How it works

Two ideas sit under everything we build: changes should be safe, and AI should assist people, not replace their judgment.

The safe-change pipeline

Every change to a CALMA site follows the same six steps. Each step exists for a reason.

  1. Completed: 01RequestA content owner asks for a change. The work starts from a clear intent, not a guess.
  2. Completed: 02ChangeIt is prepared in a branch, never directly in production. The live site is never edited by hand.
  3. Completed: 03Automated checksBuild stability, metadata, sitemap, and critical user journeys are verified automatically. If something is broken, it is caught here.
  4. Active: 04PreviewA real preview URL lets a person review the change exactly as it will appear, before anything ships.
  5. Upcoming: 05Human approvalA person decides. Always. Automation prepares and checks; it never gets the final say.
  6. Upcoming: 06ProductionThe change is deployed, recorded, and reversible.
Checks

content/services/modernization.mdx

In preview
 production build             passed
 metadata + canonical URLs    verified
 sitemap + robots rules       generated
 critical user journeys       smoke tested
 preview environment           building
· human approval               required

AI assists the work. It does not remove responsibility.

AI tools can prepare content changes, create pages, inspect the repository, and run checks. That is real leverage, and we use it.

But reliable website operations still need clear rules, deterministic validation, previews, and human approval. We build the guardrails first, then let AI move fast inside them. That is the difference between AI that helps and AI that quietly breaks things.

See the modernization service

What is making your website harder to operate than it should be?

Show us the current site. We will identify the most important problems and the smallest sensible first step.

Show us your website

No platform pitch. No obligation to rebuild everything.