Retargeting Ads: How to Bring Back Visitors Who Didn’t Convert

Why Retargeting Works

Someone visited your website. They looked at your product page, your pricing, maybe even started filling out a form. Then they left. No purchase, no enquiry, no conversion.

This happens to 96-98% of your traffic. For every 100 visitors, 2-4 convert. The rest disappear.

But they didn’t disappear because they weren’t interested. They got distracted. They wanted to think about it. They were comparing options. They were on their phone during their commute and meant to come back later. Life got in the way.

Retargeting puts your brand back in front of these people. When they scroll through Instagram, browse a news site, or watch YouTube, your ad is there reminding them to come back and finish what they started.

It works because these people already know who you are. They’ve already shown intent by visiting your site. You’re not interrupting a stranger — you’re following up with someone who raised their hand. That’s why retargeting typically delivers the lowest cost per acquisition of any campaign type.

The Lazy Approach (And Why It Fails)

Most retargeting looks like this: create one audience of everyone who visited the website in the last 90 days, show them one ad, and run it until you run out of patience or budget.

This is better than no retargeting at all, but only marginally. Here’s what’s wrong with it:

• It treats all visitors the same. Someone who spent 30 seconds on your homepage and bounced is not the same as someone who viewed three product pages and added something to their cart. They need different messages at different frequencies.
• The audience window is too long. After 30 days, most people have either forgotten about you or bought from a competitor. Showing them ads for another 60 days is money down the drain.
• One ad creates fatigue. After seeing the same ad four or five times, people stop noticing it. After ten times, they start getting annoyed. After twenty times, they actively resent your brand.
• No exclusions. If you don’t exclude people who’ve already converted, you’re spending money advertising to your own customers. Unless you’re deliberately running a retention campaign, this is waste.

Good retargeting requires more thought. Segment your audiences, vary your creative, control your frequency, and set proper exclusions.

Segment by Intent

High intent

These are people who took actions that signal they were close to converting:

• Added to cart but didn’t purchase
• Started a form but didn’t submit
• Viewed pricing or contact pages
• Spent significant time on product/service pages

This segment gets the most aggressive retargeting. Show them ads within the first 7 days. Address their likely objections — price concerns, trust issues, urgency. For e-commerce, dynamic product ads showing exactly what they left in their cart are extremely effective. For services, testimonials and case studies work well.

This audience is small but high value. It’s where most of your retargeting conversions come from.

Medium intent

These visitors showed genuine interest but didn’t take a high-intent action:

• Viewed multiple pages
• Spent more than 60 seconds on site
• Viewed specific product or service pages
• Engaged with content (blog posts, resources)

Retarget these people with educational content, social proof, and benefit-focused messaging. They’re still in the consideration phase — your job is to move them from ‘interested’ to ‘ready to act’. Show them customer stories, explain your process, answer common questions.

Low intent

Single page visitors who bounced quickly. They might have clicked an ad by accident or decided within seconds that your site wasn’t what they were looking for.

The conventional wisdom is to retarget everyone. The better approach: don’t retarget low-intent visitors at all, or retarget them only lightly with brand awareness content. The cost per conversion from this segment is high and the volume is low. Your budget is better spent on the medium and high-intent segments.

If you do retarget low-intent visitors, keep the frequency very low (1-2 impressions per week) and use broad brand messaging rather than hard conversion asks.

Retargeting on Meta Ads

Dynamic product ads

For e-commerce, dynamic product ads (DPAs) are the single most effective retargeting format. They automatically show people the exact products they viewed on your site, complete with the image, price, and a link back to the product page.

DPAs require a product catalogue connected to your Meta pixel. Once set up, they run largely on autopilot — Meta pulls the right product for each person based on their browsing history. The personalisation drives significantly higher click-through and conversion rates than generic ads.

The key is to make sure your product catalogue is accurate. Wrong prices, out-of-stock items, or poor images in your catalogue mean poor DPA performance.

Video retargeting

Video works brilliantly for retargeting because it lets you tell a story that static images can’t. Use video retargeting to:

• Show customer testimonials to address trust objections
• Demonstrate how your product or service works
• Walk through your process for service businesses
• Address the most common reasons people don’t buy

Keep retargeting videos short — 15-30 seconds. These people already know who you are. You don’t need to introduce yourself. Get straight to the point.

Sequenced messaging

Instead of showing the same ad repeatedly, sequence your messaging over time:

• Days 1-3: Remind them what they were looking at. Simple recall message. ‘Still thinking about it?’ plus the product or service they viewed.
• Days 4-7: Social proof. Customer reviews, case studies, star ratings. Address the ‘can I trust this company?’ question.
• Days 8-14: Overcome objections. FAQ-style content, guarantee information, comparison against alternatives. If appropriate, add a time-limited incentive.

Sequenced messaging requires more creative work upfront but dramatically outperforms the single-ad approach. Each message moves the prospect closer to conversion by addressing a different barrier.

Retargeting on Google Ads

Display remarketing

Google’s Display Network reaches over 90% of internet users across millions of websites. Display remarketing shows banner ads to your website visitors as they browse other sites.

The challenge with Display is ad quality and placement. Your ads might show on low-quality sites, buried at the bottom of pages nobody reads. Use placement exclusions aggressively — exclude mobile apps (which generate accidental clicks), parked domains, and any sites with poor performance.

For creative, responsive display ads are the minimum. If you can invest in custom-designed banner ads at key sizes (300×250, 728×90, 160×600), you’ll stand out more than the auto-generated responsive format.

YouTube remarketing

If you have video content, YouTube remarketing is extremely effective and underused. Show pre-roll or in-stream ads to people who’ve visited your website. The cost per view on YouTube is typically much lower than the cost per click on Search or Display.

You can also create audiences from your YouTube channel — people who watched your videos, subscribed, or engaged. These are warm audiences that work well for both retargeting and as seed audiences for prospecting.

RLSA (Remarketing Lists for Search Ads)

RLSA is one of the most powerful and underused features in Google Ads. It lets you adjust your search campaigns for people who’ve already visited your website.

Two ways to use it:

• Bid adjustment: Bid higher for past visitors. If someone searched for ‘accountant London’ and they’ve already been to your site, you know they’re further down the funnel than a first-time searcher. Increase your bid by 20-50% to make sure you win that click.
• Broader keywords: Create a search campaign that only targets past visitors but uses broader keywords you’d normally avoid. You might not bid on ‘accountant’ for cold traffic (too expensive, too vague), but for someone who’s already visited your site, that broad term is worth bidding on because you know they’re interested.

RLSA combines the intent signal of search with the warmth of retargeting. The CPA is typically the lowest of any search campaign type.

Frequency Management

Frequency is how many times someone sees your ad. Too few impressions and your retargeting has no impact. Too many and you’re wasting money and annoying people.

General frequency guidelines:

• High-intent audiences: Up to 1-2 impressions per day for the first week, then taper to 3-4 per week
• Medium-intent audiences: 3-5 impressions per week
• Low-intent audiences: 1-2 impressions per week maximum

On Meta, you can’t set hard frequency caps at the ad set level, but you can control it through budget management. A smaller audience with a lower budget naturally limits frequency. On Google Display, you can set explicit frequency caps at the campaign or ad group level — use this.

Watch your frequency reports. If average frequency exceeds 10-15 impressions per user per week across any audience, you’re almost certainly seeing diminishing returns. CPAs will start to rise as you exhaust the convertible portion of your audience and just keep showing ads to people who are never going to buy.

Audience Windows

How long should you retarget someone after they visit your site? It depends on your purchase cycle:

• Impulse purchases (e-commerce under £50): 7-14 days. If they haven’t bought within two weeks, they’ve probably bought elsewhere or decided they don’t need it.
• Considered purchases (£50-500): 14-30 days. People comparing options need more time, but beyond 30 days the intent has usually faded.
• High-value purchases (£500+): 30-60 days. Big decisions take longer, and the higher value justifies the longer retargeting window.
• B2B services: 30-90 days. B2B sales cycles are long and often involve multiple decision-makers. Staying visible for longer makes sense, but keep frequency low.

Layer your audiences by recency. A 0-7 day visitor audience is warmer than a 7-14 day audience, which is warmer than a 14-30 day audience. Allocate more budget to the most recent visitors and taper off as the window extends.

Exclusions: The Most Important Part

If you take one thing from this article, let it be this: exclusions matter more than inclusions in retargeting.

At minimum, exclude:

• People who’ve already converted. If someone bought your product or submitted your lead form, stop showing them acquisition ads. Create a conversion audience and exclude it from all retargeting campaigns. You can run separate retention campaigns for existing customers, but that’s a different strategy with different messaging.
• Very short-visit bouncers. Someone who spent under 5 seconds on your site didn’t engage at all. They probably clicked by accident or immediately realised you weren’t what they were looking for. Retargeting them has almost zero chance of converting and just wastes budget.
• People past your audience window. If your window is 30 days, don’t let Google or Meta auto-extend to 60 or 90 days. Set the window and stick to it.
• Employees and existing relationships. Your own team visiting the site, your existing clients browsing your blog — these should be excluded. Upload a list of internal email addresses as a custom audience and exclude it.

Every person you exclude who was never going to convert makes your budget go further for the people who will.

Measuring Retargeting Properly

Retargeting measurement is tricky because attribution models tend to overstate its impact. Someone sees your retargeting ad, goes to Google, searches your brand name, and converts. The retargeting ad gets view-through attribution credit, but the person might have come back anyway.

The best way to measure true retargeting impact is with holdout tests. Take 10-15% of your retargeting audience and exclude them from seeing any ads. Compare the conversion rate of the group that saw ads against the group that didn’t. The difference is the true incremental value of your retargeting.

If your holdout group converts at almost the same rate as your retargeting group, your retargeting isn’t doing as much as your reports suggest. If there’s a significant gap, you’ve validated that the spend is genuinely driving additional conversions.

Beyond holdout tests, keep your retargeting budget in context. The standard allocation is 70-80% of total ad budget to prospecting and 20-30% to retargeting. If your retargeting is consuming more than 30% of budget, you’re probably over-investing in warm audiences and under-investing in filling the top of the funnel. More prospecting means more website visitors, which means a larger retargeting pool, which means more retargeting conversions. The two work together.

Track these metrics for your retargeting campaigns specifically:

• Cost per acquisition (vs your prospecting CPA)
• Frequency per user per week
• Audience size over time (is it growing or shrinking?)
• Incremental conversion lift (from holdout tests)
• Creative fatigue indicators (CTR declining over time)