The four drivers behind AI-answer visibility

A baseline measurement tells you where a business stands in recorded AI answers. The next question is why — and that is where the four drivers come in.

Drivers are how Logres organizes a diagnosis. They group the evidence-backed findings from a site capture and the surrounding public record into four areas, so remediation work can be prioritized instead of scattered. They are worth understanding precisely, including what they are not.

The four drivers

Publisher Credibility — whether supporting public signals about the business are consistent and recognizable across independent sources. AI assistants lean on the public record; when that record is thin, inconsistent, or contradictory, the business is harder to describe confidently.

Topical Authority — whether the site clearly covers the services and decision questions people research. A buyer asking how to choose or what to expect should be able to find a clear answer; if the site does not address those questions, assistants have little owned material to draw on.

Structure & Extractability — whether a system can access and interpret useful page content and supporting markup. Clear structure, readable content, and sound technical fundamentals make the difference between material that can be used and material that is effectively invisible.

Sentiment & Reputation — whether captured evidence shows review or reputation signals that merit attention. Public reviews and reputation artifacts are part of the record assistants summarize; notable patterns in that evidence belong in the diagnosis.

What the drivers are not

This boundary matters as much as the definitions:

• They are not scores. Drivers organize findings; they do not rank a business. There is no "Publisher Credibility score," and presenting one would invent precision the evidence does not contain.
• They are not guarantees. Fixing a finding under a driver is a grounded recommendation, not a promised result. No one can guarantee that a specific change will produce a specific AI answer.
• They are not interchangeable with generic SEO checklists. The drivers describe how a business's public evidence feeds AI-generated answers, which is a different question from how a search engine ranks a page.

How drivers shape remediation

In practice, the drivers turn a measurement into a work plan. Findings from a verified site capture are grounded in sources, grouped under the relevant driver, and translated into an implementation-ready remediation bundle: proposed changes the client's team or partners can publish.

The sequence stays honest end to end. Delivery of a bundle is not publication; publication is not verification; verification is not a measured business outcome. A change counts as verified only when a later capture confirms it is live, and any later movement in measured scores is reported as an observation — not as proof of what caused it.

That discipline is what makes the drivers useful: they keep diagnosis organized, recommendations grounded, and claims bounded — so the work can be judged on evidence rather than enthusiasm.

Frequently asked questions

Are the four drivers scored?

No. Drivers organize findings; they are not rankings or per-driver scores. The scored dimensions are Visibility, Authority, Accuracy, and Influence — each reported separately, 0–100, and never averaged together.

If we fix a finding under a driver, will AI answers change?

There is no honest way to guarantee that. A remediation bundle contains grounded recommendations based on recorded evidence and cited sources. We verify that changes actually appear on the site with a later capture, then re-measure under the same pinned methodology where comparison is valid.

How do drivers relate to a remediation bundle?

Findings from a verified site capture are grouped under the relevant driver and translated into proposed, unpublished changes. Delivery of that bundle is not publication; publication is not verification; verification is not a measured business outcome.

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.