How we build

The rules this website is built under.

Apyrn is a young company with no customer logos to show, so trust has to come from somewhere checkable. These are the publishing and engineering rules the site enforces on itself — in the build, not in a style guide.

Claims discipline

If a claim can't be defended, the build fails.

Every page of this site passes through a claims validator before it ships. A registry of approved claims pairs with a list of prohibited patterns — invented customers, quantified outcomes, absolute promises, claims of replacing systems of record, autonomous-architecture language — and a match anywhere in the published content stops the release. The registry and patterns live in the same repository as the site, so the rules are versioned, reviewable, and enforced the same way for every author.

The standing rules

Four rules, machine-checked where a machine can check them.

No invented proof

No fabricated customers, logos, statistics, or outcome numbers. Proof appears when it exists and is attributable; until then, its absence is stated plainly.

Maturity, labeled

Product cards and pages carry their real release stage — flagship, controlled rollout, early access, in development — and a pre-release stage is never softened into sounding shipped.

Systems of record stay authoritative

Apyrn adds governed interfaces over ERP, CRM, SaaS, and legacy systems. Copy that implies replacement of those systems is prohibited by pattern.

Fictional data says so

Every product demonstration on this site uses fictional data and is labeled as such — down to the FICTIONAL DATA stamp in the homepage drawing.

Engineering the same way

The site is built like the product is meant to be operated.

The same discipline applies below the copy line.

Content as code

Every page, claim, and translation lives in versioned, schema-validated files. Nothing is published that a reviewer cannot diff.

Validation gates

Content structure, internal links, claims, design tokens, SEO metadata, and accessibility checks run on every change in continuous integration.

Static first

Pages are prerendered to static HTML and hydrate only when you interact, so the full content is readable without executing scripts.

Hold us to it.

If anything on this site reads as an overclaim, tell us — the fix ships through the same gates.