We Audited 11 Websites. Only Two Could Tell Us How Many Enquiries They Got.

A lot of websites can’t count their enquiries properly. If you can’t count them, you can’t tell whether or not your marketing is working.

A dashboard that says every visit converts

Let’s begin with an example: a website whose GA4 showed over 2 million key events and a 100% key event rate, all because page_view was marked as a key event. In plain English (for the blissfully SEO-unenlightened), a business had no way of differentiating between a user loading up a page and someone actually making an enquiry. Which is not ideal when it comes to tracking success. Sure, revenue generated over time can generally act as a strong barometer of good or bad fortune. That goes without saying. But without a thorough understanding of user behaviour, a business can’t implement a proactive approach to bettering their own growth.

Over the course of the last two months we’ve audited 11 different websites. Nine of them were unable to reliably count their enquiries or bookings. It’s a story we’ve seen across myriad clients throughout our years of digital marketing. What we want to get to the crux of is, why does it happen, how does it happen and – the big’un – how does one avoid it?

What we looked at

The 11 sites we looked at are naturally shielded by anonymity, but they fall across a variety of industries and degrees of economic clout. We took data from Google Analytics, Google Search Console, Google Tag Manager, SE Ranking and from the sites themselves, each from July to September of this year. These are the cannot-go-without tools in the SEO shed for any business that wants to perform well in the wild west that is a search engine results page.

Five ways tracking goes wrong

Though made easier now with AI automation and guidance, setting up Google properties for a website can indeed be a tedious and time-consuming endeavour. And the evidence is in the haphazard and rushed nature in which the majority of properties are set up. To the untrained eye, Google’s tracking tools can look like hell on a dashboard, while its recommended processes can appear menial and, frankly, skippable. Rest assured that Google will not be forgiving if any of the following cardinal sins are committed.

The wrong things get counted

One site we analysed took job applications as 239 of 553 “leads”. This was an automation firm – not a recruitment agency. Regardless, for those scanning the data in Google Analytics, a steady increase in total leads over the course of a few months might appear a case for cautious optimism – at least on a surface level without proper scrutiny. In reality, nearly half of its lead data was bogus.

That’s not to say that the company didn’t want job applications, nor that the applications weren’t worth tracking. But conflating them with leads meant that the business had no accurate way of measuring progress within that metric. The job applications rendered the entire figure redundant when it should have been one of the company’s guiding lights.

Things get counted twice

The opposite problem can be just as misleading. One site we looked at loaded GA4 twice on every page. Another tracked its form submissions through a WordPress plugin and Google Tag Manager at the same time, so the same enquiry could be counted twice.

Nothing gets counted

Three of the 11 sites we looked at had a purchase event set up in GA4 that had never fired. Another property had no key events set up at all. Another’s Google Tag Manager container had been empty since 2023. In the case of the latter, it meant complete measurement blindness going back three years – an analytics blackout, if you will – wherein visitor traffic, pageviews and user sessions are all kept from the annals of recent history.

Losing the last step

Tracking might work for part of the customer’s journey, only to fall flat on its face when an enquiry or booking is completed. Thus, the connection between website activity and an actual outcome is lost.

A manufacturer’s form showed an on-page message instead of redirecting to the /thank-you/ page, which happened to be the only page that the Google Ads conversion fired on. The page had one view all year, yet the form logged 37 enquiries in a quarter. Clinics that were booking through portals on another domain could see clicks but not appointments.

A click on a booking button does not make a confirmed appointment. If the clicking user’s journey continues via another domain, the business risks losing visibility of the outcome that they are actually trying to measure.

Counting a fraction

It can be easy to miss if GA4 is seemingly bringing in the right data – especially if the right processes have been followed during the set-up. That being said, data in GA4 still should be corroborated with Search Console. We found one retailer’s GA4 records showed only about 7% of the organic visits that Search Console was reporting. Some gap is normal. Search Console counts clicks, while GA4 only counts visitors whose browser loads its tag, and in the UK, anyone who declines cookies may not be counted at all. A small shortfall is expected. Recording just 7% isn’t. This can be an especially frustrating problem, as there might be no obvious red flags in the GA4 data boxes, and the fix might be a convoluted one.

The good news, however, is that when it’s done, it’s done. GA4 properties should be monitored consistently, but if attention and care are paid during the setup process, monitoring will be all it needs in the long-term.

Why it costs high-value businesses most

To a business that sells industrial machinery or specialist equipment, for example, a handful of high-value deals might be all that’s closed in a month. Ergo, one genuine enquiry will likely be worth more to that business than a thousand website sessions. If they can’t distinguish between genuine sales enquiries and unrelated actions, it would be unreasonable to expect them to confidently judge which campaigns or channels are giving them the most commercially valuable results.

And inevitably the issue is exacerbated if advertising platforms are involved. Platforms like Google Ads use conversion signals to help optimise campaigns and bidding, so if GA4 is reporting brochure downloads, job applications or ordinary page views as its conversions, Google Ads gets a boatload of misleading signals about which actions to prioritise. The ripple effect can impact every facet of marketing.

Why nobody noticed

The typical evolutionary process for website analytics is stunted:

1. Configure with the launch of the website
2. Leave alone forever

If a website is not abandoned, it will evolve over time. The analytics have to evolve with it. New contact forms, redesigns, caching settings and new booking platforms will all need mirroring on GA4, as GA4 itself won’t know what it’s missing. The website might still appear healthy as customers go about their business, but do those numbers match the real outcomes? That’s the question. It doesn’t help when nobody owns the tracking. On one site we audited, no one could tell us who even owned the GA4 property.

Dashboards are designed to look trustworthy. They’re awash with complicated numbers and comparisons and projections and analyses. Don’t be fooled. It can look complete while it counts the wrong things or misses the events that matter.

Six checks to run on your own website

1. Submit your own contact form: Submit a test enquiry via your contact form and open GA4 Realtime to see whether the event actually appeared. If it’s a no-show, investigate the event configuration and confirm that the submission is actually being captured.
2. Review your key events: Head to your GA4 key events list and check which user actions are marked as conversions. Pageviews and newsletter sign-ups will unlikely be meaningful business outcomes for you.
3. Check where bookings are completed: Does the pesky booking button send visitors to another website or domain? If it does, you’ll need to find out if you can measure completed appointments as well as clicks. If only the click shows, you’re measuring intrigue rather than bookings.
4. Check phone and email clicks: Do prospective customers call or email you? You’ll need clicks on the phone number and email address tracked if so – although separately to completed conversions.
5. Compare analytics with your real enquiries: The true disparity between analytics and reality will come out in this wash. Gather up one month’s reported leads and compare them with the enquiries you actually received in your inbox, CRM or booking system. Where are the differences? What unrelated actions have been included? Which enquiries are missing?
6. Check how long GA4 keeps your data: By default, GA4 deletes the detailed data behind its custom reports after two months. In Admin, find Data retention and change it to 14 months. It takes under a minute.

All being well, the number of enquiries that your analytics reports should somewhat resemble the number that you see in your own systems. If they don’t, find out why and rectify.

Once you can count every enquiry, the next problem is replying before the customer goes somewhere else. We’re building a system that does exactly that for any business, and we’re happy to talk it through.