How Consent Mode Affects Conversion Tracking in Google Ads

Consent mode changes what your Google tags are allowed to collect when a visitor declines cookies, and that is why conversions can drop in Google Ads after a cookie banner goes live. Some conversions are no longer observed. Google fills part of that gap with modeling, which produces estimates rather than recovered individual records.

If you have seen a dip in reported conversions, or a consent mode warning in your account, you are probably dealing with this. The dip is not always a performance problem. Below is what changes in your data, what modeling can and can't restore, and how to check that your setup is working. Understanding how consent mode affects conversion tracking is mostly a matter of knowing which numbers are measured and which are estimated.

What Consent Mode Does to Your Tags

Consent mode is a feature that lets Google tags (the Google tag, Google Analytics, Floodlight, conversion linker) change their behavior based on the choice a visitor makes in your cookie banner. Without it, a tag either fires normally or doesn't fire at all. With it, the tag can run in a restricted mode that respects the visitor's choice while still sending limited, non-identifying signals.

The choice is passed to the tags through consent signals. These are the main ones, as named in Google's documentation as of 2026:

• ad_storage controls whether advertising cookies or identifiers can be written and read, which affects things like click attribution and remarketing.
• analytics_storage controls whether analytics cookies can be used to measure visits and behavior.
• ad_user_data controls whether user data can be sent to Google for advertising purposes.
• ad_personalization controls whether data can be used for personalized advertising, such as remarketing.

Each signal is set to granted or denied. Your site sets a default state before tags load, then updates it when the visitor makes a choice. Tags read the current state and behave accordingly.

What consent mode is not

It is not a cookie banner. A consent management platform (CMP) is the tool that shows the banner, records the visitor's choice, and stores proof of it. Consent mode only relays that choice to Google tags. It does not collect consent for you, and it does not make a non-compliant banner compliant.

It also doesn't override the visitor. If someone declines, consent mode is the mechanism that makes your tags honor that decision. Whether your setup meets legal requirements is a question for your legal counsel, not for the tag configuration.

Basic vs Advanced Consent Mode

Consent mode comes in two implementations, and the difference decides how much data Google has to work with when a visitor says no.

In basic mode, Google tags are blocked from loading until the visitor consents. If they decline, nothing is sent to Google from that visit, so there is nothing to feed modeling from those users.

In advanced mode, tags load when the page loads, with consent defaulting to denied. If the visitor declines, the tags send cookieless pings instead of setting cookies. Google describes these pings as carrying limited information such as whether consent was granted, and uses them, along with patterns from consented users, in its modeling. If the visitor accepts, the tags run normally.

Here is how the two compare:

• When tags load: basic waits for consent; advanced loads immediately with consent denied by default.
• What is sent on denial: basic sends nothing; advanced sends cookieless pings without cookies.
• Modeling inputs: basic gives Google no signals from declining users; advanced gives it the pings to help model the gap.
• Implementation effort: basic is simpler, often a matter of blocking tags until the CMP reports consent; advanced needs correct default and update calls, and tag behavior that reacts to each state.

A common misconception

Advanced mode does not let you track declined users as individuals. The pings are not a way around the visitor's choice, and they are not used to build a profile of someone who said no. Their value is aggregated: they help Google estimate how many conversions likely happened among unconsented traffic. Google's wording on exactly what the pings contain has been refined over time, so check the current consent mode documentation on Google's developer site before describing it to a client or in a privacy policy.

What Happens to Conversions When a User Says No

Take an illustrative journey. Someone clicks your Search ad, lands on your site, and declines cookies in the banner. They browse, leave, come back later, and buy.

With consent denied, Google Ads can't use cookies to link that purchase to the ad click, so it is generally not counted as an observed conversion. In advanced mode, the cookieless pings may contribute to a modeled estimate for traffic like this, but Google Ads doesn't point to this specific purchase as attributed to this specific click. The sale happened in your store or CRM. The ad platform just can't see it directly.

That is why reported numbers can fall after a banner goes live without any real change in performance. The same visitors are buying, but a share of them are now invisible to cookie-based measurement. Compare conversions before and after the banner date with that in mind, and expect the size of the drop to track your decline rate.

Two further effects follow:

• Returning visitors: without cookies, a second visit looks like a new one, so the connection between the original ad click and a later conversion weakens.
• Cross-device paths: someone who clicks on mobile and buys on desktop is harder to stitch together when cookies are unavailable.

Where enhanced conversions fit

Enhanced conversions are a separate feature that improves measurement by sending hashed first-party data, such as an email address a customer typed at checkout, to match conversions to signed-in Google users. They are meant to work alongside consent mode, not replace it, and they still depend on the right consent signals being granted. Requirements and setup options have changed over the years, so confirm the current rules in Google Ads Help before relying on them.

How Conversion Modeling Fills the Gap

Modeled conversions are estimates. Google looks at the behavior of consented users, such as how often a click leads to a conversion, and combines that with aggregated, non-identifying signals from unconsented traffic to estimate the conversions it couldn't observe. The model works at the group level, not the person level.

Eligibility

Modeling isn't automatic for every account. As of 2026, Google's documentation points to conditions such as an advanced consent mode implementation that sends the required signals, enough traffic and consented data for the model to be reliable, and, for some uses, correct regional consent settings. Google has also required consent mode for EEA audience features. The specific thresholds and the current status of those requirements are updated in Google Ads Help, so check there rather than relying on a number from a blog post, this one included.

How it shows up in reports

Modeled conversions are included in the standard conversion columns alongside observed ones. They aren't separated into their own labeled column the way you might expect, so you can't simply subtract them out. Google has changed some reporting behavior over time, so confirm how your account currently displays them.

What modeling can't do

• It doesn't recover individual conversions. You can't see which user or which order was modeled.
• Accuracy depends on volume. Small accounts with few conversions give the model less to learn from.
• Accuracy depends on consent rate. If most visitors decline, there are fewer consented users to base the patterns on.
• It is still an estimate. Treat it as a way to restore part of the picture, not the whole one.

That last point matters when you report to clients or stakeholders. A conversion total that mixes observed and modeled data is a good directional number, but your own sales records remain the check on it.

What This Means for Smart Bidding and Optimization

Smart Bidding strategies like Target CPA and Target ROAS learn from the conversions your account reports. If observed volume falls after a banner launch, the strategy has less data to learn from, and the mix of observed and modeled conversions can make performance less stable for a while. Expect a possible learning period and some noise, particularly on lower-volume campaigns.

A few practical habits help:

• Expect a reporting shift at launch. Avoid big bid target changes in the first days based on a dip that may be measurement, not performance.
• Use back-end data as the source of truth. Compare Google Ads conversions with orders, leads, or revenue in your store or CRM to see whether real results moved.
• Judge changes over longer windows. Short-term swings are harder to read when part of the data is estimated.

The part consent mode doesn't touch

Clicks, impressions, and cost are recorded regardless of consent, and so are the search terms that triggered your ads. That means search terms analysis stays reliable even when the conversion columns are noisy. A term that spent money with no relevance to your offer is junk whether or not conversion data is complete.

This is a good place to keep working while measurement settles. Adding negative keywords for irrelevant search terms, and choosing match types for the positive keywords worth keeping, protects spend without depending on consent-affected data. Keywordme is a Chrome extension that works inside the Search Terms Report in Google Ads, so you can do this cleanup in a few clicks without exporting to a spreadsheet. It doesn't change your consent or tracking setup.

How to Check Your Consent Mode Setup

Start by confirming the signals actually fire in the right order. Use Google Tag Assistant or the preview mode in Google Tag Manager, load your site in a fresh session, and look at the consent state at each step.

1. Check that a default consent state is set before any Google tags load.
2. Decline in your banner and confirm the signals stay denied and tags behave as your chosen mode (basic or advanced) intends.
3. Accept in a separate test and confirm an update event moves the relevant signals to granted.
4. Check Google Ads for any consent mode status or diagnostics messages. The exact location in the interface has moved over time, so look under your conversion and data diagnostics areas and confirm the current path in Google Ads Help.

Common mistakes

• No default before tags load. If the banner sets consent after the tags fire, the first hit may go out with the wrong state.
• No update after the choice. The visitor accepts, but the signals stay denied, so you lose data you were entitled to collect.
• Mismatched regional settings. Defaults that differ by region should match where your banner actually applies, or visitors may be treated inconsistently.

Measure the change

Add an annotation in Google Ads on the date consent mode or the banner goes live, so future you knows why the trend line moved. Then compare conversions before and after against your back-end numbers, not against Google Ads alone.

Audit the Setup, Then Protect Your Spend

Consent mode lowers the conversions Google can observe, and modeling puts part of that picture back as estimates. It won't give you every conversion, and it was never meant to. What you can control is whether the signals fire correctly, whether you're on the right mode, and whether you judge results against your own sales data.

Run through the checks above, annotate the launch date, and cross-check Google Ads against your CRM or store before changing bids. While you wait for the numbers to settle, spend is the thing you can protect, and search terms are the same no matter who consented.

Optimize Google Ads Campaigns 10X Faster. Without Leaving Your Account. Keywordme lets you remove junk search terms, build high-intent keyword lists, and apply match types instantly, right inside Google Ads. No spreadsheets, no switching tabs, just quick, seamless optimization. Start your free 7-day trial (then just $12/month) and take your Google Ads game to the next level.