Fewer pages,
that bring in enquiries.

Generating a hundred articles in a week has become trivial, and that is exactly why it is no longer worth anything. The market for volume writing has collapsed in on itself: what is produced without effort does not differentiate, and since quality is assessed at the scale of the site, a stock of weak pages actively damages the few strong ones already online.

We take the problem the other way round. Few pieces, dense, linked to each other, written on searches that carry a verified buying intent. One page bringing two qualified enquiries a month is worth more than thirty pages bringing traffic that goes nowhere.

That assumes knowing something others do not. It is the real work, and it cannot be subcontracted by the kilo.

It is the stage that decides the return on everything else, and the one most often skipped because it produces no text.

The trap of high volume

A keyword searched a great deal but with no buying intent brings traffic and zero enquiries. It is the most disappointing scenario in search work: the figures rise, the curves look good, and nothing arrives in the inbox.

Conversely, a search made ten times a month by people looking for exactly what you sell can fund a role. So we prioritise on intent and on real difficulty, never on the volume a tool displays.

How we qualify an intent

• Look at what the search engine already shows. If the first page is made of definitions and tutorials, the query is informational: a service page will not answer it, however good it is.
• Read the variants. They reveal who is searching. A query surrounded by “example”, “template” or “free” is not speaking to a buyer.
• Test the commercial prefix. On a young domain, the variant prefixed with “agency” or “firm” is often three times less difficult for a far better intent.

What we systematically set aside

Searches too difficult for a domain without authority, however attractive they look. Fighting over a generic, highly competitive query amounts to spending a budget against sites established for ten years, while the searches that convert stay free.

The ranking of a piece of content does not rest on keyword density, a notion abandoned long ago. It rests on four things, and none of them is a trick.

Answering more completely than the others

A page ranks when it covers the question better than those already there. That assumes reading them first, identifying what they do not say, and writing precisely that. It is analytical work before it is writing.

Answering fast

The main answer is given in the first lines, not in the third paragraph after an introduction about the context. A reader who has to hunt for the answer leaves, and that departure is measured.

A structure that reflects the reasoning

Subheadings that genuinely describe the content, not hollow phrases. Lists when the information is a list. Tables when it is comparative. Formatting is not decorative: it makes the text scannable, and most readers scan before they read.

Internal links in both directions

This is the point most often missed. A set of pages all pointing to a main page without ever receiving a link back passes nothing on: they are dead ends. Every piece we write receives links and gives them, and the main page of the subject is rewritten to welcome its satellites.

It is what we apply on our own site: our comparisons, such as n8n or Make, point to the service pages, which point back to them.

The question always comes up, and it deserves a straight answer rather than a posture.

What they are good for

Roughing out a plan, rephrasing a passage that resists, spotting what has not been covered, producing a first draft you will rewrite. Refusing those uses would be an expensive affectation.

What they are not good for

Producing the published text. Not on moral principle, but for a simple economic reason: what anyone can generate in three minutes cannot distinguish you from anyone. A piece that adds nothing beyond what can already be found does not rank, and it weighs on the rest of the site.

What cannot be generated

A decision you actually made. A mistake you made and corrected. An order of magnitude you measured. A case where your solution was not the right one, and why. That is the material that makes the difference, and it comes from you, not from a tool.

Our way of working takes that into account: we interview you before writing, and the interview lasts an hour. It is the most profitable part of the engagement, and the first thing volume offerings remove.

The market for search writing is one where the promise is easy and the verification slow. A few practices are common and we do not offer them.

A fee per word

Charging by the word pushes towards writing long, when the quality of a page rests on what it says and not on its length. A 2,000-word page that answers better than a 4,000-word one costs more to produce and should cost more, not less.

A guaranteed monthly volume

Four articles a month, indefinitely, mechanically produces weak pages as soon as the good subjects run out. We prefer short batches, decided on the return from the previous one.

A ranking guarantee

Nobody controls a search engine’s ranking. We commit to what we control: the subjects chosen and why, what is written, the internal linking put in place, and what is measured.

And a reservation we own

This market is highly competitive and the basket is often small. If your need is to produce a great deal of content cheaply, we are not the right supplier and we will say so at the first conversation rather than at the third quote. Our work makes sense when a few pages have to carry qualified enquiries, not when an editorial calendar has to be filled.

The method counts more than the style, because it is what explains why one page brings enquiries and another does not.

Step 1: understand who you sell to

Before any subject, one question: who signs, and what question are they asking themselves just before they look for a supplier? The answer is not found in a tool. It is found in your sales conversations, in the objections you hear, and in the reasons you were told no.

That step lasts an hour and determines everything else. An agency that starts with the keyword list starts at the end.

Step 2: select and qualify

We build a short list of candidate searches, then qualify them one by one: what the search engine already shows, what the variants say, and the real difficulty for a domain like yours. Many candidates fall at this stage, and that is the point.

The deliverable is an ordered list with, for each subject, what it targets and why it comes before the next. You can challenge it: it is a decision document, not a formality.

Step 3: the interview, which is the real work

One hour with the person on your side who knows. We are looking for what no tool can produce: a decision you settled, a mistake you corrected, an order of magnitude you measured, a case where your solution was not the right one.

It is that material that makes a page impossible to copy. Without it, you write what everybody writes, with the result everybody gets.

Step 4: write, then cut

The first draft is always too long. The useful work is the second pass, the one that removes what adds nothing, tightens the transitions and checks that the main answer arrives in the first lines.

Step 5: link, and rewrite the neighbourhood

A new page means retouching those around it, so that they cede the intent to it and send it links. A page published without anything else moving is an isolated page, and an isolated page passes nothing on.

Step 6: measure, and derive the next batch

Six to twelve weeks later, we look at what the page produced: rankings, enquiries, the path taken before the enquiry. The next batch is decided on that, not on the list you started with. That is what separates steered production from a volume subscription.

That choice deserves an explanation, because it goes against market practice.

An isolated piece produces almost nothing

A single page, with no internal links and no neighbouring pages covering the same subject, rarely ranks. What works is a cluster: a main page surrounded by satellites each covering one precise question, linked in both directions.

Charging per page therefore amounts to selling an element that only works as part of a set. It is convenient at the point of sale and disappointing in use.

The useful work is not the writing

On a well-run engagement, the time divides roughly as follows: a third choosing and qualifying the subjects, a third understanding your trade well enough to write something others cannot, a third writing and linking. A price per page only pays for the last third, which explains the quality of the market.

What the support covers

Selecting the subjects, production, internal linking, the technical fixes that condition the return on the content, and a short monthly report. One task is handled at a time and you set the order.

When it makes no sense

On a site that almost nobody visits and whose subject nobody searches for, we say so and point the budget elsewhere. It is also what we do on the SEO and performance page: the technical side and the structure before the content, because content published on damaged foundations gives back nothing.