Conversion Tracking Mistakes Costing You Thousands

Let's be direct: if your conversion tracking isn't set up correctly, your Google Ads campaigns are essentially flying blind. The algorithm is optimising toward signals you've told it to trust — and if those signals are wrong, every pound or dollar you spend is making your performance worse, not better. We audit dozens of accounts every year at Workflow AI Advisors, and tracking errors are present in the overwhelming majority of them. Not edge-case errors. Fundamental ones.

This post covers the most common Google Ads conversion tracking mistakes we encounter, why each one damages your account performance beyond just "bad data," and what correct implementation actually looks like. If you run paid media for a business that cares about ROAS, this is worth reading carefully.

Why Conversion Tracking Errors Are More Destructive Than You Think

Bad conversion data doesn't just mean inaccurate reporting. It actively poisons your campaign optimisation. Google's Smart Bidding — Target CPA, Target ROAS, Maximise Conversions — is entirely dependent on your historical conversion data to make real-time bid decisions across millions of auctions. Feed it inflated, duplicated, or irrelevant conversion signals, and it will confidently optimise toward the wrong outcomes at scale.

We've seen accounts where duplicate conversion tracking was firing three events for every one actual purchase. The algorithm was chasing phantom conversions, CPCs were inflating, and the client couldn't understand why revenue wasn't tracking with what Google was reporting. Once fixed, their actual ROAS improved by over 60% — not because the campaigns changed, but because the machine was finally getting honest feedback.

Mistake 1: Counting Every Conversion Action as a Primary Conversion

This is the single most widespread error we see. Google Ads allows you to mark conversion actions as either "Primary" or "Secondary." Primary conversions are included in your Smart Bidding signals and reported in the main "Conversions" column. Secondary conversions report in "All Conversions" but don't influence bidding.

The mistake: teams set up multiple conversion actions — purchases, add-to-carts, button clicks, page views, phone calls, newsletter signups — and mark all of them as Primary. Your Smart Bidding strategy then tries to optimise for everything simultaneously, which effectively means it's optimising for nothing in particular.

The fix is intentional. Choose one or two actions that represent genuine business value — typically a purchase or a qualified lead form submission — and mark those as Primary. Everything else should be Secondary. This single change often produces measurable ROAS improvement within two to three weeks as the bidding algorithm reorients around meaningful signals.

Mistake 2: Duplicate Conversion Tracking Firing

Duplicate tracking happens when the same conversion event is being recorded by multiple tags simultaneously. The most common culprits:

• A conversion tag hardcoded directly into the website and also firing via Google Tag Manager
• Multiple Google Ads accounts each tracking the same conversion separately
• A Google Analytics goal imported into Google Ads and a native Google Ads tag tracking the same event
• Old legacy tags that were never removed during a site redesign

Duplicate conversions inflate your reported numbers dramatically and cause Smart Bidding to overpay for clicks it believes are converting at a rate they're not. Use Google Tag Manager's Preview mode and Google's Tag Assistant to audit every tag firing on your confirmation or thank-you pages. If you see the same conversion being reported twice, you have a problem that needs fixing before you spend another day of budget.

Mistake 3: Tracking the Thank-You Page Visit Instead of the Actual Form Submission

This sounds basic, but it's surprisingly common — especially on sites where the thank-you page URL is accessible without completing the form. If someone navigates directly to /thank-you, bookmarks it, or lands there through a browser back-forward issue, your tracking records a conversion that never happened.

The correct approach is event-based tracking triggered on actual form submission — not page view. This means your conversion tag fires when the submit button is clicked and the form successfully posts, not when a URL loads. Google Tag Manager's form submission trigger, combined with a check for form submission success, is the reliable way to implement this. If your developer has hard-coded a redirect to a thank-you page, ensure that redirect only fires after confirmed server-side form receipt.

Mistake 4: Ignoring Cross-Device and Cross-Browser Attribution Gaps

A user searches on mobile, clicks your ad, browses, then converts two days later on desktop via a direct visit. Under last-click attribution in a cookie-only world, that campaign gets zero credit. Under data-driven attribution with enhanced conversions enabled, it gets appropriately weighted credit.

Many advertisers are still running campaigns under last-click attribution without enhanced conversions, which means they're systematically undervaluing upper-funnel and mobile touchpoints. This leads to budget being pulled from campaigns and keywords that are actually driving significant assisted revenue — because the data makes them look like non-performers.

Enhanced conversions — which use hashed first-party data like email addresses to stitch together cross-device journeys — should be enabled for every account where it's technically feasible. Combine this with data-driven attribution (available once you have sufficient conversion volume) and you'll get a substantially more accurate picture of what's actually working.

Mistake 5: Not Segmenting by Conversion Value

If your business has variable order values or lead quality, treating every conversion as equal is costing you money. An e-commerce store sending £5 and £500 orders to the same campaign with the same Target ROAS tells the algorithm nothing about which customers it should prioritise. A B2B business where a demo request and a content download are both marked as conversions of equal value is optimising toward whichever is easier to get — not whichever actually drives revenue.

The solution depends on your business model. For e-commerce, pass dynamic transaction values through your conversion tag using the value parameter from your order confirmation data layer. For lead generation, implement conversion value rules based on lead characteristics, or use offline conversion imports to feed back actual closed-deal values from your CRM into Google Ads. This is a core part of what we implement through our AI automation infrastructure — connecting CRM data back to ad platforms so bidding reflects real revenue, not just raw lead counts.

Mistake 6: Using Google Analytics Goals as Your Only Conversion Source

Importing GA4 conversions into Google Ads is convenient and better than nothing, but it introduces attribution model discrepancies, session-based counting differences, and delays that make Smart Bidding less responsive. Google Analytics attribution and Google Ads attribution operate on different models, and when you import GA4 conversions, you're often getting a diluted, lagged signal.

Native Google Ads conversion tracking — implemented via Google Tag Manager with proper configuration — gives you faster, more accurate, click-attributed conversion data that integrates more cleanly with Smart Bidding. Use GA4 imports as a secondary validation layer, not as your primary conversion source for bidding decisions.

Mistake 7: No Conversion Tracking Audit Process

Conversion tracking breaks silently. A site update removes a data layer variable. A developer deploys a new checkout flow that changes the thank-you page URL. A GTM container gets published with a tag accidentally paused. These things happen constantly, and most teams have no systematic process for catching them.

At minimum, you should be running a weekly sanity check comparing conversion volume against actual business metrics — revenue from your CRM or payment processor against conversions reported in Google Ads. A sudden drop or spike in conversion rate that doesn't correlate with actual business performance is a red flag. More robust implementations include automated monitoring that alerts you when conversion volume deviates more than 20% from its 7-day average — something we build into accounts as part of our paid media management infrastructure.

Mistake 8: Ignoring Phone Call Conversions — or Tracking Them Incorrectly

For service businesses, phone calls are often higher-intent conversions than form fills. But phone call tracking is frequently misconfigured in one of two ways: either it's not set up at all (meaning calls driven by ads are invisible to the algorithm), or it's set with too short a call duration threshold (meaning a two-second accidental dial counts as a conversion).

Google's forwarding numbers are the standard implementation for call tracking from ads. Set your minimum call duration threshold to a time that represents genuine engagement — typically 60 to 90 seconds for most service businesses. Import call data from a dedicated call tracking platform (CallRail, ResponseTap, etc.) into Google Ads as offline conversions if you want to include calls from your website that weren't directly from the ad click.

What Good Conversion Tracking Infrastructure Looks Like

When we build out conversion tracking for clients, the standard we hold ourselves to is this: every conversion action should be independently verifiable against a source of truth, fire exactly once per qualifying event, carry accurate value data, and be clearly classified as Primary or Secondary based on its relationship to actual revenue.

That means a properly configured Google Tag Manager container, enhanced conversions enabled, data layer implementation on all key pages, offline conversion imports for lead-gen clients, call tracking integrated at the right duration thresholds, and a monitoring layer that catches breakage before it compounds. Across the accounts we've rebuilt from broken tracking baselines, the average improvement in attributed ROAS — before changing a single campaign setting — is significant. You're not getting better results because the campaigns improved. You're getting better results because the algorithm finally knows what's actually working.

If you're unsure whether your current tracking setup is reliable, start with a full audit. Check Tag Assistant, pull a Conversion Action report in Google Ads, compare reported conversions against your actual business data, and look for duplication or gaps. Our SEO & GEO and paid media teams regularly find that clients who came to us for campaign optimisation actually needed tracking fixes first — and that those fixes alone moved the needle before any strategic changes were made.