Do Meta and TikTok ads increase Google searches? How to measure the branded search halo

Do Facebook and TikTok ads actually increase Google searches?

In many accounts, yes. Someone scrolls past your ad on Instagram or TikTok. They do not click. Two days later they remember your name, type it into Google and book a call. In your reports, that lead looks like it came from Google. The social ad that started it gets zero credit.

This effect is usually called the search halo or search lift. It matters because it changes how you judge your ad spend. If you cut Meta because its reported cost per lead looks high, your Google numbers may drop a few weeks later, and you will not know why.

In the same Brainlabs analysis, 71% of the incremental search visits came from organic results and 29% from paid search. So the halo shows up in your free Google traffic, not only in your Google Ads account. Of the studies that split brand and non-brand terms, about 31% of the extra search traffic came from brand queries. The rest came from generic searches, like the product category.

These numbers come from the platforms and their agency partners, so treat them as evidence that the effect exists, not as a benchmark you should expect. Results vary a lot by brand, offer and budget. An older Facebook study of 23 US campaigns found that only about a quarter showed a statistically significant lift in search referral traffic. The halo is real in some accounts and small or absent in others. That is exactly why you test.

Why GA4 and Google Ads give social ads no credit

Most reports assign the sale to the last click before conversion. When someone searches your brand and clicks a Google result, that search click wins. The ad they saw on TikTok two days earlier had no click, so it does not exist in the report.

Meta and TikTok try to fill this gap with view-through attribution, which counts conversions from people who saw an ad but did not click. The problem is the opposite: view-through windows can over-credit social ads for people who would have bought anyway. Neither report answers the real question, which is what would have happened without the ads.

How do you measure the search lift from social ads?

There are three levels, from cheapest to most reliable. Most small businesses should start with the first, then move to a geo holdout once they spend steadily on social.

Level 1: watch branded search around campaign changes

Track branded search impressions in Search Console and branded clicks in Google Ads every week. Mark the dates you launch, pause or scale social campaigns. If branded search climbs two to four weeks after a new TikTok push and falls after you pause it, that is a useful signal. It is not proof, because a holiday, a news story or a competitor problem can cause the same pattern.

Level 2: run a platform lift study

Meta offers lift studies that split an audience into an exposed group and a holdout group that never sees your ads. Some versions measure search behavior. TikTok runs similar Search Lift studies. Access is usually limited and often handled through a platform rep, and you need enough spend for the test to reach significance.

Level 3: run your own geo holdout

You pause social ads in a set of regions, keep them running everywhere else, and compare what happens to branded search and leads. You control the test, it works with any platform, and it measures effects across every channel at once. The next section explains how to do it.

How to run a geo holdout test for branded search

1. 1Pick one question. For example: does our Meta spend raise branded Google searches and booked calls?
2. 2Choose comparable regions. Use US states or metro areas (DMAs) with similar past trends in branded search and leads. Avoid using only one test region.
3. 3Split the regions into test and control groups. In the holdout group, pause or exclude the social campaigns. Leave everything else, including Google Ads, unchanged.
4. 4Pull a baseline. Export at least several months of weekly branded impressions from Search Console and organic sessions by region from GA4, so you know how the regions normally move together.
5. 5Run the test for at least four weeks, and longer if you have long buying cycles. Do not change budgets, offers or creative during the test.
6. 6Compare the gap. Measure how branded searches, organic sessions and leads in the holdout regions changed against the control regions, relative to the baseline.
7. 7Turn the gap into money. Divide the social spend you held back by the incremental leads it was driving to get a true cost per incremental lead.

If you want more statistical rigor, Meta publishes an open-source tool called GeoLift for designing and reading geo tests. It uses a method called synthetic control, which builds a weighted mix of control regions that tracks your test regions before the test starts.

Which metrics should you track in GA4 and Search Console?

Pick metrics that the social platform cannot influence through its own reporting. Search Console and GA4 are good choices because they measure what people did on Google and your site, regardless of which ad they saw.

Google added a branded queries filter to the Search Console Performance report in late 2025. It separates searches that include your brand name, common misspellings and related product names from the rest, using Google’s own classification. It only works on top-level properties with enough search volume, and it does not backfill old data, so it is worth checking now whether your property has it.

What if the test shows no search lift?

A flat result is still useful. It can mean three things. First, your social ads may not be memorable enough to make people search. Weak hooks and a missing brand name in the first seconds are common causes. Second, your spend may be too small to move search volume you can measure. Third, your social ads may drive direct conversions well and the halo simply is not large for your business.

Check the creative before you cut the budget. Ads that show the brand name clearly, on screen and in voice, give people something to search. Ads that look like generic stock content do not. Then retest after a creative refresh.

How to use the results to set your budget

• Add the incremental search leads to the social channel’s results when you calculate its real cost per lead.
• Lower your expectations for branded Google Ads. If much of your brand demand comes from social, branded search is capturing demand, not creating it.
• Do not cut social spend based only on in-platform cost per lead. Check what happens to branded search first.
• Repeat the test once or twice a year, or after any big change in spend, offer or creative strategy.

The goal is not a perfect attribution model. It is a reasonable, tested estimate of what each channel adds, so you can move budget with confidence instead of guessing.

Frequently asked questions