This guide walks you through a structured conversion rate optimisation (CRO) audit using Google Analytics 4. By following each step, you will identify where visitors leave your funnel, quantify the revenue impact, and produce a prioritised fix list. Most teams complete this audit in a single working day.

What You'll Build

  • A GA4 funnel exploration that maps every stage of your conversion path
  • A segment comparison showing how different traffic sources convert
  • A drop-off report that ranks pages by lost conversion value
  • A prioritised CRO backlog ready to hand off to your design or development team

Prerequisites

  • GA4 property with at least 30 days of data (90 days preferred)
  • At least one key conversion event configured (purchase, lead form submit, or similar)
  • Editor or Analyst role in GA4
  • A spreadsheet tool such as Google Sheets for recording findings
  • Optional: Hotjar or Microsoft Clarity for session recordings, GA4 Looker Studio connector for reporting

Step 1: Confirm Your Conversion Events Are Firing Correctly

A CRO audit built on broken tracking produces wrong conclusions. Before touching a single report, verify that your key events are recording accurately.

How do you check event accuracy in GA4?

Open GA4 and navigate to Admin > Data display > Events. Look at the event count for your primary conversion event over the past 28 days. Cross-reference this figure against your CRM, Shopify order count, or form submission log for the same period.

A discrepancy above 5% signals a tracking gap. Common causes include missing event tags on thank-you pages, duplicate GA4 measurement IDs, or server-side conversions not being imported. Fix these before proceeding. Every step that follows depends on clean data.

Common pitfall: Teams often audit a secondary metric like page views while their primary conversion event is misconfigured. Always audit the event that maps directly to revenue or leads first.

Step 2: Map Your Conversion Funnel in GA4 Explorations

GA4 Explorations give you a visual funnel that standard reports cannot produce. This is where you find the exact step where visitors drop off.

How do you build a funnel exploration in GA4?

Go to Explore > Funnel exploration. Add a new exploration and select the Funnel technique. Define each step of your funnel using the following logic:

  • Step 1: Session start (or landing page view for entry-point analysis)
  • Step 2: Product or service page view (page_view event filtered by page path)
  • Step 3: Initiation event (add_to_cart, begin_checkout, or lead form view)
  • Step 4: Completion event (purchase or form_submit)

Set the funnel type to Open so GA4 counts users who skip intermediate steps. This reflects real user behaviour more accurately than a closed funnel.

Once the funnel renders, read the abandonment percentage at each step. Record these figures in your spreadsheet. A healthy e-commerce funnel typically loses 60 to 70% of users between product page and cart. Anything above 85% drop at that stage warrants immediate investigation.

Pro tip: Duplicate the funnel exploration and switch the device breakdown dimension. Mobile funnels frequently show 20 to 30% higher abandonment than desktop on the same pages. This split tells you whether the problem is UX or content.

Step 3: Segment by Traffic Source to Find the Leaky Channels

Aggregate conversion rates hide channel-level problems. A 2% overall rate could mean paid search converts at 4% while organic social converts at 0.4%. Treating these the same wastes budget.

Which segments should you compare?

In your funnel exploration, apply a breakdown by Session default channel group. GA4 will show each funnel step side by side for Organic Search, Paid Search, Direct, Email, Organic Social, and Referral.

Look for two patterns. First, channels with high entry volume but low completion rates. These suggest a message mismatch between the ad and the landing page. Second, channels with low volume but high completion rates. These are candidates for increased budget or organic investment.

Record the conversion rate for each channel in your spreadsheet. You will use this to calculate the revenue impact of fixing each one in Step 5.

Step 4: Identify High-Exit Pages With the Landing Page and Page Path Reports

The funnel tells you where drop-off happens at a stage level. The page reports tell you which specific pages are responsible.

How do you find the worst-performing pages?

Navigate to Reports > Engagement > Pages and screens. Sort by Exits in descending order. Cross-filter by users who triggered your initiation event (add_to_cart or checkout_start) but did not trigger the completion event. This isolates pages that receive high-intent traffic but fail to close.

Export this list to your spreadsheet. Add two columns: monthly exit volume and your average order value or lead value. Multiply them. This gives you a rough upper bound of monthly revenue affected by each page's performance.

Prioritise any page where lost value exceeds $5,000 per month. These are your highest-leverage targets.

Common pitfall: Sorting by bounce rate instead of exits misidentifies the problem. Bounce rate counts single-page sessions. Exits count users who left after visiting that specific page in a multi-page session. For funnel pages, exits are the metric that matters.

Step 5: Quantify Drop-Off Impact With a Simple Revenue Model

CRO prioritisation fails when teams rely on gut feel. A basic revenue model turns audit findings into a business case that gets sign-off.

How do you calculate potential lift?

Use this formula for each funnel stage:

Monthly users reaching stage × current conversion rate = current conversions
Monthly users reaching stage × target conversion rate = target conversions
Target conversions − current conversions = additional conversions
Additional conversions × average order value = monthly revenue opportunity

For the target conversion rate, use the best-performing channel you identified in Step 3 as a realistic benchmark. If paid email converts at 5% and paid search converts at 2%, targeting 4% for paid search is defensible.

Enter these calculations for your top three to five problem areas. Sort by monthly revenue opportunity. The top item becomes your first sprint.

Step 6: Collect Qualitative Evidence to Explain the Data

GA4 tells you where users drop off. Session recordings and on-page surveys tell you why. Without the why, fixes are guesses.

What tools work alongside GA4 for qualitative data?

Install Microsoft Clarity (free) or Hotjar on your top exit pages. Filter recordings to sessions that triggered your initiation event but not the completion event. Watch at least 20 recordings per page.

Look for three patterns: rage clicks (indicating broken or confusing elements), scroll depth (indicating content that is not being seen), and form abandonment (indicating friction in required fields).

Add a one-question exit survey using Hotjar or Survicate to your highest-exit page. Ask: "What stopped you completing your purchase today?" Run this for two weeks. The answers surface objections that no analytics tool can detect on its own.

Record your qualitative findings alongside each page in the spreadsheet. This evidence strengthens your prioritisation and makes the case clearer when briefing designers or developers.

Step 7: Build Your Prioritised CRO Backlog

You now have quantitative data from GA4, a revenue model, and qualitative evidence from recordings and surveys. Combine these into a backlog your team can act on.

How should you structure the backlog?

Use a simple four-column table in Google Sheets or Notion:

  • Issue: A one-line description of the problem (e.g. "Checkout page loses 78% of mobile users")
  • Evidence: The GA4 metric or recording observation that supports it
  • Revenue opportunity: Monthly figure from your Step 5 model
  • Effort estimate: Small (copy change or button move), Medium (new page section), or Large (redesign or dev work)

Sort by revenue opportunity divided by effort. High-value, small-effort items go to the top. These are your quick wins. Large redesign items belong below them regardless of opportunity size, because they take longer to ship and validate.

If you work with an agency such as Lenka Studio, share this backlog directly. A structured brief that maps problems to revenue reduces scope creep and speeds up project kickoff.

Step 8: Set Up a GA4 Comparison Report to Measure CRO Progress

An audit without a measurement plan produces one-time improvements. Set up a report now so you can track whether your fixes actually work.

How do you track CRO results in GA4 over time?

Create a Looker Studio dashboard connected to your GA4 property. Include the following metrics on a weekly trend:

  • Overall conversion rate by channel
  • Funnel step completion rate for each stage you identified
  • Exit rate for your top three problem pages
  • Revenue per session by device category

Set a comparison date range to the four weeks before your first CRO fix shipped. Use this as the baseline for every subsequent comparison. Avoid comparing month-on-month without accounting for seasonality, especially for Australian and Canadian e-commerce businesses where seasonal peaks shift dramatically.

Review this dashboard every two weeks after each fix ships. Give experiments at least 14 days of data before drawing conclusions. GA4 does not run statistical significance tests natively, so use a free tool such as AB Testguide's significance calculator to validate results.

Frequently Asked Questions

How much data do I need before running a CRO audit in GA4?

A minimum of 30 days gives you enough data to identify trends, but 90 days is more reliable for spotting seasonal patterns. If your site has fewer than 500 monthly conversions, focus on qualitative data from session recordings rather than statistical funnel analysis, since sample sizes will be too small for confident conclusions.

Can I run this audit without GA4 Enhanced Ecommerce set up?

Yes, but your funnel will have fewer steps. You can still build a funnel exploration using page_view events and any custom events you fire on key actions. Enhanced Ecommerce events such as add_to_cart and begin_checkout give you a more granular funnel, so setting them up before the audit adds significant value.

How is a CRO audit different from a UX audit?

A CRO audit starts with conversion data and works backward to find friction. A UX audit starts with usability heuristics and expert review. Both are useful, and the best results come from combining them. This guide uses GA4 as the primary evidence source, then supplements with qualitative UX methods in Step 6.

What is a realistic conversion rate improvement to expect?

Industry benchmarks suggest that a structured CRO programme, run consistently over six months, typically produces a 15 to 40% lift in conversion rate for SMBs. Quick wins from copy and button changes can show results in two to four weeks. Larger structural changes take longer to validate.

Does this process work for lead generation sites, not just e-commerce?

Yes. Replace purchase with form_submit or phone_call as your completion event. The funnel steps change (landing page, service page, contact page, confirmation page), but the audit logic is identical. The revenue model uses your average lead value or customer acquisition cost instead of order value.

Next Steps

Run this audit on your highest-traffic funnel first. Export the backlog, sort by revenue opportunity, and ship your first fix within the same week. The longer a high-exit page sits unfixed, the more revenue it drains.

If your audit reveals issues that need design or development work, that is the right time to bring in specialists. At Lenka Studio, we work with SMBs across Australia, Singapore, Canada, and the US to turn CRO findings into shipped improvements. If you would like a second set of eyes on your audit results or your backlog, get in touch with our team and we can take it from there.

Not sure whether your broader brand presence is helping or hurting your conversion rates? Take the free Lenka Studio brand health assessment to see where your brand stands before investing further in paid traffic or CRO testing.