Every beknown delivery produces the same four figures. Not because it is tidy, but because these are the four things that actually tell you whether AI systems can find your business, what they say about it, and whether that is sending you traffic. Everything else is decoration.
The Four Numbers are: isitagentready level, 16-Probe Scan score, Citation Rubric score, and AI referral traffic delta. Each one measures something different. Together, they give a complete picture of where a business stands in AI search.
Why a Fixed Format Matters
Think of a blood test. When a doctor orders one, they do not invent a new set of markers each time. They use the same panel, because the value is in the comparison: to normal ranges, to previous results, to what has changed. A bespoke set of metrics invented for each patient would be useless for spotting trends and convenient for hiding bad news.
Most digital marketing reporting is bespoke in exactly that way. Metrics are chosen after the fact. Numbers that moved in the right direction get included. Numbers that did not, quietly disappear.
The Four Numbers exist to prevent that. They are set at the start of every engagement, measured at the start, and measured again at the end. The format does not change based on how the results look.
Number One: isitagentready Level
isitagentready is a classification system that assesses whether a business's digital presence is structured for AI agent systems to understand it, recommend it, and in more advanced cases, transact on behalf of a user who has asked about it.
It returns a level, not a score. The levels run from undetectable (an AI system has no reliable way to identify or describe the business) through to agent-ready (the business has the full signal set required for agentic AI interactions). The levels in between reflect partial readiness: present but uncitable, citable but incomplete, complete but not agent-accessible.
This matters because AI search is not a single behaviour. A user asking ChatGPT for a recommendation is doing something different from an AI agent booking a service on their behalf. isitagentready measures readiness for the full range, not just the simplest case.
Most businesses start at level one or two. A successful beknown engagement typically moves a business to level three or four. Level five is the current ceiling for most sectors.
Number Two: 16-Probe Scan Score
The 16-Probe Scan is an audit of 16 specific signals that AI systems use when deciding whether to surface and cite a business. It returns a score out of 100.
The 16 probes map directly to the 5-Layer Framework: crawlability, structured data, llms.txt, entity signals, and WebMCP. Each layer contains multiple checks. A missing robots.txt directive scores differently from a missing llms.txt file, which scores differently from an absent organisation schema. The scan does not treat all absences as equal, because they are not.
The score is useful for two reasons. First, it identifies the specific gaps, not just their aggregate weight. Second, it creates a comparable baseline. A business that scores 31 at the start of an engagement and 78 at the end has a concrete record of what changed and by how much.
The scan is available publicly as a free check on beknown.world. It gives you this number before any other conversation happens.
What the Score Ranges Mean
A score below 40 means most of the signals AI systems rely on are absent or broken. The business may have a functioning website, but from an AI's perspective, it is largely invisible.
A score between 40 and 69 means some signals are in place, typically the basic ones. Structured data may be present but incomplete. An llms.txt file is probably missing. Entity signals are partial.
A score above 70 means the majority of signals are present and correctly implemented. The business is likely being cited by at least some AI tools. Above 85, the signal set is strong enough that consistent citation across major AI platforms becomes the expected outcome rather than the exception.
Number Three: Citation Rubric Score
The Citation Rubric measures what AI systems actually say about a business when they mention it. It grades four things: factual accuracy, completeness, sentiment, and source attribution.
A high search ranking tells you that a link appeared near the top of a list. It tells you nothing about whether the AI system described your business correctly, whether it mentioned the right services, whether the tone was appropriate, or whether it credited a source. The Citation Rubric measures all four.
Factual accuracy asks whether the AI's description matches what the business actually does. Completeness asks whether the description covers the key offering or leaves out critical details. Sentiment asks whether the tone is neutral, positive, or subtly dismissive. Source attribution asks whether the AI identified where its information came from, and whether that source is one the business controls.
Each dimension is scored. The aggregate is the Citation Rubric score. A business can have strong structured data and still score poorly on the Citation Rubric if the underlying content is thin, contradictory, or poorly organised for AI consumption.
Why Sentiment Is Included
Sentiment might seem like a soft measure in a framework that is otherwise structural. It is not. AI systems trained on large bodies of text absorb the aggregate tone of what has been written about a subject. If the dominant written record about a business is complaint-heavy review content, that register tends to surface in generated responses even when the AI is not explicitly drawing on review sources.
Monitoring sentiment in AI-generated citations gives early warning of reputation signal problems that structured data alone cannot fix.
Number Four: AI Referral Traffic Delta
The delta is the change in traffic arriving from AI-powered tools, measured between the baseline reading at the start of an engagement and subsequent measurement points.
Most analytics platforms now distinguish between traffic from AI sources and traffic from conventional search. The delta is calculated from this data. It is expressed as a percentage change and, where volume is sufficient, as an absolute figure.
This is the commercial number. The first three numbers measure whether the right signals are in place. This one measures whether those signals are producing visits. The two do not always move in lock-step, because AI systems index and re-index on their own schedules, and because citation does not always produce a click. But over a twelve-week horizon, a business that has moved from a 16-Probe Scan score of 31 to 78 should expect to see meaningful movement in AI referral traffic. If it does not, that is diagnostic information pointing to the Citation Rubric, not the structural signals.
Measuring the Baseline Correctly
The delta is only meaningful if the baseline is accurate. Beknown takes the baseline reading at the start of engagement, before any implementation work begins. Measuring after the first changes have been made produces a flattered baseline and an understate of the eventual result. The sequence matters.
How the Four Numbers Work Together
Each number measures a different layer of the same problem. isitagentready tells you what category of readiness you are in. The 16-Probe Scan tells you which specific signals are missing. The Citation Rubric tells you what quality of result those signals are producing. The AI referral traffic delta tells you whether the result is driving commercial outcomes.
A business can score well on the scan but poorly on the rubric. That usually means the signals are technically present but the content they point to is weak. A business can score well on both but show a flat delta. That usually means the AI tools have not yet re-indexed the changes. The four numbers together locate the problem more precisely than any single metric can.
This is the AI visibility case study format that every beknown delivery uses. It exists because vague reporting produces vague decisions. These four numbers produce specific ones.
Running Your Own Numbers
The 16-Probe Scan is available as a free public check on beknown.world. It gives you number two immediately, with a breakdown of which specific signals are present, which are missing, and what the gap means for AI visibility.
The other three require a baseline measurement at the start of a delivery. You can start with the scan.
Frequently Asked Questions
What are the Four Numbers in beknown's reporting format?
The Four Numbers are: isitagentready level (a structured readiness classification), 16-Probe Scan score (a signal audit across 16 specific checks), Citation Rubric score (a graded assessment of how AI systems describe your business), and AI referral traffic delta (the measurable change in visits arriving from AI-powered tools). Every beknown case study leads with these four figures.
Why does beknown use the same four numbers for every client?
Consistency makes results comparable and honest. Using the same format across every delivery means you can see where you started, what changed, and by how much. It removes the temptation to cherry-pick the metrics that look best on a given day.
What is a good 16-Probe Scan score?
Most businesses start below 40 out of 100. A score above 70 indicates that the majority of AI-relevant signals are in place. A score above 85 means a business is well-positioned to be surfaced and cited by AI search tools. The scan identifies the specific gaps holding a score down, not just the number.
How is the Citation Rubric score different from a search ranking?
A search ranking tells you where a link appears in a list. The Citation Rubric score tells you what AI systems actually say about your business when they mention it: whether the facts are correct, whether the description is complete, whether the sentiment is appropriate, and whether a source is attributed. These are different things.
How long does it take to see a change in AI referral traffic delta?
In most cases, measurable change appears within six to twelve weeks of implementation. AI crawlers re-index content on their own schedules, so there is no single switch to flip. The delta is measured against a baseline taken at the start of engagement, not against a theoretical maximum.
Can I see my own Four Numbers before becoming a client?
The 16-Probe Scan is available as a free public check on beknown.world. It gives you one of the four numbers immediately. The other three require a baseline measurement taken at the start of a delivery, because they need context: what your business is, how AI systems currently describe it, and where your traffic is coming from.
What does isitagentready actually measure?
isitagentready assesses whether a business's digital presence meets the structural requirements for AI agent systems to understand, cite, and transact with it. It is not a binary yes or no. It returns a level, from undetectable through to agent-ready, based on the signals present across the site and its wider entity footprint.