Intent classification
Every term tagged commercial or informational, so page type and messaging are decided before writing starts, not after.
Foundation research
Keyword research built on intent, not keyword volume alone
A high-volume term that nobody buys from is a vanity metric. We map demand by what the searcher actually wants, then build the site around that.
Keyword research done as a spreadsheet exercise produces a list of terms sorted by volume, most of which have nothing to do with how your buyers actually search. Keyword research done properly starts by asking what a person means when they type a phrase, whether they are ready to buy, comparing options, or just learning, and only then looks at volume and difficulty.
This distinction matters because it determines what kind of page you build. A commercial-intent term needs a page built to convert, with pricing context, comparison points, and a direct path to contact. An informational-intent term needs a page built to teach, with no hard sell, because that is what earns the click and the trust that eventually turns into a customer.
We treat this as the research layer underneath everything else in the SEO program. The content calendar, the internal link graph, and the priority order for building new pages all come out of this work rather than out of a separate brainstorm. If the research is wrong, everything built on top of it is aimed at the wrong target.
Difficulty scores from any single tool are treated as a starting estimate, not a verdict. We look at the actual page one results for a term, not just a number, because a term can show as difficult in a tool while the real competition is thin, outdated, or poorly matched to intent.
What it is
How the research actually gets built
Four layers, each one narrowing the last from raw demand down to a page-level plan.
We start by separating commercial intent from informational intent across every seed term connected to your business. Commercial terms are the ones closest to revenue; a searcher typing them is evaluating a purchase now or soon. Informational terms build authority and capture demand earlier in the decision, before a buyer knows which vendor they want. Both matter. They get different treatment.
From there, terms get grouped into clusters around a single shared intent rather than left as an unrelated list. A cluster might be built around one service and its variations, or around one buyer question and its related sub-questions. Clustering is what makes a content plan coherent instead of a scattered pile of blog topics competing with each other for the same click.
Difficulty gets assessed by reading the actual SERP, not just a tool's score. We look at what is currently ranking, how well it actually answers the query, what domain authority those results are carrying, and whether the SERP includes features like a map pack, a featured snippet, or an AI overview that changes what a winning page needs to look like.
The output maps directly onto execution. Each cluster gets assigned to an existing page that needs expansion, a new page that needs to be built, or a supporting page that should link into a stronger pillar. That mapping is what feeds the content plan and the internal link graph, so research never sits in a document nobody opens again.
Fit
Who this is for, and who it is not for
We would rather say no early than sell a program that cannot work.
Right fit
- You are planning a content program and want the topic list backed by real demand data before writing starts.
- Your current keyword targets were chosen by volume alone and are not converting.
- You are entering a new service line or market and need to know what the actual demand looks like first.
Not the right fit
- You already have a validated, intent-mapped keyword list and just need pages written against it.
- You want a single list of a thousand keywords with no clustering or prioritization attached.
- Your business has no service or product with search demand behind it at all.
Deliverables
What the research produces
A working plan, not a static spreadsheet.
Cluster map
Terms grouped by shared intent into pillar and supporting clusters, with the relationships between clusters made explicit.
SERP shape analysis
A read on what actually ranks for priority terms, including map packs, snippets, and AI overviews that change what a page needs to include.
Difficulty assessment
Realistic difficulty calls based on the competitive set actually ranking, not a single tool score taken at face value.
Commercial priority list
Terms closest to revenue flagged and ranked separately, so the pages most likely to convert get built first.
Page mapping
Every cluster assigned to an existing page, a planned new page, or a supporting link target, ready to hand to content strategy.
How we run it
How the research runs
Roughly two to three weeks, sized to the number of service lines and markets involved.
Step 1: Seed and expand
We start from your services, your existing rankings, and your sales team's language, then expand the term list using search data rather than guesswork.
Step 2: Classify intent
Every term gets a commercial or informational tag based on the actual language and the actual SERP, not assumptions about what a keyword sounds like.
Step 3: Cluster and prioritize
Terms group into pillars and supporting clusters, then get ranked by a combination of realistic difficulty and business value.
Step 4: Map to pages
Every cluster gets assigned an existing page, a new page, or a link relationship, producing a plan that content strategy can execute directly.
What makes keyword research intent-first?
It classifies every term by what the searcher actually wants, commercial or informational, before looking at volume or difficulty. That classification determines what kind of page gets built, which is why intent comes first and volume comes second.
Where this connects
How research feeds the rest of the program
This is input, not a finished product. Here is where it goes next.
Cluster maps turn directly into the calendar and page briefs built under content strategy so writing starts from validated demand instead of a blank page.
The page-to-cluster mapping also determines how pages should connect to each other inside the foundational SEO build since the internal link graph should mirror the cluster structure.
Commercial-intent terms with thin, weak competition become priority targets inside on-page optimization where existing pages get rebuilt to match intent directly.
SERP shape analysis often reveals where competitors are outranking on terms that matter most, which is the deeper question answered in competitor analysis as its own engagement.
Terms that surface as questions rather than searches are also the terms AI assistants get asked directly, which is why this research doubles as input to answer engine optimization and the answer-first pages that format demands.
Questions
Keyword research questions we get asked
Build your content plan on actual demand, not guesswork
We map intent and clusters before anyone writes a word.