Measuring Branded Search Lift From Influencer Campaigns

What "branded search lift" actually measures

Branded search lift is the increase in searches that include your brand name during and shortly after an influencer campaign — more people googling "[brand name] review", or just the brand name on its own, after seeing a creator's post. It's a proxy signal: it doesn't measure the campaign's sales directly, but a behaviour pattern that typically precedes a purchase — someone gets curious, searches, and may only buy days later, often on a different device than the one the post was seen on. The method needs no tracking links, discount codes or pixel setup, only access to two free Google tools.

It's already mentioned briefly as one of several "Layer 2" signals in how to measure influencer marketing when you can't track everything — this article goes deep on how to actually do it, in which tool, and exactly where the limit of what the numbers can prove sits.

Two free tools, two different numbers

Google Search Console and Google Trends don't answer the same question, even though both can be used to spot a branded search lift. The difference matters before you compare numbers from them:

How to use Search Console's branded queries filter

Google launched a dedicated branded queries filter in Search Console's Performance report on 20 November 2025, per Google's own Search Central blog. The filter automatically splits every search query into two groups: "Branded" (searches that include your brand name, common misspellings of it, or brand-specific products — Google's own example is "Google", "Gogle" and "Gmail") and "Non-branded" (everything else). Find it under the filter bar in the Performance report: select the Query filter, then choose Branded or Non-branded.

Three things are worth knowing before using it:

• The filter covers organic search only — paid search ads don't appear in Search Console at all.
• Data for the filter itself only exists from 11 March 2025 onward, per Google's own help page — so the branded filter can't be used to compare against a period further back than that, even though the Performance report generally shows up to 16 months of history.
• The filter isn't available for every site — it requires a site with sufficient query and impression volume, and it only works on top-level properties (not a subdomain or path-based property, per Google's own documentation).

If the filter isn't available yet, you can get close to the same result manually: in the ordinary Query filter, search for queries that "contain" your brand name to pull the same click and impression numbers — just without the automatic misspelling and product recognition the dedicated filter adds.

How to use Google Trends' comparison feature

Google Trends doesn't show an absolute number of searches — it shows a relative interest score from 0 to 100. Per Google's own explanation, Trends normalizes the data by dividing each data point by the total search volume for the geography and time range it represents, then scaling the result so the highest point in the selected range gets the value 100 — everything else is shown as a share of that. A value of 50 means "half of the period's peak interest", not a fixed number of searches.

That has a direct consequence for how you use the tool to measure a campaign: pick a range where the campaign week can realistically become the period's peak — the last 90 days, for example — not a multi-year range where a completely different, older event already occupies the 100 value and artificially compresses the campaign week's number.

Google is explicit about the method's limitations: the data is based on a sample of Google's searches (not all of them), with statistical noise added to protect user privacy — something most noticeable on low-volume search terms — and Google states directly that "Google Trends is not a scientific poll" and should be treated as one data point among several, not a conclusion on its own.

Search Console vs. Google Trends: which one when

The real limitation: correlation, not proof

Neither Search Console nor Google Trends can prove the influencer campaign caused the increase. Some of the people who searched the brand name during the campaign week would likely have done so anyway — for other reasons, or because they were already on their way to becoming customers. A branded search lift that coincides in time with the campaign is a strong signal, but it's observational: there's no control group showing what would have happened without the campaign. That's the exact same limitation as share of voice and sentiment — another free, observational signal with no control group. Proving an actual causal effect requires either a real brand lift study with an exposed and a control group, or an incrementality test with a holdout group.

Common mistakes

• Reading Google Trends' 0-100 scale as an absolute search count. A jump from 25 to 100 isn't "4x more searches" — it's a jump to the period's highest point, which can reflect a small absolute number of searches if the category already has low volume.
• Choosing too long a comparison range in Trends. An older, unrelated spike can occupy the 100 value and make the campaign week's real increase hard to see visually.
• Ignoring other activity happening at the same time. A branded search lift that coincides with a press mention, another campaign or a seasonal event can't automatically be attributed to the influencer alone.
• Using Search Console alone and forgetting it only covers organic search. Paid brand keywords (Google Ads) aren't counted, even though they're also affected by a branded search lift.

Decision framework: is branded search lift the right signal here?

IF you have little or no measurement budget and just want a fast, directional signal → branded search lift via Search Console and Trends is the cheapest place to start.

IF you need to document a concrete, defensible number for a report → use Search Console's absolute click and impression numbers, not Trends' relative score.

IF you need to prove the campaign actually changed people's perception of the brand (not just their search behaviour) → a brand lift study is the right tool.

IF you have enough volume for a holdout test and need to prove an actual sales effect → an incrementality test beats both free signals here on documenting causation.

Worked example: reading a branded search lift

The numbers below are hypothetical and illustrate only the calculation principle — this is not a real Make Influence customer case.

A fashion brand runs a two-week influencer campaign. The week before the campaign, Search Console shows a weekly average of 340 branded clicks and 2,100 branded impressions. During the campaign week, branded clicks rise to 612 (an increase of (612−340)/340 = 80%), while impressions rise to 3,150 (an increase of (3,150−2,100)/2,100 = 50%).

In Google Trends, over a 90-day range, the brand name's interest score averages around 25 in the eight weeks before the campaign. During the campaign week, that same week becomes the range's highest point and is therefore scored 100. That is not the same as saying search interest quadrupled — it means no other week in the selected 90 days had more searches. Search Console's 80% rise in actual clicks is the number that can genuinely be defended as a concrete figure; the Trends number is best used to confirm direction and timing, not magnitude.

Make Influence's operational perspective

In our experience, branded search lift is the signal brands skip most often — because it requires actually going and looking, rather than appearing automatically in a report. We recommend checking Search Console's branded filter (or the manual "contains brand name" query) one week before and one week during any larger campaign, as a standing, low-cost check — not a replacement for real tracking, but a fast health check that often shows an effect long before the first tracked sales come in.

FAQ

Can I use branded search lift as the only proof a campaign worked?

No. It's an observational signal with no control group — use it as a fast, free directional signal, and pair it with a brand lift study or an incrementality test if you need to document an actual causal effect.

Does Search Console's branded queries filter work for every site?

No. Per Google, it requires a site with sufficient query and impression volume, and it's only available for top-level properties, not subdomains or path-based properties.

How far back can I see branded data in Search Console?

The general Performance dataset typically goes back up to 16 months, but the branded queries filter itself only shows data from 11 March 2025 onward, per Google's own help page.

Can I measure branded search lift if the campaign ran organically with no paid spend at all?

Yes — the method needs no ad account or paid budget at all, only access to your own Search Console and the publicly available Google Trends.

What if the brand name is a common word used in many other contexts?

Then the signal gets harder to read cleanly, because searches mix with unrelated volume. Try combining the brand name with a product name or a campaign-specific search phrase in Trends, and in Search Console look at queries that contain both the brand name and one of your product names.