Most teams do not have a keyword problem. They have a decision problem: which terms deserve a page, which deserve a section, and which deserve nothing at all. Our keyword research services turn a raw export into that decision — seeds expanded, intent scored, and a cluster map your writers can pick up on Monday morning.
Scope-based quotes only. No seat licences, no dashboard you have to learn.
collectedindustrial fasteners
expandedstainless steel hex bolts supplier
assigned/stainless-steel-hex-bolts/
field note 01
Almost every export we are handed looks the same. Four hundred rows, a volume column, a difficulty score, and no owner. Nobody has decided which row belongs to a service page, which belongs to a blog post, and which belongs to a comparison table buried halfway down an existing page. So the file is renamed, shared, and forgotten.
The second failure is volume-first sorting. Broad head terms look like the prize, so a manufacturer with twelve pages chases "industrial fasteners" for a year while the phrase "stainless steel hex bolts supplier" — one hundred and forty searches a month, low difficulty, obvious commercial intent — sits unclaimed and converts inside a quarter.
The third is intent by assumption. A page is built for "what is a torque wrench", then fitted with a request-a-quote form and an instant chat prompt. The searcher wanted a definition. They leave. Meanwhile a page built for "torque wrench supplier" is given two thousand words of explainer and no price band, no lead time, no certifications — and it lands on page two because nothing on it answers a buying question.
And the quiet one: nobody writes the negative list. Job seekers, students, DIY hobbyists and people looking for a repair shop all get counted as demand. If those clicks never had a chance of becoming an inquiry, remove them from the plan before they shape the plan. When existing pages are already misaligned, we usually pair this work with SEO Content Writing Services so the fix list and the keyword map agree with each other.
A keyword without an assigned page is a note. A keyword with an assigned page, an intent and a CTA is a plan.
what you get
We start from your product families, quotation language, and the words buyers use in RFQs. Then we expand through autocomplete, related searches, People Also Ask, supplier directories, trade forum threads and your own sales inbox. Every seed is traceable, so you can see why a term is on the list.
Each term is checked against the live SERP: what ranks, what the top three pages actually are, and whether an inquiry form would feel natural there. Transactional, commercial investigation, informational, navigational — the label decides the page template before the brief is written.
We group terms by how far the buyer has walked: problem-aware, solution-aware, vendor-aware, ready-to-buy. A B2B buyer reads four to eleven pages before they ever fill a form, so a plan with only ready-to-buy terms starves the top of that path and makes every later channel work harder.
Volume tells you the size of the room. Difficulty tells you whether you can get in the door this year. We read them together against your domain's current link profile, then place each term in one of three buckets: build now, build after authority, or ignore and cover inside an existing page.
A written exclusion list for the terms you never want to rank for — employment, repair, classroom, free, second-hand. It keeps content briefs honest, keeps paid search from re-learning the same lesson, and stops future writers from re-adding the same low-value blog topics.
One core commercial term per cluster, four to nine supporting pages, and the internal linking direction drawn as arrows. The map is the same artefact your developers will read, so new URLs stop appearing because someone had a hunch on a Friday.
Every mapped page gets a one-page brief: target term, secondary phrases, intent, page type, word range, questions to answer, proof points to include, and the CTA it should carry. Our SEO Audit Services team writes from that brief; yours can too, which is often the cheaper route.
field note 02
Intent is not a dropdown you set from memory. We read the SERP and describe what the winning page is doing, then attach the template and the call to action it implies. The matrix below is the actual working artefact — the column on the left is where the budget conversation starts.
One row per intent, one template per row. When two pages claim the same row, we merge or rewrite — never publish a third.
field note 03
A cluster is not a tag. It is a promise: one commercial page carries the buying intent, and a small set of supporting pages answers the questions that arrive before the buying decision. The arrows below matter as much as the boxes, because internal linking is how a small site tells a search engine which page is the parent and which pages are the support act.
We keep clusters deliberately small — four to nine supporting pages. A cluster of thirty is a content factory with no owner, and it usually collapses at page six when the writer runs out of genuine questions to answer. Small clusters ship, and shipped clusters earn links.
Location and map terms get their own branch rather than being sprinkled everywhere. "Fastener supplier Ohio" and "fastener supplier" are not the same page: one needs a service-area block, a local phone number and driving-distance proof, the other needs export terms and container loading data.
Once the map is signed off, the order of production is set by commercial value per hour of work, not by volume. That single change is usually worth more than any tool subscription in the stack.
how the work runs
Read your site, your quote templates, your sales objections and your top twenty inquiries. Pull export data from two sources minimum, plus autocomplete and forum wording. This is where the words buyers actually type come from — not from a template list.
Every surviving term is opened in a browser and judged by what ranks. If the top three results are all marketplace listings, a product page can compete. If they are all government PDFs, a blog post will not.
Terms are sorted into build now, build after authority, and fold into existing pages. The output is a short list you can fund, not a museum of everything the tool found.
URL, target term, intent, template, word range, CTA, internal links in, internal links out. Owners named. Anything without an owner is deleted from the map rather than left to rot.
Briefs go to writers with a rejection checklist. Thirty days after publishing we compare position, impressions and assisted inquiries, then re-cut the map based on what the data says — not on what the tool predicted.
Want the map before the invoice? Send us a URL and the three products you actually want to sell.
from the case notes
An industrial fasteners exporter, 6 months: mapped pages went from 34 to 118, and monthly organic inquiries moved from 4 to 17. Nine of the new inquiries came from a single long-tail cluster we nearly deprioritised.
manufacturing · EU + USA packaging supplier had eleven pages chasing the same commercial term. We merged them into three, redirected the rest, and impressions dropped briefly before clicks rose 38% on the surviving pages.
cannibalisation cleanupA test-equipment brand moved from volume-first to intent-first ordering. Same publishing budget, same writer, three times the inquiry-assist rate per published page over two quarters.
prioritisation onlyscope, not seats
We quote on scope: how many product families, how many existing pages need re-mapping, whether local branches are in play, and whether you want briefs written or just the map. There is no per-seat licence and no annual platform fee — you are buying decisions and documents, not access.
Most projects land in one of three shapes: a single-cluster sprint for a product launch, a full-site re-map for a site with sixty pages or more, or a quarterly refresh where we re-cut the map against fresh SERP data and your own inquiry log. If you also need the technical groundwork fixed before content ships, that work sits under our SEO Services team and runs in the same calendar.
— the field notes desk
questions we get asked
Mapping takes about three weeks. Ranking movement on low-difficulty commercial terms usually appears between week six and week twelve after the pages are published and internally linked. Competitive head terms are a two-to-four-quarter story, which is why we rarely lead with them.
Both, and neither is trusted alone. Paid exports give coverage; the live SERP, autocomplete, forum threads and your own inquiry emails give truth. Every term we recommend has a human check against what currently ranks for it.
Fewer than you expect, and that is deliberate. A typical single-cluster sprint produces 60 to 140 scored terms, of which 12 to 25 become pages and the rest become sections, FAQs or negative entries. Volume of rows is not the deliverable; assignment is.
Yes, and it is usually the fastest win. We audit existing pages for intent mismatch and cannibalisation first, then decide whether each page should be kept, merged, redirected or retired. New URLs only get created where the map proves a genuine gap.
Yes, as a separate branch with its own template requirements: service-area copy, consistent business details, location-specific proof and a phone number that a human answers. We do not stuff city names into a national page and call it local SEO.
Access to analytics and Search Console, a list of the products you actually want more of, and one person who can approve the map. A thirty-minute call with someone from sales is worth more than another tool export, because they know the phrases buyers use on the phone.
Three numbers: how many mapped pages are indexed and ranking in the top twenty, how many distinct commercial terms drive impressions, and how many organic inquiries the CRM attributes to those pages. Rankings alone are a leading indicator, not the score.
who does the work
The people who build your map are the people you talk to. Our strategists have run search programmes for exporters, industrial suppliers and B2B software teams since 2017 — the kind of accounts where a single well-mapped product page is worth more than a hundred lifestyle blog posts.
We publish our method, show the artefacts, and put a name and a phone number on the work. If a term is not worth chasing, we say so in writing, with the reason, because a plan that survives contact with your sales team is the only plan that gets executed.
Every deliverable is reviewable by someone who is not an SEO: a sales manager should be able to read the map and disagree with it intelligently. That is the test we hold ourselves to, and it is why our briefs are one page rather than twelve.
If your site is already ranking but the inquiries are wrong, the problem is usually intent assignment, not authority. Start with the map, not with more content.
