AI Overviews answer the question directly, and the searcher moves on. If your keyword strategy is still built around "find high-volume terms and write content for them," you're optimizing for a search landscape that's already changed underneath you.
The keywords worth targeting in 2026 aren't necessarily the ones with the biggest numbers. They're the ones tied to intent specific enough that a generic AI-generated answer doesn't fully satisfy the searcher, and commercial enough that the clicks which do happen are worth winning.
That's the shift my keyword research is built around: not chasing volume, but finding where real opportunity still exists once you account for how people actually search now.
"Effective keyword research in 2026 focuses on intent mapping and topic clusters rather than volume alone. With AI Overviews absorbing generic queries, the highest-value opportunities are specific, commercial-intent terms where searcher needs aren't fully satisfied by a quick summary..."
A staggering share of published web content gets zero organic traffic, ever. Not low traffic, zero. The common thread isn't bad writing. It's keyword research that stopped at search volume and competition score, without asking whether that traffic was reachable or worth having in the first place.
Three shifts make the old approach unreliable on its own:
If a query can be fully answered in two sentences, Google's AI increasingly answers it directly in the search results, and there's no click left for you to win. Targeting purely informational, easily-summarized keywords is a shrinking strategy.
A single intent now spawns multiple related sub-queries that AI systems expand on conversationally. Ranking for one exact phrase matters less than comprehensively covering the topic ecosystem around it.
A high-volume keyword with broad, ambiguous intent often converts worse than a lower-volume keyword with clear commercial intent. Chasing the bigger number without checking the intent behind it is how businesses end up with traffic that doesn't turn into revenue.
Keyword research today is not about pulling the biggest search volumes from a tool and calling it strategy. I prioritize the intent behind the query first, then the search entities Google connects to it, such as products, problems, brands, services, and topics. That is how you find keywords that are not just searchable, but actually relevant, rankable, and commercially useful.
Before search volume, I look at what the searcher is actually trying to do, learn, compare, evaluate, or buy. That helps filter out keywords that may bring visits but never bring qualified leads, enquiries, or revenue.
Google no longer treats keywords as isolated phrases. It connects them to related entities like services, products, brands, locations, pain points, and topic relationships. I use that context to build keyword targeting that aligns with how search engines and AI systems actually interpret a topic.
Six steps that turn keyword research from a list-building exercise into a strategy that actually maps to revenue.
Keyword research that starts with a tool instead of your business gets the targeting wrong from day one. I start with your actual offer, your customers, and your competitive landscape, so the keyword list that comes out the other end is built around what you actually sell, not just what's searched for.
Every keyword gets evaluated against what the searcher actually wants: are they learning, comparing, or ready to buy? I prioritize keywords where your content can be the clearest, most useful answer available, and where the searcher's next step is plausibly choosing a business like yours.
Rather than handing you fifty disconnected keywords, I group them into topic ecosystems: a core pillar topic with the related questions, comparisons, and sub-intents that real searches actually spawn around it. This is also what builds the topical authority both Google and AI systems increasingly reward.
I look at what's actually ranking for your target terms now, including in AI Overviews, to identify where competitors have a real content advantage versus where there's a genuine opening for something better.
Every keyword gets assigned to a specific page, existing or planned, with a primary target and supporting terms. This avoids the common problem of multiple pages quietly competing against each other for the same query.
Search behavior shifts, AI Overview coverage expands, and competitors react. I revisit keyword performance regularly rather than treating research as a one-time deliverable that goes stale within months.
Two recent engagements where rebuilt keyword strategy produced measurable, business-moving outcomes.
A B2B SaaS platform in the drone technology space needed visibility in a search landscape being reshaped by AI Overviews and generative search. Rebuilding their keyword strategy around AI-era search behavior helped 13 of 15 target keywords show measurable AI Overview impact, while securing top rankings for high-volume commercial terms in their category.
A US health e-commerce brand had stagnant organic traffic despite heavy ad spend. Shifting their keyword targeting toward high-intent, long-tail queries their actual buyers were searching, instead of broad, high-volume terms, helped grow organic revenue from $10K to over $100K per month within a year, with organic sessions up 450%.
Common questions about how I approach keyword research, what's changed in 2026, and what you'll actually get.
Yes, but the target shifts. Generic, easily-summarized queries are increasingly absorbed by AI Overviews with no click left to win. The keywords still worth targeting are ones with clear commercial intent, specific enough that a quick AI summary doesn't fully satisfy the searcher, or where being cited as a source still delivers visibility even without a click.
Tools generate data. They don't tell you which of those keywords are actually reachable for your specific site, which ones convert versus which just generate traffic, or how to structure them into a coherent content strategy rather than a disconnected list. That judgment is where the actual value is.
Both. I look at how target topics show up in Google's traditional results and AI Overviews, since increasingly the same keyword research needs to account for visibility in ChatGPT, Perplexity, and Gemini as well, not just classic blue links.
Rather than a flat list, you'll get keywords organized into topic clusters mapped to specific pages, each with a clear primary target and supporting terms. The exact count depends on your site size and goals, but the structure matters more than the raw number.
At minimum, every 6 months, and sooner if you're in a fast-moving industry or if AI Overview coverage expands into more of your target queries. Search behavior doesn't hold still, and neither should the strategy built around it.
Usually, yes, if that content strategy wasn't built on the same intent-first, topic-cluster approach. Content built without proper keyword mapping often ends up competing against itself or targeting terms that don't convert. I can audit your existing content and tell you honestly where the gaps are.
Search has changed enough that yesterday's keyword strategy is actively costing you traffic today. Let's find the keywords that are actually worth building content around.