Attributed Branded Searches is a metric in Google Ads that counts how many people saw your video ad and later searched for your brand on Google, without clicking the ad. It covers Video, Performance Max and Demand Gen campaigns, it's free and always on, and it's the first everyday measure of whether your top of funnel is creating demand. It's also easy to oversell, so here's what it tells you and what it doesn't.
If you've ever tried to justify a YouTube budget to a client (or to yourself), you know how the conversation goes. You show views, reach, maybe a few clicks. And then comes the question:
"But... does any of this actually sell anything?"
And that's where you got stuck. The honest answer has always been some version of "yes, but not in a way I can prove in this report". Now Google has released something that actually starts to fix that, and it's as useful as its name is boring.
What does Attributed Branded Searches actually measure?
Put simply: how many people saw your video ad and then searched for your brand on Google without having clicked the ad.
Read that last part again, because it's the whole point. Nobody clicks a YouTube ad. People sit there with a coffee, watch it to the end, think "oh, them" and move on. Later (that evening, or a week later) your name pops back into their head and they search for you. That's what a real customer journey looks like: rarely a straight line from ad to purchase, more often an idea that's left to grow for a while.
Until now, that behaviour has been more or less invisible in your reporting. Now you get a number for it.
How does it work in practice?
A few details are worth knowing, so you don't end up selling it as something it isn't:
- It's based on views, not clicks. If a user sees the ad and later searches for the brand, it counts. The credit goes to the last viewed ad before the search.
- The attribution window is adjustable from 1 to 30 days, with 7 days as the default. You'll see people write "30 days" as if that were the rule. It's the maximum, not the default. Worth knowing before someone corrects you.
- It's for reporting only. You can't bid towards it, and it doesn't show up in the regular Conversions column. Think of it as a gauge on the dashboard, not the steering wheel.
- It works for Video, Performance Max and Demand Gen campaigns on YouTube, Discover and Gmail.
- You have to set up brand mapping first, which means telling Google which search terms and spellings count as "your brand". Get that wrong and the measurement is skewed.
Why is this a bigger deal than it sounds?
Until now there were really only two ways to measure the top of the funnel: expensive Brand Lift studies or pure gut feeling. Brand Lift studies are survey-based, usually need a sizeable budget and a fixed campaign period, and sit in their own little silo, separate from your normal reporting.
Attributed Branded Searches, on the other hand, is always on, free and built right into Google Ads. That lets you keep a constant eye on whether your video work is actually creating demand. Not "awareness" as a vague feel-good word, but people who go and search for you afterwards.
Google itself puts a figure on it: each additional branded search is worth, on average, about $31 in extra sales. Take that with a pinch of salt (it's Google doing the maths), but the direction is right. Someone who searches for your name is further along in their decision than someone who's never heard of you.
What's the catch?
Here's the part most people selling this skip, and the part that makes you more credible when you bring it up yourself.
It shows correlation, not causation. If you also run TV, radio, outdoor or other digital channels at the same time, you can't say for certain that it was the YouTube ad that triggered the search. Google credits the last viewed ad, but that person may have seen your advertising in four other places first.
So treat it as a strong signal and a trend over time, not the final answer. It's still vastly better than what we had before, which was more or less nothing.
One obvious but important point: this only works if your ad is worth remembering. Nobody goes looking for a brand from an ad they've already forgotten. By all means measure, but make something worth measuring. If the video is the weak link, that's a job for ad production before it's a job for measurement.
What to do next
Want to try this in your own account? Here's the Adcoacher way (you do it yourself, no agency needed):
- Set up brand mapping in Google Ads. Add your brand with every reasonable spelling and variation.
- Find the metric. Availability has varied between accounts during the rollout. If you can't see it, ask your Google contact or Google Ads support.
- Collect a 30-day baseline before you draw any conclusions. You want a trend, not a single day.
- See which campaign drives the most branded searches. Does your Performance Max campaign drive more than your YouTube-only campaign? Then you know where awareness is really being built, and you can shift budget accordingly. (Read up on when Performance Max works before you do.)
- Make sure the rest of your measurement holds up. If you're not sure it does, that's what my analytics and tracking work covers. Or start with the quick analysis: five questions, two minutes.
It isn't perfect. But for the first time in a long while there's a clean behavioural signal that connects "someone saw us" with "someone got curious". And that makes for a far more enjoyable conversation with a client than pointing at a reach figure and shrugging.
