An engineer can tell in one
paragraph who wrote it
Developers do not read marketing prose. They open the docs, try to get something working, and close the tab on the first paragraph that sounds like nobody wrote it.
Written for developer tools and API-first companies, where the person you have to convince has a terminal open.
One full article free. No card required, no time limit.
What changes here
The evaluation starts in the docs, and you are not there for it
A developer sizing up your tool skips the landing page. They go to the docs, look for the shape of the request, and try to get one call to return something. If the answer is not there, they search. They land on whichever page solves it and form their opinion of you on that page, which is often not one of yours.
So the docs are the discovery surface, not a support artefact. They are also the part of your site written in specifics, which is why an assistant reaches for them when somebody asks it what to use for a job. The articles that do any work sit right beside them and get held to the same standard: the walkthrough for the integration nobody documented, the reason the first attempt fails, the migration written from the outside in.
And the honest one wins. A tutorial that says which step is fiddly, and what breaks once real traffic hits it, beats a confident one that pretends the path is clean. The reader finds out either way. Only one version of you is still standing afterwards.
Answer in the first screen
They came for one thing. Put it at the top and explain it underneath. A preamble is a page nobody scrolls past.
Say where it breaks
The step that catches people, what the failure looks like, what to do about it. That is the part they bookmark.
Nothing that smells synthetic
Hedging, filler and the giveaway phrasing come out before the draft reaches you. One of the six checks, and the one this reader tests first.
What cpywrk does not do is write your reference docs, and it does not read your codebase. It writes the long-form pages around them, in your voice, and you read every one before it goes out. If a draft gets a detail wrong, send it back with the detail.
The work itself
They searched a specific problem, not your category
The queries are narrow and literal. How to authenticate from the runtime you never got round to documenting. Why the webhook fired twice. What the rate limit does under load, and what to do when it bites. How to move off the tool they are on now without losing a week. Whether this can run in their environment at all.
These are the pages a developer reads to the end, and the ones an assistant can quote when somebody asks it what to use, because they answer something rather than position something. They are also the pages nobody at your company writes, since the person who knows the answer is shipping.
Take the question literally
The article targets a search somebody actually types, worked out before anything is written rather than picked in a planning meeting.
Show the failure, not just the path
Edge cases, the thing that only appears in staging, the assumption that quietly does not hold. Trust is built in that paragraph.
Say when the answer is no
The cases your tool is wrong for. Ruling yourself out where you should is what makes the rest of the page believable.
Get your own names right
Product names, terms and the phrasings you refuse to use are rules in your brand profile, applied to every draft instead of caught in review.
The technical read still has to happen, and it should. What changes is how often. A correction your engineer makes this month sticks, so nobody explains the retry behaviour twice, and the read gets shorter instead of staying exactly as long as it was the first time.
Where it ends up
A finished article still has to get onto the site
Everything above assumes the article reaches the site. On a bespoke site it does not get there by itself: the repo is the CMS, publishing is a pull request, and the only person who can merge it is mid-release. So finished work waits behind a ship date. The fix is to make publishing an input to the build rather than a task on somebody's branch.
Webhook, into your own build
The finished article, its metadata and its images are handed to whatever builds your site, so publishing stops waiting for an engineer to have a free afternoon.
WordPress, direct
If the marketing site runs on WordPress, connect it once. After that it is one button: press Publish and the article is live, formatted.
Publishing can also run on a schedule you set: a date and time per article in your own time zone, or a batch at one or two a day. Any scheduled article can be rescheduled or cancelled.
The LinkedIn, X and Instagram versions come out of the same brief, written for how each one reads. Those are yours to post when you choose. We write them; we do not post them for you.
Before it reaches you
Six checks, every article, before anyone reads it
This reader stops at the first hedge, so of the six checks every draft passes before it reaches you, the one that decides whether the page works here is the removal of machine tells.
Search
It targets a question people actually type, and it is built to rank for it.
AI answers
Buyers now ask an assistant to draw up the shortlist. The article is structured so those systems can quote it.
Brand voice
It sounds like your company, checked against rules you set once rather than re-explained every time.
Machine tells
Hedging, filler and the giveaway phrasing are removed before the draft reaches you.
Brief compliance
It covers what the brief said it would, rather than drifting somewhere easier to write.
Learnings
Corrections you made last month are applied this month, so you stop repeating yourself.
Before a draft reaches you it has also been checked for claims the research does not support, and every source link has been checked live. See every step up close.
You approve the outline before writing starts and read the draft before it publishes. Nothing goes onto your site that a person has not said yes to.
Proof
Read something it made and judge for yourself
We are not going to show you another vendor's traffic chart. What we can show you is our own work. Every article on this blog came out of the same system this page is describing: researched, written, checked, and published, then read and approved by a person before it went out.
The subject is content and search rather than your category, because that is what we sell. Read one anyway and answer the only question a sample can settle: is this the standard you would put your company's name on?
Read the blogQuestions worth asking first
- Will an engineer be able to tell that we did not write it by hand?
- The check that matters most on this page runs before you see the draft: hedging, filler and the phrasing that gives a machine away are taken out. That is one of the six, and this is the audience it exists for. Then you read it, because the technical judgement is yours and stays yours. If it reads wrong, send it back until it is right.
- Will it invent a flag or an endpoint that does not exist?
- It does not invent figures. A number comes from the brief you approved or from a source the article names, and every draft is checked for claims the research does not support. That check runs against the research behind the article, not against your codebase, so whether the code is right is still your call. It is one reason the outline comes first: you settle what the article will show while it is still a list of headings.
- Our marketing site is a repo and nobody in marketing can deploy to it. How does an article go live?
- By webhook. The finished article, its metadata and its images are handed to whatever builds your site, so publishing becomes an input to the build instead of a branch waiting for a merge between releases. An engineer wires it up once and is done with it. If your blog runs on WordPress instead, cpywrk publishes to it directly.
- Can it write our documentation?
- No, and you would not want it to. Reference docs are a product surface and they belong with the people shipping the product. What cpywrk writes is the long-form work around them: the walkthrough for the integration nobody has documented, the explanation of why the first attempt fails, the answer your support channel keeps typing out by hand. Those are the pages that reach somebody who has not heard of you yet.
A different kind of software
Or start from what content has to do for any software company.
Run one article on your own site and see
You bring the topic, approve one outline and read one draft. The research, the writing, the checks and the publishing happen without you, and if the draft misses, you send it back: revisions are unlimited and cost nothing from your plan.
Start free trialOne full article free. No card required, no time limit.
Would rather not run it yourself? We can run it for you.