Search Engine Optimisation
Keyword Research
The output is not a spreadsheet of 4,000 phrases. It is a decision about which pages exist and what each one is for.
Keyword research establishes what your customers actually search, how hard each phrase is to win, and which are worth the effort at your size. The Nexclick delivers it as a keyword map assigning one primary phrase per page, which is what stops a site competing with itself.
Is this you?
What usually prompts the call
- You have a keyword list from a tool and no idea what to do with it.
- Several of your pages rank in similar positions for the same phrase and none of them ranks well.
- You are chasing head terms your budget cannot realistically win.
- You are about to build a site and need to know what pages it should have.
What we do
The actual deliverables
Things that appear on an invoice, not adjectives.
- Seed research from your actual customers
- Sales calls, enquiry forms, site search logs and Search Console queries first. What customers type rarely matches what a business calls its own services.
- Demand and difficulty assessment
- UK-specific volume, with difficulty judged from who currently ranks rather than from a tool score alone. A tool difficulty of 40 means nothing if the top five are national brands.
- Intent classification
- Every phrase marked as informational, commercial or transactional, because they need different pages. Targeting a transactional phrase with a blog post is the most common wasted effort in SEO.
- SERP feature analysis
- Which phrases return a map pack, a shopping carousel or an AI overview above the organic results — and are therefore worth less than the volume suggests.
- The keyword map
- One primary phrase per URL, with secondaries and long-tails clustered beneath it. No two pages target the same primary. This is the deliverable that prevents cannibalisation.
- Content gap and cannibalisation report
- Phrases competitors rank for that you do not, and places where two of your own pages are already competing and should be merged.
- A prioritised order to build in
- Which pages to create first, based on demand, difficulty and how close you already are. Ranking on page two is worth far more than starting from nothing.
Decision tree
Should you chase this keyword? A decision tree
Run any phrase you are considering through these in order. The first one that matches gives you your answer. This is the reasoning The Nexclick applies internally, written out so you can apply it yourself.
01The top five results are all national brands, marketplaces or Wikipedia
Do not chase it directly. Target a longer, more specific variant and build toward it. Competing head-on costs years and usually fails.
02The results are dominated by a map pack, and you have a physical location
This is a local SEO job, not an organic one. Your Google Business Profile will move faster and cost less than a content page.
03The results are all informational and you want to sell something
The intent does not match. Either write the informational page and accept it will not convert directly, or find a commercial variant.
04You already rank between 8 and 25 for it
Highest priority on the list. Improving an existing page to page one is several times cheaper than creating one from nothing.
05Volume is low but the phrase describes exactly what you sell
Chase it. Twenty searches a month with clear purchase intent beats two thousand from people who will never buy.
06An AI overview answers the question completely above the results
Expect low click-through even from position one. Worth targeting for citation and visibility rather than for traffic — see our AI search optimisation service.
07One of your existing pages already targets it
Do not create a second page. Improve the existing one. This single rule prevents most self-inflicted cannibalisation.
How it works
Step by step, with timeframes
Timeframes are typical rather than guaranteed, and they assume we get account access and approvals when we ask.
- 01Week 1
Listen first
Your Search Console queries, site search, enquiry forms and a conversation with whoever answers the phone. The best keyword ideas come from customers, not tools.
- 02Week 1–2
Expand and qualify
Tool expansion, competitor phrase extraction, and manual SERP checks on anything commercially important.
- 03Week 2
Map
Phrases assigned to URLs — existing ones where they fit, new ones where they do not. One primary each, no exceptions.
- 04Week 3
Prioritise and hand over
Sequenced build order with reasoning, walked through on a call so the logic transfers rather than sitting in a file.
What you get
Reporting and ownership
- The keyword map as a spreadsheet you own — one primary phrase per URL, secondaries clustered beneath.
- A build order with reasoning attached to each decision.
- The cannibalisation report, listing pages that should be merged or retired.
- A named list of phrases we recommend NOT chasing, and why.
- A walkthrough call so your team can maintain the map themselves.
Tools and platforms
- Google Search Console
- Ahrefs Keywords Explorer
- Semrush
- Google Keyword Planner
- AlsoAsked / AnswerThePublic
- Manual SERP analysis
- Site search logs
Timeline
How long this actually takes
Two to three weeks. The research is fast; the useful part is the argument about which phrases to ignore, and that benefits from a conversation rather than a delivery. Acting on it is the long part — a keyword map typically implies three to twelve months of content and optimisation work, and we will tell you which items pay back soonest rather than pretending it can all happen at once.
Pricing model
Fixed-price project
Fixed price by site and market size. Included at no extra cost as the first month of work within an SEO retainer.
Questions
Keyword Research questions
Is search volume data accurate?
It is directionally useful and precisely wrong. Tool figures are estimates from clickstream sampling, and they can be out by a factor of two or more on low-volume UK phrases. Use them to compare phrases against each other, not as a forecast. Your own Search Console impression data is more reliable for anything you already rank for.
How many keywords should a page target?
One primary, plus whatever variants and long-tails naturally sit beneath it — often dozens. What matters is that no two pages share a primary. A well-written page ranks for many phrases; a page trying to be about several things ranks well for none.
What is keyword cannibalisation and does it really matter?
It is when two of your own pages compete for the same phrase, so Google alternates between them and neither builds authority. It is real, it is common after a few years of publishing, and the fix is usually consolidation rather than optimisation. We find it on most established sites we audit.
Should we target low-volume keywords?
Often, yes. A phrase with thirty searches a month describing precisely what you sell is worth more than one with three thousand from people comparing definitions. This is especially true in B2B and specialist trades, where the total addressable search market is small and highly qualified.
Do keywords still matter now AI answers questions?
The underlying demand does. People still have questions and still express them in words — whether typed into Google or into ChatGPT. What has changed is that phrasing is more conversational and the click-through on informational queries is falling. Research now needs to account for which phrases still produce a visit.
Can we do this ourselves?
The data gathering, yes — the tools are available to anyone. The judgement is harder: deciding what to ignore, spotting when a SERP is unwinnable, and resisting the head term with the impressive volume. Our walkthrough is deliberately built so your team can maintain the map afterwards.
Last reviewed 28 July 2026.
Tell us what you are trying to fix
A 20-minute call, no pitch deck. The Nexclick will tell you what we would do, roughly what it costs, and whether we are the right people for it.