How your website becomes a source that AI systems can trust.

We read your website the way an AI reads it—you don’t need to install anything. The compact audit counts your conflict groups and initially estimates the number of your core entities—supported by one or two examples. These figures form the basis of the later offer and the first measurement baseline.

We read your website the way an AI reads it—you don’t need to install anything. The compact audit counts your conflict groups and initially estimates the number of your core entities—supported by one or two examples. These figures form the basis of the later offer and the first measurement baseline.

Onboarding as a single sum: The counted conflict groups determine the fixed-price tier of conflict resolution, plus the connection as a fixed item and the two options (remove clutter, machine-readable framework). The sum is a binding cap: If clutter is removed, we recount the conflicts afterward and credit a lower tier—so the sum can decrease, but only increase with your new signature.

Old and duplicate pages burden the picture that machines draw of you: the trade show archive, the pages maintained twice by two departments, orphaned campaign landing pages. We deliver the removal list and the redirect plan; your team or your agency implements; we approve. The double effect: Your message becomes clearer—and conflict groups that disappear with the pages are credited after the recount.

Each conflict group receives exactly one decision—prepared by us, signed by you. Corporate truth is not generated. It is owned. Then it is corrected. Your team implements; we direct: precise specifications, weekly office hours, verification of each group. If we are to make the text changes ourselves—everything from one source—the 1.5× rate applies. Building new pages or renewing technology is not included; that is a separate project, typically for your agency.

“Never before—otherwise you cement errors.”

The gate verifies: All conflict groups identified within the verified scope are decided, implemented, and verified closed—your website is contradiction-free within the verified scope. Only then is building for machines begun. Contradiction-freedom is verified, not completeness; there may still be gaps, just no competing truths. The model may vary. Your facts may not. Anyone who has cleaned up with their own team enters here: We verify the result (QA gate approval) and release it.

Make your domain the place where the correct version begins: Our software reads in the entire website; the WordPress plugin (free) or your CDN (technology free, introduction and operation see additional services) is set up. Then the completion scenario runs through the full monthly cycle of governance—not a showcase, but the first completed scenario of your build plan: Together we determine your core entities (typically twenty to thirty identity-defining entities such as your organization, locations, services, key people, and products) and agree on the build plan. The organizational facts are established as the binding version, completed, linked, and externally confirmed; the datasets of the organizational entities go live manually verified—they are included in the package. And attestation #1 measures across 9 AI systems what has changed: You leave onboarding with a documented before/after comparison, not with a promise.

Optional addition: the machine-readable framework—all other core entities also receive their verified dataset at the level “contradiction-free within the verified scope” (technical format: JSON-LD), delivered on every page that mentions them. Still without enrichment: This framework becomes valuable in governance scenario by scenario. No dataset leaves us without human verification.

If you book governance, the “Build” plan works through your agreed build plan—one scenario per month. The compact audit estimates your scenario fields (for mid-sized companies typically five to eight: “What services do we offer?”, “How does our service work?”, “What do applicants find?”); in onboarding, this becomes the prioritized build plan. A scenario bundles up to ten involved entities and up to ten verification questions—larger topics are divided by the build plan into clearly named sub-scenarios. For each scenario, the involved entities are completed, linked, and externally confirmed, the datasets enriched, manually verified, and delivered—and the measurement series documents the impact: measured before the work, after implementation, and again in the following attestation. Search-supported answers respond in weeks, the foundational knowledge of models in months—we document both honestly.

Your pace belongs to you: Our monthly deliverable—findings, decision templates, recommendations—is always delivered; if your team implements later, measurement happens later. At most two scenarios are open simultaneously; if both are not yet implemented, no third starts: Your contract automatically switches to your capacity-appropriate maintenance plan (“Maintain” €800 or “Maintain Plus” €1,200) until a scenario is implemented—then the build continues. Nothing expires, and you never pay build price for waiting. And the build ends as planned: Once the build plan is worked through, you switch to maintenance. A build plan is also an exit plan from the build price.

After the build, the “Maintain” plan holds the state—with three clearly separated instruments:

New scenarios arise from your business, not from our imagination: a new product, a new language, a relaunch, an acquisition—or an entity that gains weight on the watch list. Then the “Build” plan returns for the counted new scenarios.

Your scope never grows silently—neither does your price.