Just saw that Google added Query Groups in Search Console Insights. It uses AI to group similar search queries so you can see broader intent, like how “best hiking shoes for men” and “top trail running shoes” might fall into one cluster.
I like the idea. I already group keywords by landing page when planning content for my ecommerce brand, so this could be a nice shortcut. But it looks like there’s still no export option, which makes it hard to use beyond a quick glance.
Has anyone actually seen this show up in their account? Just wondering how accurate it feels in practice. My gut feeling is you get more insight from the individual terms.
Yeah, I saw that update too. On paper it’s a solid idea, but I think you’re right ... without export or API access it’s more of a dashboard toy than something you can actually build strategy around.
I’ve played with a few AI-based clustering tools lately, and most seem to hit the same wall: they sound smart but don’t match how Google actually groups intent in the SERPs. Sometimes totally unrelated keywords end up in one cluster because of phrasing.
For me, and as you already mentioned, the best proxy still comes from grouping by landing page in GSC, then checking SERP overlap if I want to go deeper. That already tells you most of what this “Query Groups” thing promises.
Would be nice if Google gave us a proper export though. Even a CSV would go a long way.
Haha, I love that we’re cautiously poking at this like it’s a new gadget we don’t quite trust yet 😅
I had the “Query Groups” thing show up in one of my GSC accounts last night. It’s… interesting? The clusters kind of make sense, but they’re not consistent. For one of my posts about eco-friendly kitchen swaps, it grouped “best bamboo utensils” with “eco-friendly dishwasher detergent,” which made me laugh.
Close, Google… but not quite.
That said, I do like the concept. I’ve spent way too many evenings sorting keyword spreadsheets, trying to find patterns manually. If this ever gets export access, it could actually save a ton of time for quick audits.
For now though, I agree with Rohan and Jordy. The landing-page view in GSC still gives cleaner insights. The AI just feels like it’s guessing intent based on vibes 😂 and AI makes mistakes!
It'll be interesting to see how the output changes week to week. Wondering if the groupings evolve as Google collects more data.
Good points, both of you. I was reading a Reddit thread about this earlier and someone mentioned that the only reliable way to cluster is still by checking SERP overlap. Sounds technical but basically, if two keywords trigger similar results, they probably belong in the same intent group.
Makes sense when you think about it. Google already does that internally, so anything purely AI-based is kind of guessing until it can verify through rankings.
If Query Groups ever reflect real SERP intent (and not just keyword similarity), that’s when it’ll actually be useful for planning clusters. Until then, I think I’ll keep doing it the old-fashioned way.
Although, to be honest, ChatGPT does a pretty good job of semantic clustering if you give it a list of keywords ;-) I don't do too much Google SEO yet either as my Amazon stuff takes up most of my time.
Okay, bear with me… I’m feeling a bit out of my depth here 😅
You all keep mentioning “intent clustering,” and I think I get the general idea. But I’m not totally sure why it matters in practice.
When I look at my own site’s data, I mostly just see different phrases people use to find the same post. So if Google’s already grouping those together, is that just a fancy way of saying “here are all the ways people asked the same question”?
Also… “checking SERP overlap” sounds very manual. Is it something you’d need tools for?
Sorry if these are basic questions, but I’m still learning how to read the data beyond just impressions and clicks. I can already feel the brain fog coming on!
Totally fair question. “Intent clustering” sounds more technical than it really is. My understanding is, it just means grouping keywords that reflect the same search goal.
For example, if people search “how to make compost at home,” “DIY composting for beginners,” and “best way to compost kitchen scraps,” they all want the same thing: to learn how to compost. That’s one intent cluster.
It helps you plan content more strategically. Instead of writing three separate posts, you can create one strong guide that covers the main topic well. Google usually rewards that approach because it matches how people search and avoids overlap.
When people talk about “checking SERP overlap,” it just means comparing the search results for two keywords. If most of the same pages show up for both, Google sees them as the same intent. Some SEO tools automate that, but you can do a quick version manually too (GPT can help a lot here).
You’re right though, the new Query Groups feature basically tries to do this for us automatically.
That’s a great breakdown Sarah -some awesome responses on this thread :-)
Rita, to give you a real world example, on one of my ecommerce sites I sell kitchen gear. I used to target keywords like “best coffee grinder,” “coffee grinder for home,” and “manual coffee grinder reviews” with separate posts. When I checked Google, the same competitors showed up for all three, which told me they were being seen as having the same intent.
Now I just have one solid page that covers all of those angles. It actually ranks better and saves me from trying to manage three similar posts. That’s intent clustering in action and I was also thinking another reason Google may favor this approach is it's less bandwidth from their side to spider/index/rank one page as opposed to three!
If Query Groups ever gets good at spotting that pattern automatically, it’ll be a big help for planning new content.