Google Ads Remarketing: Practical Setup Guide for Ecommerce

70.22% of shopping carts are abandoned before the purchase is completed (Baymard Institute, September 2025). If your store gets 10,000 visits a month, that means more than 7,000 users leave without buying. Most Google Ads budgets chase cold traffic, ignoring the people who already know your brand and have already shown purchase intent.

This guide covers everything you need to set up and optimise Google Ads remarketing in 2026: campaign types, GA4 configuration, RLSA, the real state of cookies, and how to decide what to use when you also have Performance Max running.

In 30 seconds:

70.22% of carts are abandoned before payment (Baymard Institute, 2025)

Remarketing recovers abandoned carts: conversion rate rises from 8% to 26% (DemandSage, 2025)

Remarketing CTR in display is 10x higher than standard display (0.7% vs 0.07%)

Third-party cookies remain active in Chrome: Google reversed deprecation in July 2024

Allocate between 10% and 40% of your prospecting budget to remarketing

What is Google Ads Remarketing and How Does It Work?

Google Ads remarketing achieves a display CTR of 0.7%, compared to 0.07% for standard display ads — ten times higher (DemandSage / Meazy, 2025). The reason is straightforward. You’re showing an ad to someone who already visited your website, already saw that product, already considered buying it. It isn’t cold traffic; it’s traffic that knows who you are.

The mechanism works like this: when someone visits your site, a tag (the Google Ads tag or the GA4 tag) records that visit in a cookie or identifier. That person gets added to an audience. When they browse other websites on the Google Display Network, watch YouTube, or return to search on Google, your ads can appear specifically for them.

What about third-party cookies? There’s been a lot of confusion in the industry. Google announced their removal multiple times between 2020 and 2024. In July 2024, they reversed that decision definitively: third-party cookies remain active by default in Chrome (CookieYes, July 2024). For 2026, cookie-based remarketing remains the foundation. That doesn’t mean ignoring Consent Mode — more on that shortly.

Person looking at a screen showing an abandoned shopping cart, illustrating the online shopping abandonment problem

What Types of Google Ads Remarketing Exist and When Should You Use Each?

The average remarketing conversion rate is 3.8%, compared to 1.5% for cold traffic (DemandSage / Oracle, 2025). But you don’t hit 3.8% by using the wrong campaign type. Each format has its own logic, and mixing them without a clear rationale is one of the most common mistakes I see when I audit accounts.

Standard Display Remarketing. Shows banners to previous visitors as they browse the Display Network. Useful for maintaining brand visibility and reactivating visitors who landed but didn’t engage much. Recommended minimum audience: 1,000 users.

Dynamic Remarketing. The most powerful format for ecommerce. Instead of generic banners, it shows the exact product the user viewed or added to their cart. Requires a product feed linked to Google Merchant Center. The difference in relevance is significant.

RLSA (Remarketing Lists for Search Ads). Not display: it adjusts your bids in search based on whether the user already knows you. A visitor who was on your site and comes back searching your category has much higher purchase intent. We cover this in detail in the next section.

Video Remarketing. Reaches people on YouTube who have already visited your site or watched your videos. Works well for products with longer decision cycles or brands with active video content.

Customer Match. Upload your customer list (emails) and Google matches them against Google accounts. Ideal for upsell campaigns, reactivating lapsed customers, or exclusions (avoiding spend on people who already bought).

Type	Channel	Best use case	Minimum audience
Standard Display	Display Network	Brand visibility and general reactivation	100 users
Dynamic	Display / Shopping	Cart recovery, product page visitors	100 users
RLSA	Search	Adjust bids based on visit history	1,000 users
Video	YouTube	Long decision cycles	1,000 users
Customer Match	Search / Display / YouTube	Upsell, retention, exclusions	1,000 emails

Diagram of audience segmentation on a computer screen, representing the different audience types for remarketing

How Does RLSA Work and How Should You Use It in Search Campaigns?

The average retargeting CPC on Google is $0.66–$1.23, versus $1–$2 for prospecting search (DemandSage / WebFX, 2025). With RLSA you don’t automatically pay less, but you can bid more intelligently: bid up for users who already know your brand and are more likely to convert, without changing anything for cold traffic.

RLSA requires a minimum of 1,000 users in the remarketing list during the last 30 days. If you don’t reach that volume, the bid adjustment simply doesn’t activate and the campaign runs normally.

Bid adjustment framework by segment:

Recent visitors (last 7 days, no conversion): +20% to +50%. They know you, but there’s no strong intent signal yet.
Product page visitors (no cart add): +50% to +100%. Showed specific interest in a particular product.
Cart abandoners (last 72 hours): +100% to +200%. Very high purchase intent; the moment to reconnect is now.
Previous customers (Customer Match): +150% to +300%. The probability of repeat purchase justifies aggressive bids.

Does it make sense to apply the same adjustment to everyone? No. A user who saw the homepage 20 days ago doesn’t deserve the same adjustment as someone who added to cart yesterday.

There are two modes for RLSA:

Overlay (observation). The campaign still reaches everyone, but applies different bid adjustments depending on whether the user is in a list or not. This is the recommended starting point.

Target only. The campaign only reaches users who are in your lists. More restrictive, but useful for creating specific campaigns with entirely different messages and bids for each segment.

If you use automated bidding strategies alongside RLSA, review how Smart Bidding interacts with manual adjustments before combining them.

CTR CVR

Cold traffic 0.07%

Remarketing 0.70%

Cold traffic 1.5%

Remarketing 3.8%

Cold traffic Remarketing

Remarketing delivers 10x higher display CTR and 2.5x higher conversion rate than cold traffic. Source: DemandSage / Meazy / Oracle, 2025.

How Do You Set Up Dynamic Remarketing for Ecommerce with GA4 in 2026?

Dynamic remarketing can lift abandoned cart conversion rates from 8% to 26% (DemandSage / ReadyCloud, 2025). That’s not a minor tweak — it’s tripling the recovery of sales you nearly had closed. The key is showing the exact product the user left behind, at the right moment, with the right message.

Since December 2025, Google activated native GA4 integration with dynamic remarketing for ecommerce. Previously, you had to configure custom event parameters and map them manually in Google Ads. Now, if your GA4 has standard ecommerce events active (view_item, add_to_cart, purchase), Google Ads can read that data directly and build dynamic remarketing audiences without additional tag configuration.

Setup steps for 2026:

Verify that GA4 is recording standard ecommerce events: view_item, add_to_cart, begin_checkout, purchase.
Link your GA4 property to your Google Ads account (Google Ads Admin → Linked products).
In Google Ads, import GA4 audiences: “Users who added to cart without completing purchase” (30-day window) and “Product page visitors” (7–14 day window).
Link the Google Merchant Center product feed to the dynamic Display campaign.
Create the dynamic remarketing campaign and assign the imported GA4 audiences.

On Consent Mode v2 in the UK and EU. Since January 2024, Consent Mode v2 is mandatory for any advertiser using Google tags in the EU. In 2026 there’s no grey area: if you don’t have Consent Mode v2 implemented with your CMP (Cookiebot, Axeptio, CookieYes, or another), your tag can’t legally be used to build remarketing audiences from users who declined cookies. Google can model behaviour for conversions, but remarketing audience lists are affected by consent fragmentation.

Get Consent Mode v2 working before worrying about any bid optimisation. Without the legal foundation, everything else is built on sand.

Computer screen showing data analytics and GA4 audience configuration for remarketing

Performance Max vs Standard Remarketing: Which Should You Use for Ecommerce?

In retail Performance Max campaigns, 90% of median cost share comes from feed-based ads (Smarter Ecommerce, 4,000+ campaigns, April 2025). That means PMax in ecommerce is largely a dynamic Shopping machine. And yes, PMax includes automatic remarketing — but that “automatic” comes at a price: you lose visibility and control over who sees what, how often, and at what stage of the buying cycle.

When you have PMax running, Google includes audience signals (including your recent visitors) in its overall optimisation. It’s not that it skips remarketing: it does it opaquely, mixing cold and warm audiences without letting you separate performance or tailor messages.

When to keep standard remarketing alongside PMax:

When you have more than 1,000 cart abandoners per month: the volume justifies a dedicated campaign with a specific message.
When you want to control impression frequency — something PMax doesn’t let you configure directly.
For high-value Customer Match lists: PMax can use them as a signal, but a dedicated RLSA or Display campaign gives you full control over the upsell message.
If you sell products with decision cycles longer than 7 days, standard remarketing with progressive message sequences outperforms PMax’s automated logic.

Budget allocation framework. Assign between 10% and 40% of your prospecting budget to remarketing, depending on your traffic volume and store abandonment rate. With high traffic and high abandonment, 40% makes sense. With low traffic or impulse-purchase products, 10–15% is enough.

If you want to understand how PMax works under the hood, the complete Performance Max guide for ecommerce covers its structure, audience signals, and recommended configuration.

90% feed-based

Feed-based (Shopping/Dynamic): 90% Text/image/auto video: 10%

In retail PMax, 90% of median cost comes from product feed-based ads. Source: Smarter Ecommerce, 4,000+ campaigns, April 2025.

Common Remarketing Mistakes That Are Costing You Money

Only 15% of global marketing professionals felt fully prepared for a cookieless world in March 2025 (Deloitte, via Workshop Digital, 2025). The remaining 85% were working, to some degree, with incorrect or incomplete configurations. What’s the result? Fragmented audiences, wasted budget, and optimisations built on a broken foundation.

The most common mistakes I find:

No audience exclusions. Showing remarketing ads to people who bought yesterday is direct waste. Always exclude recent converters from your cart abandonment campaigns. Create a “buyers — last 30 days” list and exclude it from your prospecting campaigns, not just remarketing.

No frequency cap. A user can see your banner 30 times in a day if you don’t set a limit. It saturates, irritates, and damages brand perception. In Display, start with a maximum of 3–5 impressions per user per day.

Bid adjustments never revisited. Many accounts set up RLSA at the start and never touch it again. Adjustments that made sense with data from 6 months ago may be overestimating or underestimating current audiences. Review bid adjustments every 4–6 weeks.

Ignoring Consent Mode v2. If your CMP isn’t properly configured, you’re building audiences on incomplete data. It’s not just a legal problem — it’s a performance problem.

If you want to see what other configuration failures affect campaign performance, check the most common Google Ads mistakes I find when auditing accounts.

Want to Review the Remarketing Setup in Your Account?

If you have active Google Ads campaigns and aren’t sure whether remarketing is properly configured, there’s a quick way to find out: let’s review it together.

If you’d like to go through how remarketing is set up in your account, the first session is a diagnostic call, at no cost.

Frequently Asked Questions about Google Ads Remarketing

How much traffic do I need to run Google Ads remarketing?

For Display remarketing, the technical minimum is 100 users in the list. For RLSA (search), the minimum is 1,000 active users in the last 30 days. In practice, you need at least 500–1,000 monthly visits for Display remarketing to have real impact and for lists not to be fragmented into segments that are too small.

How much does Google Ads remarketing cost?

The average retargeting CPC on Google is $0.66–$1.23, versus $1–$2 for prospecting search (DemandSage / WebFX, 2025). The actual cost depends on your sector and competition, but remarketing generally has a lower CPC than cold traffic because the audience is smaller and more qualified. Cost per conversion is usually significantly lower.

Does remarketing work with third-party cookies in 2026?

Yes. Google reversed the deprecation of third-party cookies in Chrome in July 2024 (CookieYes, 2024). Cookies remain active by default. That said, Consent Mode v2 remains mandatory in the EU: users who decline cookies cannot be included in remarketing lists without their consent. The technical foundation works, but the legal foundation requires proper configuration.

What’s the difference between remarketing and retargeting?

In practice, they’re synonymous. “Retargeting” is the generic industry term for reaching users who have already visited your site. “Remarketing” is the term Google uses for its own system. Meta calls it “retargeting” in its interface. The difference is terminological, not functional.

Should I use Performance Max or standard remarketing campaigns for ecommerce?

It depends on volume. If you have fewer than 500 cart abandoners per month, let PMax handle remarketing as part of its overall optimisation. If you exceed that threshold, keep standard dynamic remarketing campaigns running in parallel to control the message, frequency, and bid adjustments by segment. 90% of PMax median cost in retail comes from feed-based ads (Smarter Ecommerce, 2025), confirming that PMax is powerful for shopping but opaque in managing warm audiences.

Sources

Baymard Institute. Cart Abandonment Rate Statistics. https://baymard.com/lists/cart-abandonment-rate. September 2025.
DemandSage / ReadyCloud. Retargeting Statistics 2025. 2025.
DemandSage / Meazy. Display Retargeting CTR Benchmarks. 2025.
DemandSage / Oracle. Remarketing Conversion Rate Data. 2025.
DemandSage / WebFX. Retargeting CPC Benchmarks, Google Ads. 2025.
WordStream. Google Ads Benchmarks 2025 (16,000 campaigns). 2025.
CookieYes. Google Cookie Deprecation U-Turn: What’s Next for Marketers?. https://www.cookieyes.com/blog/google-cookie-deprecation/. July 2024.
Deloitte / Workshop Digital. Cookieless Readiness Survey. March 2025.
Smarter Ecommerce. Performance Max Retail Benchmarks (4,000+ campaigns). April 2025.