Remarketing with Google Ads: Practical Guide 2026

70% of online carts are abandoned before completing the purchase (Baymard Institute, 2025). That’s not a bug — it’s consumer behaviour. People browse, compare, get distracted, and move on. The question isn’t whether they’ll leave your store without buying. It’s whether you’ll be there when they come back to decide.

Remarketing in Google Ads is the system that allows you to show ads to people who have already interacted with your site or app. Done right, it means appearing at the right moment — when purchase intent reactivates — with the right message for that specific user. Not the same banner a hundred times.

Key takeaways

• 70% of online carts are abandoned before purchase (Baymard Institute, 2025)
• Remarketing audiences convert at 3–5x lower CPA than cold traffic (Google, internal data)
• Dynamic remarketing achieves CTR 2–3x higher than standard display ads (Google, 2024)
• RLSA (Remarketing for Search) allows adjusting bids in search campaigns based on site history
• Performance Max has integrated remarketing audiences natively since 2024

---

What is remarketing and how does it work in Google Ads?

Remarketing works via a code (Google Tag or Google Analytics 4) that registers visitor behaviour on your site: pages visited, products seen, cart additions, abandoned checkout. That data is used to create audience lists in Google Ads, which you then use to show specific ads.

When a user visits your product page and leaves without buying, they’re added to the “product page visitors” list. For the next 7, 14 or 30 days (you define the duration), that user sees your ads on Gmail, Display, YouTube, or even in Google search results if you use RLSA.

There are five main types of remarketing in Google Ads:

• Standard remarketing: shows ads to previous visitors across the Display Network and YouTube. The most basic format.
• Dynamic remarketing: shows the exact products or services the user saw on your site. Requires a connected product feed. It’s the most effective format for ecommerce.
• RLSA (Remarketing Lists for Search Ads): adjusts bids on search keywords for users with specific browsing history on your site.
• Customer Match: uploads a customer email list to Google Ads. Allows targeting existing customers or lookalike audiences with similar profiles.
• Video remarketing: targets users who interacted with your YouTube channel or watched your videos. Ideal for brand reinforcement and high-ticket abandoned carts.

---

Why remarketing converts better than cold traffic

Someone who already visited your product page has done something that cold traffic hasn’t: evaluated your site, seen your product, and decided it’s at least interesting enough to spend time looking at it. That’s a much higher baseline than someone who sees your ad for the first time.

In accounts where I’ve compared acquisition campaigns against remarketing campaigns with similar budgets, the CPA difference is usually 3–4x. This isn’t a figure I pulled from a case study — it’s what I consistently see month after month across ecommerce clients. The reason is straightforward: the purchase decision is already 40–60% made in a user who visited and left without buying. You’re closing the deal, not opening it.

Dynamic remarketing pushes this advantage further. When you show a user the exact product they were looking at — same image, same price, same product name — you’re eliminating cognitive friction. They don’t have to find the product again or remember where they saw it. The conversion path is: click ad → land on product page → buy. Google reports that dynamic remarketing achieves CTR 2–3x higher than standard display ads (Google, 2024).

Remarketing audiences consistently deliver 3–5x lower CPA than cold traffic in ecommerce accounts (Google internal data, 2024). The reason isn’t algorithmic magic: users who already visited your site have a much higher baseline purchase intent than someone seeing your brand for the first time. Dynamic remarketing amplifies this effect by showing the exact product the user viewed, eliminating the search friction of cold acquisition.

---

How to set up dynamic remarketing for ecommerce step by step

Dynamic remarketing requires three elements before you launch: an event tag correctly configured on your site, a connected product feed, and audience lists with sufficient volume. Get any one of these wrong and you end up with generic banner ads instead of product-specific ones — which defeats the purpose entirely.

Start with the tag. If you use Google Tag Manager, add the Google Ads remarketing tag with these custom parameters: ecomm_prodid (product ID), ecomm_pagetype (page type: home, category, product, cart, purchase), and ecomm_totalvalue (value of viewed products). These parameters let Google match what users viewed with your product feed. Without them, the feed connection is useless.

Next, link your product feed. In Google Merchant Center, connect the feed to Google Ads. Each product needs a unique ID (id) that exactly matches the ecomm_prodid parameter firing on your site. A mismatch here is the most common reason dynamic remarketing shows generic ads instead of specific ones — worth double-checking before launching.

Then build your audience lists. In Google Ads → Tools → Audience Manager, create these as a minimum:

Audience	Duration	Priority
Cart abandoners	3–7 days	Very high
Product page visitors (no cart)	7–14 days	High
Category visitors (no product page)	14–30 days	Medium
Previous buyers	90–180 days	Cross-sell/upsell
Homepage only visitors	30 days	Low

With the lists in place, create a Display campaign and target “Remarketing”. Enable Responsive display ads and let Google generate dynamic versions from your feed — this is where the product-specific creative comes from. Then set your bids to reflect actual behaviour: cart abandoners convert 5–10x better than generic product visitors. If your standard CPA is €20, you can afford tCPA €10–12 for cart abandoners and €25–30 for category visitors. The funnel matters here. Bid accordingly.

---

RLSA: how to use search remarketing to increase ROAS

RLSA (Remarketing Lists for Search Ads) is underused relative to how much it can move the needle: it lets you adjust bids in your search campaigns based on whether the user has visited your site before.

Here’s the thing — two people can type the exact same query and represent completely different levels of purchase intent. A user searching “running shoes” who abandoned a cart on your site last week is not the same prospect as someone who’s never heard of your brand. RLSA lets you recognise that difference at bid time and act on it.

The simplest way to start is bid adjustment on existing campaigns. Add your remarketing lists as “Bid only” — meaning you keep the same keywords and targeting, but increase bids by 15–30% when the search comes from someone who already visited your site. You’re not changing what you target, just how aggressively you bid on people you already know.

Once that’s working, you can go further. Create search campaigns with broader keywords — terms you’d normally consider too generic — but restrict them to remarketing audiences only. A user searching “shoes” who browsed your running section yesterday is a different prospect than someone seeing that keyword cold. Broad match plus RLSA can unlock traffic you’d otherwise write off as unprofitable.

The cross-sell angle is the one most accounts miss. Add your “previous buyers” list and run campaigns for complementary categories. Someone who bought running shoes might search for compression socks or a running watch three weeks later. With RLSA, you bid specifically on that search at that moment — which is about as targeted as search advertising gets.

The RLSA configuration I recommend most in ecommerce accounts with €5,000+/month budget: a duplicate of the main search campaign, limited to the “cart abandoners last 7 days” audience, with a 40–50% bid increase and the same keywords. The CPA on that campaign is usually 2–3x better than the general campaign, and it uses a small fraction of the total budget.

---

Remarketing in Performance Max in 2026

Since 2024, Performance Max has integrated remarketing audiences natively. You don’t need to create separate remarketing campaigns if you’re using PMax: you configure audience signals (including remarketing lists) directly in the campaign, and the algorithm uses that information to optimise across all available inventory.

How to configure remarketing audiences in PMax:

In your PMax campaign, in “Audience Signals”, add:

• Your custom remarketing lists (cart abandoners, product visitors, buyers)
• Customer Match lists (customer emails)
• Similar audiences to your buyers (Google generates them automatically)

Worth noting: in PMax, audience signals are suggestions, not filters. Google uses them as starting points but can show ads beyond those segments if it identifies conversion opportunities. This is the fundamental difference with standard remarketing campaigns.

When does it make sense to have a separate remarketing campaign alongside PMax?

When you need precise control over the message, bid, or frequency for specific segments. For example:

• Win-back campaign for buyers who haven’t purchased in more than 90 days, with a specific discount offer
• High-frequency campaign for cart abandoners in the last 48 hours with a specific promotional message

In these cases, a specific Display or Search remarketing campaign gives you control that PMax doesn’t provide.

For the overall view of how Performance Max fits into a complete ecommerce strategy, the complete Google Ads for ecommerce guide covers the structure and prioritisation of campaign types.

---

Remarketing mistakes that waste budget

Most remarketing problems trace back to the same root cause: treating all past visitors as if they’re equally valuable. They’re not.

Showing ads to everyone who visited the homepage is the most common version of this mistake. Homepage visitors are exploratory. They haven’t shown intent for any specific product. Lumping them into the same campaign as cart abandoners dilutes your performance data and inflates your CPA. Segment by page type from the start — homepage visits, product views, and cart additions should each be their own audience.

Frequency is the second thing that goes wrong. A user who sees your ad 15 times in three days doesn’t convert better — they develop banner blindness, or worse, start associating your brand with annoyance. Three to five impressions per day per user is enough for Display. More than that and you’re paying to irritate people.

Membership duration is where accounts quietly bleed money. A 30-day cart abandoner has very different purchase intent from a 3-day one. Using a single duration for all audiences ignores this completely. The tighter the window, the hotter the audience: 3–7 days for cart abandoners, 14–30 days for product visitors, 90–180 days for buyers looking at cross-sell.

Two more worth checking: first, add your “recent buyers” list as an exclusion in all non-remarketing campaigns — otherwise you’re paying acquisition CPA for someone who already converted. Second, if you ran a promotion, segment by promotional period. Showing a standard-price ad to someone who added a product at 20% off creates friction you don’t need.

---

How to measure remarketing performance

Remarketing campaigns need different metrics from acquisition campaigns. Comparing CPA directly between the two is misleading — remarketing has an asymmetric advantage because audiences are already warm. What should you actually look at?

View-through conversion rate matters more here than in acquisition, because Display clicks are rare. Users see the ad, go back to your site later via a direct or organic visit, and convert. GA4’s attribution reports will show you how much of that is being influenced by prior remarketing impressions.

ROAS by audience segment is the clearest performance signal. Cart abandoners should have 2–3x better ROAS than category visitors. If they don’t, the issue is usually the ad message or the landing page for that segment, not the audience itself.

Finally, watch frequency against conversion rate. If conversion rate drops as you push more impressions per user, that’s the algorithm telling you to back off. More impressions past a certain point means more money spent on people who’ve already decided not to buy.

For understanding how remarketing fits into the complete conversion measurement system, the Google Ads conversion tracking guide explains how to configure all conversion types in GA4 and Google Ads.

---

Frequently asked questions about remarketing in Google Ads

What’s the difference between remarketing and retargeting?

They’re practically synonymous in the current marketing context. Retargeting is the general practice of targeting previous visitors. Remarketing is Google’s specific term for that same practice in Google Ads. Outside Google Ads (in Meta, LinkedIn, or TikTok), it’s usually called retargeting.

How many users do I need on a list to launch a remarketing campaign?

Google Ads requires a minimum of 100 users for Display and Search (RLSA) campaigns, and 1,000 users for YouTube campaigns. For Customer Match lists (email), 1,000 matched users are required. If your traffic is low, focus on extending membership durations (up to 540 days for Display) to build sufficient lists faster.

Does remarketing affect privacy or GDPR compliance?

Yes. You need explicit consent from users to use their data for remarketing. In practice, this means a cookie consent banner that allows users to accept or reject tracking cookies. Google Tag Manager must be configured to fire the remarketing tag only after consent is given. If your site doesn’t have a compliant consent system, remarketing could violate GDPR.

Can I do remarketing without Google Analytics?

Yes, using the Google Ads tag (global site tag or Tag Manager) instead of GA4. The Google Ads tag allows creating audience lists based on page visits and conversion events. However, integrating GA4 gives you access to richer audiences (users who completed specific events, users with more than X sessions) that the Google Ads tag alone doesn’t provide.

How long should remarketing campaigns run before evaluating performance?

At least 30 days for Display campaigns and 14 days for RLSA. Remarketing relies on accumulating users in lists and on Google learning who converts within those lists. During the first two weeks, data may be insufficient for meaningful decisions. Evaluate trends, not individual days, and give the algorithm time to optimise.

---

Sources

1. Baymard Institute. 49 Cart Abandonment Rate Statistics 2025. https://baymard.com/lists/cart-abandonment-rate. 2025.
2. Google. Dynamic remarketing: Reach past visitors with personalised ads. https://support.google.com/google-ads/answer/3124536. 2024.
3. Google. About remarketing lists for search ads. https://support.google.com/google-ads/answer/2701222. 2024.
4. Google. Performance Max for ecommerce. https://support.google.com/google-ads/answer/11189316. 2025.