The claims we refuse to make

Every industry has its snake oil, and AI visibility is young enough that the snake oil is still being invented. The easiest way to evaluate anyone in this space — including us — is to listen for what they promise.

Here is what Logres does not claim, as a matter of policy.

We do not guarantee outcomes

We do not guarantee rankings, citations, recommendations, traffic, leads, revenue, or a specific return on investment. AI answers vary by user, provider, model version, and time. Anyone who promises a specific answer from a system they do not control is selling certainty they do not have.

We do not dress estimates up as measurements

A free teaser observation is a bounded, market-level signal — it is never presented as a paid score. When the evidence needed for a score is missing, the score is reported as unavailable with a reason, never filled in with an estimate. And no operator internals — costs, warmup configuration, outreach mechanics — appear in client-facing claims.

We do not blur the implementation lifecycle

A delivered remediation bundle is not a published change. A published change is not a verified implementation. A verified implementation is not a measured business outcome. Each step has a distinct name because collapsing them is how vendors turn "we sent a document" into "we improved your visibility."

We do not call correlation causation

A before/after sequence can document that remediation was delivered, verified live, and later re-measured. It cannot, by itself, establish that the work caused a result. A changed score is an observation to review, not proof of cause.

Why this protects the client

Bounded claims mean every number in a Logres report can be traced to a recorded prompt, response, source, and date. You never have to wonder which part of the story is marketing. When someone else promises you a guaranteed AI ranking, apply the same test: ask which recorded evidence the promise rests on.

Frequently asked questions

If you guarantee nothing, what are we buying?

A dated, inspectable measurement; a diagnosis grounded in stored evidence; grounded recommendations; verification that changes actually appear on your site; and later re-measurement under a pinned methodology. What you are buying is evidence and disciplined work — not a promise.

Do you ever show before/after results?

Yes — as observational records. We document the sequence from remediation to verification to later measurement, and we say plainly that the sequence alone does not prove causality.

Why refuse to estimate missing scores?

Because a number that looks precise but is not supported is worse than no number. An unavailable result tells you what evidence to gather; a fabricated estimate tells you nothing while pretending otherwise.

This note is for information only. Logres measurements describe recorded AI responses for a defined set of questions at a point in time; they are not a guarantee that any AI system will name, cite, or recommend a business in the future, and they do not predict rankings, traffic, leads, or revenue.