Structure for Scale
Systemantic is an evidence-backed organisational-diagnosis platform. It keeps three things apart — what people say happens, what your documents specify, and what your evidence actually shows — and then shows you where they disagree.
Ask three people how a process works and you get three answers. Read the procedure and you get a fourth. Look at the data and you often get a fifth.
Most assessment work resolves that tension early — someone decides which account is authoritative, writes it down, and the disagreement disappears into a document. What disappears with it is the most useful thing you had: the knowledge that your organisation does not agree with itself about how it operates.
Systemantic refuses to make that decision for you. Disagreement stays visible, attributed to whoever said it, for as long as it remains unresolved.
"We check every payment file before it goes out."
An interview, a walkthrough, a conversation. Real, and not the same as proof.
basis: testimonial
Control register: KMC-05, second-line review, monthly.
Policies, registers, procedures. What is meant to happen, which may be out of date.
basis: documentary
Run log: four settlement runs recorded in January.
Logs, exports, testing records. What actually left a trace.
basis: measured
Each step is separately checkable, and each one is recorded. Nothing in the chain relies on trusting the step before it.
A file, exactly as supplied
Identified by the hash of its own bytes, so a changed file is a new source rather than a quiet edit.
This file, at this spot, says this
Anchored to an exact span. Anyone can re-read it and check. Cheap to verify, and it settles nothing about the organisation yet.
The organisation is like this
A statement about how things work. Contested, carries its own qualifiers, and never merged with a claim that disagrees with it.
What holds the claim up
A link from claim to observation, carrying the basis it came from and how independent it is of everything else.
A named pattern worth your attention
Produced by a published rule, graded no more strongly than the weakest support beneath it.
Findings come from named, published rules — not from a model's impression of your documents. The same material always produces the same findings.
9 rules run today, each one asking a single question and stating plainly what it will not conclude. More will follow as real use demands them.
A policy or procedure says one thing; the operational record says another.
Two systems hold the same record and nothing checks that they still agree.
Something moves between two processes because a person carries it.
The control is written down. Nothing measured shows it has ever operated.
The control runs after the point where the mistake it looks for can still be undone.
Someone suspects two records describe the same system. The evidence does not settle it.
Part of the picture is held up by an assumption with no evidence behind it.
Nothing in the material says who is accountable for this control.
The control exists on paper but nothing connects it to the work it governs.
Below is a real finding from our proving corpus, followed all the way back to a single empty cell in a control register. Nothing is summarised away.
Kelmarsh Mutual is synthetic. It is not a customer.
We invented the organisation, wrote its 8 documents, and seeded known problems into them — then declared in writing what a competent analyst should expect Systemantic to notice, before running it. No real organisational material appears anywhere on this site or in the repository.
Finding
control:kmc_05 is in the model and no actor is recorded as accountable for it. The corpus does not say who owns it.
The detector asks: “Is there a control in the model that no actor is recorded as accountable for?”
Rests on the claim
controls(control:kmc_05, process:remittance_file_release, review)
Admitted by deterministic_derivation by map-control-register@1 — not typed in by hand.
Which the mapping read as
The control register records KMC-05 as covering the process remittance_file_release, as a review control.
Mapping rule, stated before the run: “The covers_process column names the process the control governs.”
basis: documentaryFrom this exact location
| control_ref | control_kind | owner_role | covers_process |
|---|---|---|---|
| KMC-03 | reconciliation | Finance Control Lead | empty |
| KMC-04 | review | Payments Operations Manager | exception_handling |
| KMC-05 | review | empty | remittance_file_release |
| KMC-06 | reconciliation | Finance Control Lead | empty |
Every other control in the register names an owner. KMC-05 does not. That empty cell is the whole finding — and in the workbench the anchor is re-checked against the file each time the page is rendered, rather than trusted.
Reproduced from the Kelmarsh Mutual proving corpus — synthetic material, described in full on this page. Finding id d9bdbb535e18… is stable: the same corpus produces the same id every run.
The record is append-only. Corrections are added, never written over the top, so you can always reconstruct what was believed before a correction arrived.
Sources are identified by the hash of their own bytes. A changed file becomes a new source rather than a silent edit of an old one.
A finding is graded no more strongly than the weakest evidence beneath it — and where it rests on an assumption with nothing behind it, it says so.
One run over the Kelmarsh Mutual corpus (synthetic)
8
documents in
73
claims admitted
9
findings out
0
unexpected findings
What it reads today
These are deliberate. Each one is a decision we have made and enforced in the product, not a feature we have yet to build.
Entering private beta
Systemantic is entering private beta and is not open for signup. We will begin with a small number of organisations.