How to Track Conversions in Google Ads?

What “conversion tracking” really means in Google Ads (and why it’s the foundation of ROI)

In Google Ads, conversion tracking is the system that tells you which clicks (and eligible views) turned into real business outcomes—like a purchase, a lead form submission, a phone call, or an offline sale. Without it, you’re optimizing on shallow metrics (clicks, CTR, traffic) instead of outcomes (revenue, qualified leads, profit).

Practically, Google Ads is built around two layers: conversion actions (the specific things you measure, like “Purchase” or “Lead”) and conversion goals (the groupings used for campaign optimization and reporting). The most important concept to get right early is which conversions are Primary (used for bidding and shown in the main “Conversions” columns) versus Secondary (kept for observation in “All conv.” without steering automation).

Step-by-step: How to set up website conversion tracking in Google Ads

Step 1: Create a conversion action from the Conversions area

Inside your Google Ads account, go to the Goals section, open Conversions, and start the workflow to create a new conversion action for website activity. Google Ads will prompt you to add your website domain and then scan it to detect whether you already have a Google tag in place and/or whether there’s an existing Analytics setup that can be connected.

This “scan-first” approach is the fastest way to avoid duplicate tags and to reuse existing site measurement where appropriate—especially if your business already has mature Analytics event tracking.

Step 2: Choose the right measurement method (URL-based, code-based, or Analytics-based)

You generally have three practical options, and the “best” one depends on how your site behaves.

If your conversion is completed on a dedicated confirmation page (like /thank-you), a URL-based setup is usually the quickest. You define the final URL (or a URL rule like “contains thank-you”), and the Google tag can recognize that page load as the conversion.

If you need to track something that doesn’t reliably create a unique URL—like a button click, an embedded form, or an AJAX submission—use the manual code-based option (often referred to as implementing an event snippet / event tag). This method also gives you the most control for passing details like transaction value, currency, order ID, and custom parameters.

If you already manage events in Analytics (GA4), you can create Google Ads conversions based on Analytics key events. This route can reduce duplication and keep event definitions consistent across teams. Two important realities to plan for: imported Analytics-based conversions may be set up in a way that prevents double-counting for bidding, and some Google Ads-only measurement features (like certain view-through reporting) require native Ads tagging rather than relying exclusively on imported Analytics conversions.

Step 3: Install the Google tag correctly (and avoid duplicate measurement)

Google Ads conversion measurement is typically a combination of (1) a sitewide Google tag and (2) a conversion-specific event implementation (either a codeless/URL rule or an event snippet / tag firing on the conversion action). The Google tag should be present across your site so Google Ads can attribute conversions back to ad interactions as reliably as possible.

If you deploy through a tag management system, make sure you’re using the modern Google tag setup inside your container (not a patchwork of legacy implementations). For many accounts, the missing piece that breaks attribution isn’t the conversion tag—it’s the click data handling. That’s why setups using a tag manager often require a conversion linker configuration so ad click information can be stored properly and then read when the conversion happens.

If you use an ecommerce platform integration (for example, a native “Google & YouTube” style app), treat it like a full measurement system, not “just another tag.” Platform integrations can automatically create conversion actions and may set some as account defaults. The most common mistake I see is leaving old tags running in parallel, which inflates results and confuses Smart Bidding.

Conversion settings that matter most (because they directly affect Smart Bidding)

Primary vs Secondary: decide what should actually steer automation

Google Ads reporting and bidding will only use the conversion actions that are configured to be eligible for optimization in your campaigns. As a rule, keep Primary conversions limited to outcomes you genuinely want the algorithm to chase. Everything else—micro-conversions, upper-funnel engagement, early-stage actions—often belongs as Secondary so you can still analyze it without letting it distort bidding.

Also note a nuance that trips up advanced advertisers: if you build custom goals for campaigns, actions included in that custom goal can be used for bidding even if they were otherwise treated as secondary in other contexts. This is powerful when used intentionally, and dangerous when done accidentally.

Conversion value and Count: choose what “success” means mathematically

For lead gen, you’ll often start with a flat value per lead (even if imperfect) so you can compare campaigns and inform automation. For ecommerce, you want dynamic revenue values and transaction IDs to reduce duplication risk and improve profit-based optimization.

The Count setting is equally important. If you sell one-time services or want to count one lead per ad interaction, “one” may be appropriate. If multiple purchases per click are normal, “every” is usually the right choice. This single setting can dramatically change reported performance and bidding behavior.

Attribution and conversion windows: match settings to your real sales cycle

Google Ads lets you control key timing and credit-allocation settings at the conversion action level, including click-through conversion windows and (where applicable) view-through and engaged-view windows. If you have a longer consideration cycle, tightening windows too far can underreport performance and train Smart Bidding on incomplete data. If you have an impulse purchase, windows that are too long can over-credit campaigns that weren’t actually decisive.

For attribution, many advertisers now rely on data-driven attribution (when available for their setup) or choose a simpler model when needed. The point isn’t to chase a “perfect” model; it’s to pick one that stays stable enough for optimization and matches how you make budget decisions.

Beyond the website: calls, offline conversions, and privacy-first measurement

Track phone calls (from ads and from your website)

If calls are part of your funnel, set up call measurement intentionally—because “calls happened” and “calls attributed correctly” are not the same thing.

For calls driven directly from ads (call assets/call-focused formats), Google can report call-based conversions when forwarding is enabled and your settings support it. For calls that happen after someone lands on your site, you can track calls to a phone number on your website by implementing the appropriate website call measurement setup (often requiring number replacement and careful formatting so the number on the page matches what’s configured).

If your sales team qualifies calls in a CRM, you can also import call outcomes later—just remember that importing only works for eligible call sources and must be done within the conversion window you’ve set (with an upper limit that can be up to 90 days depending on configuration).

Track offline leads and sales (and why “enhanced conversions for leads” is now the practical default)

If you generate leads online but close deals later (phone, showroom, sales team, invoicing), you should not stop at “lead submitted.” Importing offline outcomes is one of the biggest ROI unlocks in Google Ads because it trains Smart Bidding on what actually becomes revenue.

Modern best practice is to use enhanced conversions for leads rather than older offline conversion import alone, because it can improve durability and matching by using consented, hashed first-party data in addition to click IDs where available. Operationally, you still need clean processes: capture the right identifiers at lead time, map lifecycle stages (lead → qualified lead → converted lead), and upload within required time limits. Also, be disciplined about deduplication rules so reuploads don’t inflate performance.

Consent mode, user consent, and keeping measurement resilient

Measurement has become more privacy- and consent-dependent, especially for traffic in regions with strict requirements. At a minimum, your business needs a clear consent collection mechanism where required, and you should implement a consent-aware tagging approach so your measurement doesn’t collapse when users decline cookies.

Consent mode can be implemented in a basic approach (tags blocked until the user chooses) or an advanced approach (tags load with consent signals and adjust behavior). The advanced approach is often what serious advertisers move toward because it can preserve more modeled measurement while respecting user choices—assuming your consent signals are configured correctly for advertising use cases such as measurement and personalization.

Verify, troubleshoot, and keep your conversion data clean (the checklist I use in audits)

How to confirm conversions are recording correctly

Don’t rely on “I installed the tag” as proof. Use the platform’s diagnostics and a proper debugging workflow. In Google Ads, you can review conversion action status and open diagnostics for deeper details. On the site side, use Tag Assistant-style debugging to confirm the Google tag loads, the conversion event fires, and the right parameters (value, currency, transaction ID, identifiers) are present.

• Check the conversion action status and confirm it shows as recording (not inactive, unverified, or tag missing).
• Test an end-to-end path (ad click → landing page → conversion) rather than triggering a conversion event in isolation.
• If using a tag manager, confirm the Google tag configuration is present and that click data handling is correct (often via a conversion linker setup), especially on landing pages.
• Look for duplicates: old hardcoded tags, platform-app tags plus GTM tags, multiple conversion actions measuring the same event, or both Ads-native and imported Analytics conversions counting the same outcome for bidding.
• Cross-domain journeys: if users land on one domain and convert on another, ensure cross-domain linking is configured so attribution isn’t broken mid-funnel.
• Consent and blockers: verify what happens when consent is denied, and test with/without browser ad blockers because they can prevent tags from running during QA.

The “measurement maturity” upgrade path (what to do after basic setup)

Once your core conversions are stable, the next level is quality. First, tighten the definition of what counts as a primary conversion so Smart Bidding learns the right goal. Next, add conversion values (even estimated values for lead gen) so you can optimize toward value, not just volume. Then, implement enhanced conversions (web and/or leads) where appropriate and compliant, and finally, close the loop with offline outcomes so Google Ads optimizes toward revenue and not just form fills.

When you do this in the right order—accuracy first, then depth—you’ll see cleaner reporting, more stable automated bidding, and better ROI decisions that hold up when budgets scale.

‍

Once your Google Ads conversion tracking is set up correctly—choosing the right conversion actions, avoiding duplicate tags, and aligning Primary vs Secondary goals—you still need a reliable way to keep everything clean as campaigns evolve (new landing pages, new forms, changing attribution settings, offline imports, and consent constraints). Blobr connects to your Google Ads account and continuously analyzes performance and setup signals, then translates best practices into practical recommendations you can review and apply at your pace; and if you want help beyond tracking, its AI agents can also tackle day-to-day optimization work like matching keywords to the right landing pages or improving landing-page alignment with your ads, so your measurement stays actionable, not just “installed.”