twice. Menu path: none, this is your spreadsheet.
Set a floor before you rank. Pick a minimum number of sessions and purchases and write it down. Channels below it go on a list called "too small to rank". A channel with a handful of sessions can top any table on luck. Menu path: none, this is your spreadsheet.
Read the order from the first touch. User acquisition groups users by the channel that first brought them. You can narrow its key events to purchase too. Google advises against comparing metric values across the two reports, because their scopes differ. Compare the order of the channels instead. Menu path: Reports > Acquisition > User acquisition.
Split a channel that does two jobs. Click the plus next to the dimension drop-down and add Session campaign. Brand search, retargeting and win-back emails can carry a whole channel's rate on people who were coming anyway. Menu path: Traffic acquisition > plus icon next to the dimension drop-down > Session campaign.

A worked example

For illustration, here is a made-up quarter for four channels, as it looks after step 5.

For illustration	Sessions	Purchases	Purchases per session	Total revenue	Revenue per session
Email	2,000	80	4.0%	€4,000	€2.00
Organic Search	12,000	240	2.0%	€30,000	€2.50
Paid Social	15,000	150	1.0%	€13,500	€0.90
Referral	150	9	6.0%	€450	€3.00

Sort by purchases per session and Referral wins, at 6.0% on 150 sessions in this worked example. Step 6 benches it. Nine purchases is a coin toss, not a trend.

In this worked example, Email then leads the rest at 4.0% and Paid Social trails at 1.0%. Revenue per session reorders the top. For illustration, Organic Search earns €2.50 a session against Email's €2.00, because its orders are bigger. So Email is the best converter on orders and Organic Search the best on euros. Write down which one you mean before anyone says "best".

Step 7 can reorder it again. Suppose User acquisition puts Paid Social first: the people it brought went on to buy in sessions other channels closed. Its weak session rate would then describe the closing visit, not the channel.

Here is how much revenue sits in that gap in one store's anonymised GA4 export, 1 January 2024 to 21 August 2026.

What the export shows	Value	Source cell
Journeys with 2 or more touches, three rows added	20.6% of revenue	Journeys sheet, 2-3, 4-9 and 10+ touches rows
Journeys with 10 or more touches (16.0 days to buy)	3.0% of revenue	Journeys sheet, 10+ touches row

In that one store, journeys with two or more touches hold 20.6% of revenue: 12.2% plus 5.4% plus 3.0% (Journeys sheet). Those are the journeys where the buying session can come from a different channel than the first one. The longest, with 10 or more touches, took 16.0 days to buy (Journeys sheet). If your export looks like this, step 7 can reorder about a fifth of your revenue, and steps 5 and 6 settle the rest.

What should I check when the ranking looks wrong?

A channel you pay for is missing. Widen the date range first, since thresholds hide low counts. If its visits land in Direct, Google suggests linking your ad accounts or tagging your destination URLs.
Total revenue is blank. Google says the purchase event needs both a value and a currency parameter, or Total revenue shows no purchase data.
The rate looks high on every row. Check which events carry a key event star. With Ads Personalization on and a linked Google Ads account, Google also marks browsing events such as view_item as key events. Rank on purchases per session instead.
The two acquisition reports disagree. They are supposed to. User acquisition is scoped to new users and Traffic acquisition to sessions, so only the order of the channels is worth comparing.
The leader flips from quarter to quarter. Use the date picker's Compare option against the previous period. A lead that changes hands every quarter is noise, and budget should not chase it.

What to do this week

Audit your key events. In Admin, under Data display, click Events and open the Key events tab. Pass: purchase carries a star and every other star is an action you would pay for. Fail: browsing events carry stars. Unstarring only stops future counting, so check with whoever runs Google Ads first, and rank older ranges on purchases per session.
Compare with the previous period. Rebuild your table with the date picker set to Compare, previous period. Pass: the same channels lead in both periods. Fail: the order flips, so the gap is noise and no budget should move on it yet.
Look behind the leader's landing pages. Filter to your top channel and add Landing page + query string as a secondary dimension. Pass: its purchases spread across many pages. Fail: one page carries the rate, so the channel's "conversion" is really that page.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Frequently asked questions

Where is the conversion rate in GA4?
In Traffic acquisition it is called Session key event rate: sessions with a key event divided by all sessions. User acquisition shows User key event rate, the same idea per user. Both count any key event, so count purchases separately when you mean buying.
Why does Shopify show a different conversion rate than GA4?
They divide different things. Shopify's rate is the share of its sessions that resulted in an order, while GA4's counts sessions with any key event its tag recorded. Shopify's session measurement rollout ran from September 21 to 23, 2026, so its rate can move while orders stay flat.
Can I see conversion rate by campaign in GA4?
Yes. In Traffic acquisition, click the plus next to the dimension drop-down and add Session campaign. Campaign names come from your UTM tags or auto-tagging, so untagged links will not show up there. Then divide each campaign's purchases by its sessions, as you would for channels.

Go deeper: Causal attribution, explained.

Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.

Keep reading

Terms in this article

Browse the full glossary