Fix Your Conversion Tracking Before You Judge SEO or Ads

Before I will argue with anyone about whether their SEO or their Google Ads is paying off, I want to see how their conversions are tracked. Not because I am stalling - because if the tracking is wrong, every ROI number in the conversation is wrong, and we would be debating fiction. This is the least glamorous part of the job and the one I refuse to skip.

It matters more in B2B than almost anywhere, because the sales are large and the decisions made off the back of these numbers are big - cut this channel, double that budget. Getting the foundation right is part of my broader approach to SEO for B2B and technical companies. Here is what usually goes wrong and how I check it.

ROI is only ever as good as the conversion under it

Return on investment is a fraction: value out over cost in. The "cost in" is usually easy - you know what you spent. The "value out" depends entirely on counting conversions correctly, and that is where it falls apart. If your conversion count is inflated, ROI looks great and you over-invest. If it is undercounting or miscounting, you might kill a channel that was actually working. The number feels precise because it has decimal places. That precision is worthless if the input is wrong.

A confident ROI figure built on broken tracking isn't a measurement. It's a decision made on sand with a spreadsheet stapled to it.

The mistakes I see most often

Counting the wrong action

The classic one: a button click gets counted as a conversion when the actual meaningful action happens later. Picture a site where clicking "Request a quote" fires the conversion - but plenty of people click it, see the form, and never submit. You are counting intent as if it were outcome. In a technical B2B setting this is everywhere: opening a configurator versus actually completing it, starting a datasheet download versus finishing it, clicking a mailto link versus a real enquiry landing. Distinguishing the button click that opens a tool from the completed action inside the tool is the single most common fix I make.

Double-counting and firing on every page

A tag placed carelessly in Google Tag Manager can fire on every page load, or fire twice, or fire on a thank-you page that people refresh. Suddenly you have three conversions from one enquiry, and organic search looks like a hero. Nobody notices because the number went up and up is the direction everyone wanted.

Not filtering bots and internal traffic

If your own team, your developers, and a fleet of bots are all in the data unfiltered, your conversions and your traffic are both polluted. Clean this out first - it is the same hygiene point I make about tracing B2B leads honestly. Dirty data doesn't just add noise, it quietly changes the conclusion.

Every conversion weighted the same

A newsletter signup and a request-a-quote for a bundled order are both "conversions" in a default setup, counted equally. In B2B they are worlds apart in value. If you don't separate them, a channel that drives lots of low-value signups can look better than one driving a handful of genuine sales enquiries.

A practical setup order

When I set this up in GA4 (Google Analytics 4, Google's current analytics platform) and Google Tag Manager (the tool that controls when tracking tags fire), I work in this order:

1. Decide what a real conversion is. Write down the handful of actions that genuinely represent value - a submitted quote request, a completed enquiry - not every click on the way there.
2. Filter out the noise. Exclude internal IPs, known bots and referral spam before you trust a single figure.
3. Fire tags on the true completion, not the intent. Track the form submission or the confirmation, not the button that opens the form.
4. Separate conversions by value. Distinguish a high-value sales enquiry from a low-value signup so you never average them together.
5. Verify with real tests. Actually complete each action yourself and confirm one clean conversion is recorded - no double counts, no misfires.
6. Only then calculate ROI, and only then compare channels.

Why this comes before the SEO-versus-ads argument

People love to debate SEO versus Google Ads and which delivers a better return. It is a fair question, but it is unanswerable until the tracking underneath both is trustworthy. I have seen a channel written off as a failure purely because its conversions were being undercounted while another channel's were being inflated. The strategy debate was moot - the scoreboard was broken.

The same goes for the deeper measurement work. Once tracking is solid you can have honest conversations about the things that genuinely can't be pinned down - the assisted paths and multi-touch reality I cover in measuring SEO results - and separate them from the things that now can. But you earn the right to those nuanced conversations by getting the basics right first.

So before you judge any channel, spend the unglamorous day fixing the tracking. It is not the exciting part of the work, but every number you care about is sitting on top of it. Get it wrong and you will confidently make the wrong call. Get it right and, for the first time, your ROI figures actually mean something. Cheers.