Kenaptic is coming
We are building something specific, for a problem most teams have learned to live with. Before we show you the product, it is worth asking whether you recognise the problem.
Here is a test you can run this week without buying anything.
Take the last twenty support tickets you closed. For each one, ask a narrow question: was the answer already written down somewhere we own? Not “could we have written it” — was it already there, in the documentation, a knowledge-base article, a blog post, or a forum thread from eighteen months ago that solved exactly this.
Most teams who try this are surprised by the number. Not because their content is bad, but because the answer was in a different place from where the customer was standing.
The shape almost everyone ends up with
It happens the same way every time, and nobody decides it.
Product documentation goes on one platform, because the docs team picked it. The knowledge base goes on another, because support bought a help desk. Marketing runs the blog. The community lives in a forum, or Discord, or GitHub Discussions — and that is where the real troubleshooting knowledge accumulates, the workarounds and the “actually, in version 4.2 you need to…” replies that never make it back into anything official.
Four teams. Four tools. Four search boxes that each see a quarter of the picture.
None of it is anyone’s fault, and none of the individual decisions were wrong. The problem is the space between them, and nobody owns the space between them.
What that costs, without anybody logging it
A reader arrives on a documentation page that describes the feature but not the failure they are hitting. The page is accurate. It is well written. It is simply not the page they needed — and there is nothing on it pointing at the page they did need, because the two live on different platforms and no tool can see both.
So they do one of three things. They open a ticket you did not need to receive. They ask in the community, and someone answers a question that was already answered. Or they quietly decide your product is harder than it is, and you never hear about that one at all.
That third outcome is the expensive one, and it is invisible in every dashboard you have.
Four questions
If you want to know whether this applies to you, these are more useful than any demo:
Can a reader get from your documentation to your community without going back to a search engine? For most estates the honest answer is no — the two are linked at the navigation level, if at all, not at the level of the specific page and the specific thread that answers it.
When your community solves something your documentation does not cover, what happens next? If the answer is “someone might notice and file an issue”, that is a demand signal you are collecting and not reading.
How would you find out if two of your own pages contradicted each other? A changelog says a default changed; a tutorial still teaches the old one. Both were correct when written. No linter, test or reviewer of a single page can catch it, because the error only exists between two documents nobody reads together.
And how much of your content can an AI assistant actually reach? Coding agents and answer engines navigate by following links. A page nothing points to is, to them, very nearly a page that does not exist.
If those questions landed, you already know the problem. You have probably tried to fix it by hand at some point, and found that it is not a task that stays finished — the estate changes every week, and a link added today is stale next quarter.
What we are building
Kenaptic connects a content estate at the source. It reads every silo you point it at, works out how the pages genuinely relate — this one explains that, this thread troubleshoots that feature — and proposes the links that ought to exist between them.
Two things we will say now, because they are the parts people are right to ask about first.
A person approves every change. We are not asking anyone to let a tool edit their published pages unattended. Kenaptic proposes; your team decides.
Everything it adds can be removed in one click, exactly, back to what the page was before. A tool that writes into your content should be boringly easy to undo, and we would rather be judged on that than on a feature list.
Not yet, but soon
We are not open for signups. This is a note to say what is coming, and to leave you with the twenty-ticket test, which costs an afternoon and tells you more about your own estate than we could.
If it turns out you have the problem, we would genuinely like to hear how it shows up for you — what you have tried, and what did not work. That is more useful to us right now than another sign-up.
Kenaptic proposes; a person decides. Every link is reviewed before it publishes and retractable in one click — see Security.
