SEO keyword and search intent research
Keyword and intent research identifies the searches your potential customers make, groups them by the need behind them, and maps each group to a page that should satisfy it. The output is a plan for which pages to create, improve or merge, not a spreadsheet of terms.
By ROI SynthUpdated 11 min read
The problem this solves
Search volume estimates from tools are approximations, and many valuable queries are too specific to show any volume at all. Intent is the more reliable guide: a query that shows guides and definitions in its results needs a different page from one that shows product listings or comparison tables.
Research that stops at volume produces content aimed at broad terms the business cannot satisfy, while missing the specific, lower-volume questions buyers ask late in their decision. It also causes duplicate pages when similar keywords are each given their own article.
What the work includes
- Query collection from Search Console, tools, site search and customer conversations
- Clustering of queries by shared intent and result type
- Mapping of each cluster to an existing or new page
- Prioritisation by business relevance, competition and effort
How the work is done
Gather queries
Combine tool data with your own first-party sources, which reveal language competitors cannot see.
Read the results
Review what currently ranks for representative queries to understand the expected content type.
Cluster and map
Group queries that one page can satisfy and assign each group to a page.
Prioritise
Order the plan by value to the business, not by volume alone.
When your pages and your buyers use different words

Keyword research is often treated as a list-building exercise: export thousands of terms from a tool, sort by volume, hand over a spreadsheet. That approach produces numbers but rarely decisions. The research we do has one purpose — to decide which pages your site needs and what each one must answer — so it starts from your buyers and ends with a page plan.
It is most valuable when your website was written in your own vocabulary rather than your customers'. A firm that calls its service "digital transformation advisory" while buyers search for "help moving our accounts to the cloud" will struggle no matter how technically sound the site is. Research closes that gap with evidence.
- You are about to build new pages or a new site and want to know which pages to build.
- Search Console shows impressions for queries your pages do not really answer.
- Several of your pages seem to target the same searches, and none performs well.
- Your team disagrees about what customers search for, and nobody has checked.
- You are entering a new market, service line or audience.
Why intent matters more than volume
Two searches can share most of their words and want completely different things. "Payroll software" is usually someone comparing products. "Payroll software login" is an existing customer. "How to run payroll for one employee" is someone who may not be ready to buy at all. Sending all three to the same page serves none of them well.
We group searches by the need behind them, then decide what kind of page satisfies each group: a service page, a comparison, a guide, a pricing page, a location page, or no page at all. Search results themselves are part of the evidence. If the top results for a query are all step-by-step guides, a sales page is unlikely to be what searchers or search engines want there.
| Intent | Example search | Page that usually fits |
|---|---|---|
| Understand a problem | why is my site not indexed | A guide that explains causes and fixes |
| Compare options | headless cms vs traditional cms | A fair comparison page |
| Find a provider | technical seo agency for next.js | A service page with clear scope and price |
| Check a detail before buying | how much does a website migration cost | A pricing or FAQ section on the service page |
| Reach something specific | brand name login | No marketing page; make sure navigation works |
Where the evidence comes from

No single source gives a full picture, so we combine several and weight them by reliability. Your own data comes first, because it reflects people who actually found you or contacted you.
Search Console
The queries you already appear for, with impressions and clicks, which reveal both demand and gaps in your current pages.
Your enquiries and sales notes
The words customers use when they write to you, which are often closer to real searches than industry terms.
Search results themselves
What ranks today, what format it takes and which questions appear around it.
Keyword tools
Useful for discovering related terms and rough relative demand, read with caution because volume estimates vary widely.
Competitor sites
Which topics and page types others cover, as a check on blind spots rather than something to copy.
What you receive: a page plan, not a keyword dump
The deliverable is a map from search needs to pages. Each row names an intent group, the representative searches in it, the page that should serve it (existing or proposed), the action needed, and a priority based on commercial value and effort. It is short enough to read in one sitting and specific enough to act on.
For a hypothetical commercial cleaning company, one row might read: "Office cleaning contracts — searches about regular office cleaning, contract terms and pricing — served by /services/office-cleaning, which exists but does not mention contract lengths or what is included; action: rewrite with pricing factors and an FAQ; priority: high." A developer, a writer and a manager can all work from that line.
- Intent groups with representative searches and the evidence behind each
- The page that should own each group, existing or proposed
- An action for each: create, improve, merge, or leave
- Priorities ranked by commercial value, confidence and effort
- Notes on gaps where demand exists but you have no offer, for you to decide
Turning research into templates and content models
On a coded site the research often reveals page types, not just pages. If buyers search consistently by industry, the site may need an industry template with fields for sector terminology, regulations and examples. If they search by integration, an integration template. Those decisions shape the content model and routes, which is why research is best done before a build or restructure rather than after.
The research also informs the metadata rules in your templates: what each page type's title should lead with, how descriptions should be written, and which headings belong in the page structure. Those rules are then implemented once and apply to every page of that type.
A hypothetical research project
A hypothetical recruitment software company believes its buyers search for "applicant tracking system". Search Console agrees that the term matters but shows something the team had missed: many impressions for searches about specific problems — "how to schedule interviews automatically", "reduce time to hire", "recruitment software for small business" — landing on the homepage or a blog post, with few clicks.
Grouped by intent, the evidence suggests four needs the site does not serve well: small businesses looking for a simple system, teams trying to automate interview scheduling, buyers comparing the company with two named competitors, and people researching pricing. The page plan proposes a small-business landing page, a feature page for scheduling, two fair comparison pages and a clearer pricing page with an FAQ. Each is tied to the searches it should serve and to the evidence that those searches exist.
How the research runs
Research is delivered within the SEO offer, usually in the first month, and refreshed periodically as new data arrives and the business changes.
| When | What happens | What you see |
|---|---|---|
| Week 1 | Search Console and enquiry data reviewed; conversation about services, priorities and language | Initial demand themes |
| Week 2 | Searches gathered from all sources and grouped by intent; results pages reviewed | Draft intent groups with evidence |
| Week 3 | Groups mapped to existing and proposed pages; priorities scored | Draft page plan for review |
| Week 4 | Plan agreed; first pages scheduled; metadata rules updated where needed | Final page plan and the next month's scope |
How we know the research was right
Research is a set of hypotheses about what people search for and what they want. The pages built from it test those hypotheses. We track, per intent group, whether the intended page becomes the one shown for its searches, whether impressions and clicks grow, and whether visitors from those searches take the next step.
Where a group underperforms, we revisit the evidence: perhaps the intent was misread, the page does not yet answer the need well, or the competition is stronger than expected. The page plan is a living document, updated as the data comes in.
Research mistakes that lead to the wrong pages
These are the patterns that most often send content effort in the wrong direction.
Chasing volume over fit
Targeting the biggest terms in your industry, which are often informational or dominated by large publishers, instead of the specific searches buyers make.
One page per keyword
Creating separate pages for close variants of the same need, which then compete with each other.
Ignoring the results page
Building a sales page for a query where every result is a how-to guide, or the reverse.
Using internal vocabulary
Naming pages with the terms your team uses rather than the ones your customers type.
Treating research as finished
Searches, products and competitors change. A plan that was right two years ago may now miss whole needs.
When research differs by country or language
If you sell in several countries, research has to be done per market, not translated. People in different countries use different words for the same thing, care about different details, and see different competitors in their results. A term that dominates searches in one English-speaking market may be rare in another, and a direct translation of a keyword into another language often misses how native speakers actually search.
For multilingual sites we research each priority market separately, with native-language input where possible, and then decide whether the page plan can be shared across markets or needs to differ. That decision feeds directly into international SEO strategy and, later, into how pages are connected with hreflang.
What we need from you
The most valuable input is often the simplest: a sample of real enquiries, with personal details removed, and an hour with someone who talks to customers. Those two sources regularly reveal language and needs that no tool shows. We also need read access to Search Console and analytics, and a clear list of the services and markets you want to grow, so the research is prioritised around your business rather than around search volume.
The deliverables, listed
Research ends in a document your team can work from immediately.
- Intent groups with representative searches and the evidence behind each.
- A page plan mapping each group to an existing or proposed page, with an action and a priority.
- Notes on demand for services or markets you do not currently offer, for you to consider.
- Recommendations for template metadata rules where page types need them.
- The source data, so the research can be revisited and extended later.
Seasonal, local and emerging patterns
Search demand is not constant. Some searches peak at predictable times — tax deadlines, holidays, the start of a heating season — and planning pages ahead of the peak gives them time to be found. Some vary by place, so a service may be searched for very differently in a city and in rural areas. And some are new: emerging terms for technologies, regulations or problems that tools have little data on yet.
We look for these patterns explicitly. Seasonal demand shapes when pages should be published or refreshed. Local variation informs whether location pages are worthwhile. Emerging terms, often spotted first in your own enquiries or Search Console, can be an opportunity to publish the clearest answer before competitors notice the topic at all.
Why low-volume searches can matter most
Keyword tools often show specific searches as having little or no volume. That does not mean nobody makes them; it often means the numbers are too small or too varied for tools to estimate. Yet these specific searches — a service plus a sector, a problem described in a buyer's exact words — are frequently made by people close to a decision. A page that answers them well can bring fewer visits and more enquiries than a page chasing a popular general term. Your own enquiries and Search Console data are the best evidence for them.
How research connects to other work
Research is the foundation for several other pieces of work. It shapes information architecture, because intent groups often become sections and hubs. It drives the content strategy and roadmap, which sequences the pages over months. And it guides on-page optimization, where existing pages are improved to serve the needs the research identified.
Competitor analysis is a natural companion: organic competitor analysis shows why other sites win the searches you care about, which informs what your pages must do better. For businesses with many similar pages to create, the research also decides whether programmatic SEO is justified.
Which offer includes keyword and intent research
Research is part of the SEO offer, from $200 per month, and of All-in-One SEO + AEO + GEO, from $300 per month, where it also feeds the question list for answer-first content. AEO, from $150 per month, includes question research focused on the questions buyers ask before contacting you.
If you are planning a new site, doing the research before the Next.js + TypeScript Website Build, from $1,000 one time, means the page list and templates are designed around real demand from the first day.
Questions about keyword research
Common doubts, answered.
"We already have a keyword list"
Good — it is a useful input. We test it against your own search data and map it to pages and actions.
"Will you tell us how much traffic we will get?"
No honest forecast is possible from tool estimates. We tell you where demand exists and how confident the evidence is.
"Do we need to target high-volume terms?"
Not necessarily. Specific, lower-volume searches often bring buyers closer to a decision.
"How often should research be repeated?"
Revisit it when you change services or markets, and review Search Console data regularly. Monthly SEO work does the latter routinely.
How progress is evaluated
- Impressions and clicks for the mapped clusters
- Number of clusters with a dedicated, appropriate page
- Conversions from organic landing pages
Who this is for
- New sites planning their content
- Established sites whose content growth has stalled
- Teams commissioning content from writers
Limits and honest caveats
Common questions
questions
What is SEO keyword and search intent research?
Keyword and intent research identifies the searches your potential customers make, groups them by the need behind them, and maps each group to a page that should satisfy it. The output is a plan for which pages to create, improve or merge, not a spreadsheet of terms.
What does a SEO keyword and search intent research engagement include?
The work includes query collection from Search Console, tools, site search and customer conversations, clustering of queries by shared intent and result type, mapping of each cluster to an existing or new page and prioritisation by business relevance, competition and effort.
What problems does SEO keyword and search intent research solve?
Research that stops at volume produces content aimed at broad terms the business cannot satisfy, while missing the specific, lower-volume questions buyers ask late in their decision. It also causes duplicate pages when similar keywords are each given their own article.
How is SEO keyword and search intent research carried out?
It runs in 4 stages: gather queries, read the results, cluster and map and prioritise. Combine tool data with your own first-party sources, which reveal language competitors cannot see.
Who is SEO keyword and search intent research a good fit for?
It suits new sites planning their content, established sites whose content growth has stalled and teams commissioning content from writers.
How do you measure the results of SEO keyword and search intent research?
Progress is measured with impressions and clicks for the mapped clusters, number of clusters with a dedicated, appropriate page and conversions from organic landing pages.
What are the limits of SEO keyword and search intent research?
Third-party volume figures are estimates and vary between tools. Intent can shift over time as results change.
Can ROI Synth carry out the SEO keyword and search intent research work, or only recommend it?
Implementation is included. Within the agreed scope we fix templates, metadata, internal links and redirects, and build the landing pages ourselves, on your existing platform or on a Next.js and TypeScript build when a rebuild is part of the project, starting with steps such as “Query collection from Search Console, tools, site search and customer conversations”. Every completed item is recorded with evidence in a weekly progress sheet.
How long does SEO keyword and search intent research take?
The 4 stages (gather queries, read the results, cluster and map and prioritise) fit the proposal's four weekly phases. When the agreed scope can be delivered in a month, the proposal commits to completing it in the first month, subject to timely access and approvals. Search engines then take weeks or months to re-evaluate the changes.
Is SEO keyword and search intent research guaranteed to bring more traffic or leads?
No. Rankings, traffic, leads and sales depend on search engines, competition and your market, so they are not guaranteed. What is committed is the agreed work, completed and evidenced, and measurement of impressions and clicks for the mapped clusters and number of clusters with a dedicated, appropriate page.
Sources and further reading
Contact us about SEO keyword and search intent research
Tell us about your website and goals. We will say plainly whether this is the right offer, what it would include, and what it can and cannot achieve.
Discuss SEO keyword and search intent research for your business
Tell us about your website and goals. We will confirm whether this work is the right priority, what it would include and how it would be delivered.
