Intent retargeting: following what they searched

A person searches "best accounting software for freelancers." They click your site, look around, and leave. Tomorrow they search "how to track business expenses." They are the same person. They have the same problem. But now they see your competitor's ad instead.

This is the difference between behavior retargeting and intent retargeting. Behavior retargeting shows ads based on what someone did (visited your site). Intent retargeting shows ads based on what someone is thinking right now (their search query).

The pattern: Someone who searched for your solution once will probably search for related solutions again. If you can show up in those moments, you catch them when they are actively looking — not when they are watching videos or scrolling social media.

In Google Ads, this is called RLSA — Remarketing Lists for Search Ads. Instead of showing display banners on random websites, you bid higher when a previous visitor searches again.

The basic setup:

1. Build a list of people who visited your site (using a tracking pixel).
2. When someone on that list searches relevant keywords, your ad shows.
3. You can bid higher for these searches because these people already know you.

Example:

A software company sells project management tools. Someone visited their pricing page but did not buy.

• Next week, that person searches "project management for remote teams."
• Because they are on the remarketing list, the software company bids 50% higher.
• The ad shows at the top. The person clicks and finally buys.

Without RLSA, that person might have clicked a competitor. With RLSA, the software company won because they recognized returning intent.

People who already visited your site are more valuable than strangers. They know what you offer. They just need another nudge.

Higher bids for warmer audiences:

• First-time searcher: Normal bid
• Visited homepage: Bid +20%
• Visited pricing page: Bid +50%
• Added to cart but did not buy: Bid +100%

The logic: Someone who looked at pricing is much closer to buying than someone who just read a blog post. Adjust bids accordingly.

Broad keywords become safe:

Normally, broad keywords are risky because they attract random people. But with RLSA, you can bid on broad keywords and only show to previous visitors.

Example:

• Keyword: "software" (too broad normally)
• With RLSA: Only show this ad to people who already visited the project management site
• Result: You catch returning visitors searching vague terms, without wasting money on strangers

When someone is already familiar with your brand, they might search differently. They might use generic terms because they forgot your exact name.

Search patterns of returning visitors:

• First search: "invoice software for consultants" (specific)
• Return search: "invoice software" or "billing tool" (broader)
• Later search: Just your brand name or a competitor's name

How to capture these:

Create separate campaigns for:

1. Non-brand keywords + RLSA (catch them when they search generically)
2. Competitor keywords + RLSA (catch them when they compare)
3. Brand keywords (catch them when they remember you)

Without RLSA, bidding on competitor names is expensive and often fails. With RLSA, you only target people who already visited you — the message becomes "Remember us? Here is why we are better."

Not all returning visitors are equal. Someone who visited yesterday is different from someone who visited three months ago.

Recent visitors (1-7 days):

• Still actively shopping
• Aggressive bids make sense
• Message: "Still deciding? Here is what customers say."

Medium-term visitors (8-30 days):

• May have forgotten or found another solution
• Moderate bids
• Message: "Come back — we added new features."

Long-term visitors (31-90 days):

• Probably moved on
• Lower bids
• Message: "Things have changed. Give us another look."

Why this matters:

Bidding the same for all returning visitors wastes money. Someone who visited yesterday is ten times more likely to buy than someone who visited two months ago. Split your lists by recency.

This approach works best when people search multiple times before buying. It does not work for every business.

Low search volume categories

If your product is so niche that people only search once, RLSA will not help. There are no return searches to capture.

Impulse purchases

People buying cheap or simple things do not research. They decide in one session. RLSA works better for considered purchases like software, services, or expensive items.

Small website traffic

RLSA requires enough visitors to build meaningful lists. If your site gets 100 visitors a month, your remarketing lists will be too small for Google to use effectively.

Privacy and tracking limits

Browsers increasingly block tracking. Some visitors will not be added to your lists at all. Plan for shrinking audience sizes over time.