A software company that writes its own products and operates them on hardware it owns. Nobody here has a title, and every decision is voted on.
Structure
Owner-operated · no titles
Funding
Revenue only · no investors
Where
Three sites, one country
Trading for
— days
Scroll
Part two · On the record
Six questions, answered without a communications department.
Question 01 / 07
Where did this company come from?
From irritation, mostly. Years of watching software shipped by people who had never had to run it, and infrastructure sold by people who had never had to talk to whoever was using it. Shardly started in 2025 to do both halves of that job in one building, with a spreadsheet that had to make the electricity bill work before anything else was allowed to happen.
Question 02 / 07
Who owns it?
One person, outright. There is no board, no investor, no second shareholder and no holding company above it. Every decision recorded on this page was made by somebody who has to live with it, and who is named in the imprint as the law requires.
Question 03 / 07
How big are you, honestly?
Between one and ten, depending on the month. We publish a range rather than a number, because inflating a headcount is the first lie a small company tells and you have to keep telling it afterwards. Nobody here has a title that exists to impress a procurement form.
Question 04 / 07
Who pays for all of this?
The people who buy from us. There has never been an investor, a grant or a line of credit — including the year it would have been very convenient. Everything the company owns was paid for out of revenue, which is slow and makes the argument about what to do next very short.
Question 05 / 07
So what kind of company is this?
A software company that operates what it writes. Two competences, held on purpose: knowing how to run things properly, and knowing how to deal with the people using them. Most companies are decent at one of those and outsource the other; doing both is the whole reason this one exists.
Question 06 / 07
What are you actually trying to build?
A company that is still here in ten years and still small. Not a category leader, not a platform play, not an exit. The measure we care about is whether somebody who bought from us in the first year would buy from us again in the tenth.
Question 07 / 07
Why believe any of this?
Don’t. Check it. The registration, the operator details and the commitments on this site are all verifiable, and every figure we publish carries its measurement method beside it. Where we cannot show our working, we leave the sentence out rather than soften it.
Where did this company come from?
From irritation, mostly. Years of watching software shipped by people who had never had to run it, and infrastructure sold by people who had never had to talk to whoever was using it. Shardly started in 2025 to do both halves of that job in one building, with a spreadsheet that had to make the electricity bill work before anything else was allowed to happen.
Who owns it?
One person, outright. There is no board, no investor, no second shareholder and no holding company above it. Every decision recorded on this page was made by somebody who has to live with it, and who is named in the imprint as the law requires.
How big are you, honestly?
Between one and ten, depending on the month. We publish a range rather than a number, because inflating a headcount is the first lie a small company tells and you have to keep telling it afterwards. Nobody here has a title that exists to impress a procurement form.
Who pays for all of this?
The people who buy from us. There has never been an investor, a grant or a line of credit — including the year it would have been very convenient. Everything the company owns was paid for out of revenue, which is slow and makes the argument about what to do next very short.
So what kind of company is this?
A software company that operates what it writes. Two competences, held on purpose: knowing how to run things properly, and knowing how to deal with the people using them. Most companies are decent at one of those and outsource the other; doing both is the whole reason this one exists.
What are you actually trying to build?
A company that is still here in ten years and still small. Not a category leader, not a platform play, not an exit. The measure we care about is whether somebody who bought from us in the first year would buy from us again in the tenth.
Why believe any of this?
Don’t. Check it. The registration, the operator details and the commitments on this site are all verifiable, and every figure we publish carries its measurement method beside it. Where we cannot show our working, we leave the sentence out rather than soften it.
Part three · The ledger
Every decision that made the company, and what it cost us.
A history that lists only the wins is a brochure. This one is kept the way an accounts book is: each entry names the decision on one side and the price we paid for it on the other.
2025 · Q1
Registered a company instead of keeping a side project
What it cost
Paperwork, liability and the obligation to be findable and answerable. In exchange: people could sign something, and we could be held to it.
2025 · Q2
Chose to run our own software rather than ship and leave
What it cost
Capital tied up in hardware, an on-call rota, and no way to shrink in a quiet month. In exchange: the people who wrote a thing are the people who get woken up by it, which changes how it gets written.
2025 · Q3
Ended the first investor conversation after twenty minutes
What it cost
The growth we could have bought, and a year that was tighter than it needed to be. In exchange: nobody outside the company has ever needed a return from it.
2025 · Q4
Wrote our own tooling instead of licensing somebody else's
What it cost
Two engineering quarters and a maintenance burden that never ends. In exchange: when something is wrong, the fix is ours to make rather than ours to request.
2026 · Q1
Published our own numbers, including the bad month
What it cost
One customer left the week we did it, and said so. In exchange: every figure on this site can be argued with, which is the only reason to print one.
2026 · Q3
Separated the company from what it sells
What it cost
A second site to write and keep honest. In exchange: this page can be about who we are rather than about a sales funnel.
Part four · How decisions get made
Nobody here is a CEO.
There is no chief anything. No head of, no VP of, no lead. The company has one structure and it fits in a sentence: every decision is proposed, argued and voted on, one vote per person, and the founder holds a final say that is meant to be expensive to use.
Chief executive
No titles at all
Not flat-ish, not “flat with clear ownership”. Nobody has a title, on a business card or in an email signature, because a title is a claim about whose opinion counts before the argument has happened.
One vote each
Everything is voted
Hardware, pricing, what to build next, whether to take a meeting. Anyone can propose; everyone votes; the tally is written down. Seniority does not weight a vote, and neither does having been right last time.
One veto
The founder can override it
Somebody has to carry the legal and financial responsibility, and that person can overrule a vote. It has been used once. Every use is recorded below, because a veto nobody can count is just a rumour about who is really in charge.
The voting record
Five decisions, and how the room actually split.
01
Buy the machines instead of renting them
2025 · Q2Carried
3 for1 against0 abstained
The one against was about cash, and was right about the cash.
02
End the investor conversation
2025 · Q3Carried
4 for0 against1 abstained
The abstention wanted to hear the second meeting first.
03
Shorten the billing cycle to a week
2025 · Q3Carried
3 for2 against0 abstained
Closest vote we have had. Revenue predictability was the argument against.
04
Publish the bad month
2026 · Q1Carried
4 for1 against0 abstained
We lost a customer that week. The vote would go the same way again.
05
Start a second project before the first was comfortable
2026 · Q2Vetoed
3 for2 against0 abstained
Carried on the vote, overridden by the founder. Revisited a quarter later and started properly.
06
Write our own control plane rather than license one
2025 · Q2Carried
4 for1 against0 abstained
Two engineering quarters. The vote against was about the opportunity cost and it was not wrong.
07
Open a third site
2026 · Q1Carried
5 for0 against0 abstained
The only unanimous one. Backups sharing a building with the thing they protect had been bothering everybody.
08
Never work with recruitment agencies
2026 · Q1Carried
4 for0 against1 abstained
The abstention pointed out we have never hired anybody, which was fair.
09
Split the company site from the product
2026 · Q3Carried
3 for1 against1 abstained
The vote against thought it was vanity. Partly it was.
10
Put the voting record on the public site
2026 · Q3Carried
4 for1 against0 abstained
You are reading the result. The argument against was that we would have to keep doing it.
Part five · The refusals
Six things this company will never do.
Values are cheap in the affirmative. These are written as refusals because a refusal is the only kind of promise that costs something to keep.
We will not take outside investment.
The moment somebody outside the company needs a return, the company starts optimising for them instead of for the people paying it. There is no round planned and no exit planned.
We will not publish a number without its method.
Every figure on this site says how and over what window it was measured. If we cannot show the working, the sentence gets cut rather than softened.
We will not make leaving difficult.
Exports work, configuration stays portable, and cancelling takes one click with no retention flow behind it. Somebody who stays because leaving is hard has already left.
We will not run analytics on you.
No analytics product, no tracking cookie, no third-party script anywhere on this site. That is written into the privacy policy, so enabling one would mean rewriting it in public.
We will not claim a certification we do not hold.
No badge wall, no borrowed compliance, no award nobody has heard of. When we hold something, it will be named with its scope and its date.
We will not grow faster than the revenue.
Headcount is the easiest thing to add and the hardest to undo. The company stays small enough that the person who built a thing is the person who answers for it.
We will not hand anybody a title.
No chief anything, no head of anything. Every decision is voted on instead, and the tally is published — including the one the founder overrode.