How a GA4 audit revealed where a business was losing tens of thousands of pounds — and why nobody noticed

I'm not writing about a business that wasn't working. I'm writing about one that was working quite well — and that's precisely why nobody went looking for problems. Revenue was growing year on year, traffic was there, the marketing budget was being spent. Meanwhile, email wasn't converting, the blog was bringing in traffic without generating sales, and the majority of customers were browsing on smartphones — on a site that wasn't optimised for smartphones.

This is a case study from my GA4 audit. I'm not naming the company, sector, or any specific identifiers — the data has been anonymised and scaled. But the numbers are real, the proportions are real, and the mechanics of the problem are real.

Where the numbers come from — the basis for this analysis

Over the course of more than a year, I produced regular GA4 reports for an e-commerce business. I had access to raw data from four sources: monthly Master Analytics Reports (MAR), a deep e-commerce audit, a year-on-year analysis, and weekly operational reports. Together that's several hundred pages of first-party data — not external studies, not benchmarks without context, but real data from a real business over a real period of time.

Where the business was losing money — three places GA4 shows directly

Problem one: an email list of 17,000 subscribers that barely sold anything

The business had a list of 17,000 subscribers and sent around 15 newsletters per year. On paper — a solid channel. In GA4 — a different story entirely.

The CTR (click-through rate — the percentage of recipients who clicked a link after receiving the email) was 0.3–0.5%. The benchmark for e-commerce newsletters is 2–5% according to Campaign Monitor and Mailchimp data for 2024. That's a gap of 5 to 10 times. In the December Christmas campaign, where four emails were sent to over 17,000 subscribers each, the total number of clicks across all four sends was 133. From nearly 70,000 messages sent in total.

But there was a deeper problem still: the business was running its email marketing through a tool that provided no data at all — zero open rate, zero bounce rate, zero visibility into subscriber behaviour. Sending a newsletter without being able to measure its results is operating in the dark. You have no idea whether people are opening, whether messages are going to spam, whether the list is active or dead. In that situation, every decision about content, frequency, and segmentation is a guess. Without data there is no optimisation — there is only spending a budget in an unknown direction.

The Christmas campaign pattern was the most telling symptom: the email sent with 8 days to go until Christmas had a CTR of 0.31%; the one sent on Christmas Eve itself — 0.09%. The more urgent the call to action, the fewer people responded. That is a classic sign of email fatigue — a list that has stopped trusting the sender's communications.

An additional problem: email tracking was not working correctly. For most of the year GA4 could not see any traffic from the newsletter — sessions attributed to the Email channel in the standard view showed zero, because the links in the messages had no UTM parameters. The estimated real traffic was around 2,500 sessions per year, against a potential of 12,000–24,000 sessions if CTR had been at benchmark level.

Problem two: the blog was generating 20–30% of all traffic and almost no sales

The business ran a blog. The articles had solid organic visibility — the top posts were getting 2,000–3,000 sessions per month each. Together the blog accounted for 20–30% of all site traffic.

The conversion rate from that traffic: practically zero. Articles with several thousand sessions each were producing zero sales — without exception.

For comparison: the homepage at a similar session volume generated hundreds of purchases and over £33,000 in revenue — at a conversion rate of 2.19% versus 0% from the blog. When I looked more closely at those articles, the answer was obvious: they had no CTAs, no product links, no mechanism whatsoever to guide the reader any further. An article could mention a specific service by name multiple times — without a single link to that service in the shop. People arrived, read, left. The backlinks were working, the SEO was working, the traffic was there. Nobody had simply thought about what happened to that traffic once it arrived on the page.

This is not a content problem — it is a conversion architecture problem. A blog without CTAs, without product recommendations, and without internal linking to the offer is effectively free advertising for competitors who will show up in the user's next search.

Problem three: 78% of traffic came from mobile, which converted 37% worse than desktop

This is a number that appears in many audits and never fails to make an impression on business owners.

From my data: mobile generated nearly 78% of all sessions. Conversion rate on mobile: 0.95%. Conversion rate on desktop: 1.51%. Desktop converts 59% better. At these traffic volumes, the mobile-desktop gap alone costs the business sums that make an impact in every management report.

This pattern is well established across the industry. Baymard Institute data from 2024 indicates that the mobile conversion gap in e-commerce averages 30–50% — which aligns precisely with what I found in this business. Mobile users abandon their baskets more readily, move through checkout more slowly, and give up more often at payment forms.

And there is a broader issue here beyond this one business: most small business websites are still not genuinely optimised for mobile. Mobile-first has been the standard for years, but in practice many businesses have a site that "works on a phone" in the sense that it loads — not in the sense that it guides a user through to purchase. When I visit a site on my phone and have to fight my way through it, I leave. And I don't come back. That business has lost me — regardless of what it's selling.

A separate observation from the YoY analysis: after the site was optimised for the mobile funnel, performance improved — the add-to-cart rate rose by over 50%, and checkout completion after adding to the basket improved noticeably. That is proof the problem lay with the site, not the users. When the site started working properly on phones — people started buying.

The full picture of losses — and how I calculated it

Summing up the audit calculations: email was running at a CTR 5–10 times below benchmark. Mobile converted 37% worse than desktop — with a 78% share of traffic. The blog generated 20–30% of all traffic at near-zero conversion. The basket was being abandoned more frequently than the industry average by over 10 percentage points. Returning users — who convert three times better than new visitors — accounted for less than 5% of traffic.

Each of these gaps is measurable. Together they paint a picture of a business leaving tens of thousands of pounds on the table every year — not as a guaranteed profit figure, but as the cost of data being available and nobody looking at it.

Implementing fixes is a separate story: new tools, people who know what they're doing, time for testing and monitoring. None of this happens in a month. But the starting point is always the same — knowing where the problem is. And roughly what it's costing.

Why nobody noticed — what stops businesses from reading their own data?

I ask myself this question at every audit. The answer is always similar.

First: the business was watching revenue, not channel efficiency. If sales are growing year on year, there's no obvious reason to go looking for problems. But GA4 doesn't just say "how much did you sell" — it says "where did the people who bought come from" and "where did the people who didn't buy come from, and why".

Second: no UTM tags on email links meant GA4 couldn't see newsletter traffic. That is a classic trap — if you're not measuring a channel, you can't see that it isn't working. In this case, for most of the year the Email source in GA4 showed zero sessions, because the links in the messages had no UTM parameters.

Third: the blog looked fine in SEO reports (organic traffic growing, articles in the top 10). No SEO report measures whether that traffic actually buys. For that you need GA4 with landing page segmentation and conversion attribution.

Fourth: mobile "works" — you can place an order, pay, receive a confirmation. The fact that it converts 37% worse than desktop is invisible without comparing CR per device. Most businesses don't make that comparison.

What this means for you — how to check whether you have the same problem

Three questions you can check in GA4 in 15 minutes:

1. Email channel: go to Acquisition → Traffic Acquisition, set a 12-month range, and find the Email row. If you see zero or less than 1% of total traffic — you have a UTM problem or a newsletter problem.
2. Mobile vs desktop: go to Explorations → create a Free Form with the Device Category dimension and Sessions + Conversions metrics. Calculate the CR for each device type. If the gap exceeds 40% — you have a mobile UX problem.
3. Blog: go to Engagement → Landing Pages, sort by sessions, and check the conversions against URLs containing /blog/. If you see thousands of sessions and zeros in conversions — the blog is attracting the wrong search intent and not driving any sales.

If the answer to all three is "I don't know" or "I've never checked" — that is precisely the moment to change that.

FAQ — common questions about GA4 audits

What is a GA4 audit and what can it reveal?

A GA4 audit is a systematic analysis of Google Analytics 4 data — traffic, sources, user behaviour, and conversions. It reveals where you are losing sales: which marketing channels aren't working, where users abandon the site, and what that costs in real money. Simply having GA4 installed is not enough — you need to know what to look for.

What should email CTR be in e-commerce?

The benchmark CTR for e-commerce newsletters is 2–5% according to Campaign Monitor and Mailchimp data for 2024. Results below 1% indicate a serious problem — with content, list segmentation, or send frequency. A CTR of 0.2% means that for every 10,000 messages sent, only 20 people click.

Why does mobile convert worse than desktop?

In e-commerce, mobile typically converts 30–50% worse than desktop due to harder navigation, smaller tap targets, slower loading, and a less comfortable checkout process on a small screen. Baymard Institute data from 2024 confirms this pattern as an industry norm. The problem is not the users — it is the website.

Can a blog hurt an online shop's sales performance?

A blog does not hurt directly, but it can be a budget trap. If it generates 20–30% of all traffic with near-zero conversion, the question is why. The most common answer: no CTA, no product links, no mechanism that guides the reader further. An informational article that points to no next step is a dead end. GA4 lets you measure this precisely and decide: fix the blog or redirect the content budget elsewhere.

How do I check in GA4 how much my shop is losing through weak email marketing?

In GA4, go to Acquisition → Traffic Acquisition and filter by the Email channel. Check sessions, conversions, and revenue. Then calculate: how many sessions should email generate at a CTR of 2–5% × average store CR × AOV = the potential revenue you are leaving on the table. If your email generates a fraction of that — the difference is a concrete loss you can quantify to the penny.

Sources

External data used in this article:

• 2–5% — e-commerce newsletter CTR benchmark. Source: Campaign Monitor, "Email Marketing Benchmarks Report 2024"
• 2–5% — CTR benchmark confirmed independently. Source: Mailchimp, "Email Marketing Statistics and Benchmarks", 2024
• 30–50% — mobile conversion gap in e-commerce as industry norm. Source: Baymard Institute, "Mobile E-Commerce UX" — study across over 40,000 shopping sessions, 2024
• 73% — average basket abandonment rate in e-commerce. Source: Baymard Institute, "Cart Abandonment Rate Statistics", 2024
• No UTM = no data — channel attribution best practice. Source: Google Analytics Help, "About attribution and attribution modelling", 2024
• 15–30% — share of returning users in mature e-commerce as a reference point. Source: proprietary data from multi-year audits and Klaviyo industry benchmarks, "E-commerce Benchmarks Report", 2024