Finding out what people actually type
Most businesses guess at this, and they guess using the words they use
internally. Customers do not search that way. They search using the
problem they have, in the words they would use to describe it to a
neighbour — and there is often a sizeable gap between the two.
Keyword and content strategy is the work of closing that gap: finding
the searches that actually happen, working out which of them are worth
competing for, deciding which page should own each one, and then
writing those pages so they answer the question properly.
Done well, it stops you competing with yourself, stops you writing
pages nobody was looking for, and gives every piece of content a job.
It is the part of SEO that decides what gets built — which is why we
usually do it first.
- Real search data,
not assumptions about what people call things
- Every target search
assigned to one page that owns it
- Written to be read
by people and understood by machines
- Built up over time
into genuine depth on your subject
what's included
What the work involves
Keyword research
The searches people really make around what you sell, with
the volume behind each one and an honest read on how hard
it would be to compete for it.
Intent, not just volume
A big search term is worthless if the people typing it are
not buying. We separate the ones looking to hire from the
ones doing homework.
Competitor gap analysis
What the businesses beating you are ranking for that you are
not — usually the fastest way to find the pages missing
from your site.
Mapping searches to pages
One search, one page that owns it. This is what stops two of
your own pages competing for the same term and splitting
what they would each be worth.
Service and location pages
The pages that carry most local businesses: one per thing
you do, one per place you do it, each written to stand on
its own rather than swapping the town name.
Supporting content
The articles and answers that build authority around your
money pages, planned as a structure rather than published
whenever someone has an idea.
Writing that sounds like you
Written to be read. Pages stuffed with repeated phrases read
badly to customers and have not worked on Google for a very
long time.
Built for AI answers too
Questions answered directly and structured clearly, so
assistants like ChatGPT and Google's AI Overviews can quote
you rather than someone else.
how it works
From guesswork to a plan
01
Research
We pull the real search data for your market, look at what
competitors rank for, and find the gaps worth going
after.
02
Map
Every search worth having gets assigned to one page, and we
agree what already exists, what needs rewriting and what
has to be built.
03
Write & Review
Pages get written, published and then watched — what ranks
tells us where the next round of work should go.
This is a good fit if
- You are not sure which searches are worth chasing
- You have pages that get traffic but no enquiries
- Competitors outrank you and you cannot see why
- You are about to build a site and want the structure
decided by data
Usually part of something bigger
This work can be bought on its own, and sometimes that is
exactly right — you want the research and the plan, and
your own people do the writing.
More often it runs inside an ongoing SEO engagement, because
a plan is only worth what gets published from it, and
because what ranks in month three should change what gets
written in month four.
SEO services
questions
Questions we get asked
You can, and it is a useful starting point — you know your business far better than we do. What we will do is check it against real search data, because the terms an industry uses internally are frequently not the ones customers type. Some of your list will be exactly right. Some will turn out to have almost no one searching for it.
Either. We can write it, or we can hand you the plan and a brief for each page and let your own people write it — which often produces better copy, because they know the work. What matters is that somebody publishes it. Plans that sit in a document do nothing for your rankings.
However many distinct things you want to be found for, roughly. A business with four services in two cities has a different answer from one with thirty products. The mapping stage is where that number falls out, and it is usually smaller than people expect — a few pages that properly own their subject beat fifty thin ones.
No, and it has not worked for a long time. Repeating a phrase until the page reads badly does not help you rank and costs you the customers who do land there. What matters now is whether the page genuinely answers what was searched for, and whether the rest of your site backs it up.
Only for places you genuinely serve, and only if each page has something real to say about that place. Twenty near-identical pages with the town name swapped is a well-known pattern and it does not hold up. A handful of proper location pages will out-earn a directory of thin ones.
Google's position is that it cares about quality, not about how the text was produced. In practice the risk is that unedited AI output says nothing only you could have said — and that is exactly what separates a page that ranks from one that does not. Used to draft, with someone who knows the work editing hard, it is a tool like any other.