The baseline is only as clear as the question behind it

The broad question, 'Are we visible in AI?' sounds sensible but leaves too much unresolved. Visible to whom? For which service? At what point in the buying journey? Compared with which alternatives?

If those choices are not made before tracking begins, the resulting dashboard can become a mixture of unrelated observations. A mention for one service is placed beside a recommendation for another. A global competitor is compared with a local specialist. Early research questions are treated like supplier-selection questions.

The data may be accurate at answer level while the commercial frame remains weak.

One brand still needs a clear market identity

A brand can trade under different names, use several domains or describe itself in inconsistent language. A proper setup needs enough clarity to know what counts as the company being represented.

This is not a branding exercise. It is a measurement decision. If the tracking environment cannot recognise the relevant brand context consistently, leadership may misread absence, duplication or partial inclusion.

The setup therefore starts by defining the one brand in scope and the practical identity that should be observed.

The primary ICP gives the measurement a buyer

ICP means ideal customer profile. In plain English, it is the main type of company and buyer context the measurement is meant to represent.

A baseline for a procurement-led enterprise buyer should not be built around the same questions as one for a founder choosing a specialist partner. Their concerns, comparisons and language differ.

Choosing one primary ICP creates discipline. It does not claim the business has only one audience. It creates one coherent measurement foundation rather than mixing several buying situations into the same starting view.

A clear offer focus stops the data becoming generic

Many B2B companies sell several services, products or variations. If every part of the portfolio is included at once, the tracked questions can become too broad to reveal a meaningful pattern.

A defined focus connects the baseline to a commercial priority. It lets the company ask whether it appears around a particular need rather than whether the market has heard of the corporate name in general.

This is why prompt quality begins before wording. The commercial choice of what to measure is more important than producing clever variations of a vague question.

Next step: Explore the AI Visibility Setup >

Relevant competitors provide context, not a league table

A company appearing five times tells leadership little in isolation. The same count could be encouraging in one market and weak in another. Relevant competitors give the observation context.

The setup includes three to five selected competitors. Selection should reflect the alternatives a buyer might genuinely consider, not simply the largest or most familiar names in the category.

The purpose is not to create a competitor strategy. It is to make the initial measurement easier to interpret later.

Structured buyer questions connect the setup

The 25 structured buyer questions are the working bridge between the commercial context and the tracking environment. They should cover meaningful ways a buyer explores a problem, looks for suitable providers and compares options.

Publishing the complete framework would encourage people to copy questions without making the decisions that give those questions meaning. A useful set is not a bag of prompts. It is a designed measurement structure.

The initial run then creates a starting baseline. It establishes what was observed in the configured context. It does not diagnose why the pattern occurred or prescribe what to change.

Coherence matters more than prompt volume

It is tempting to assume that a larger prompt list creates a stronger baseline. Volume can improve coverage, but only when the questions belong to the same commercial frame. Fifty unrelated questions can create more noise than 25 structured questions built around one buyer, offer and competitive context.

Coherence means the questions can be understood as parts of one measurement design. They may cover different stages and needs, but leadership can explain why each belongs. The results can then be compared without mixing unrelated services or audiences.

This is also why the complete prompt framework is not a public checklist. Copying question wording without the setup decisions would reproduce the surface of the method while losing the part that makes the output useful.

The baseline needs stable definitions

Before the first run, the team should agree what counts as the brand being present and how variants will be treated. It should also agree which competitors are in scope and why, and how the question groups relate to the buyer journey.

These definitions reduce arguments after the data arrives. Without them, people can redraw the boundary to fit the result: a partial brand reference counts when it helps, an inconvenient competitor is dismissed as irrelevant, or a broad question is treated as important only after the company appears.

Stable definitions do not make the answer engines stable. They make the business's interpretation more consistent.

Next step: Explore the AI Visibility Setup >

What leadership should receive from the initial baseline

The most useful initial output is not a long list of raw prompts. Leadership needs a clear explanation of the configured context, the starting observations and the limits of the evidence.

That means being able to see where the company appeared, where competitors appeared, how the signals differed and which questions may deserve attention. It also means stating plainly that the initial baseline does not prove causes or create an implementation priority.

A baseline succeeds when the next conversation becomes more precise. Instead of asking, 'How do we improve AI visibility?', the team can ask, 'Is this repeated absence in a commercially important question group strong enough to investigate further?'

How to protect the baseline after launch

Once the tracking environment is live, changes to the configuration should be controlled. Adding unrelated questions, switching competitors frequently or broadening the offer focus can make later comparisons difficult to interpret.

That does not mean the setup can never evolve. It means changes should be recorded and made for a clear commercial reason. If a new audience or service deserves measurement, it may need a separate coherent context rather than being added casually to the existing baseline.

Leadership should also avoid judging every daily movement. The operating guide describes a daily run, but availability of frequent data does not make each movement significant. The useful task is to look for repeated, decision-relevant patterns.

A protected baseline gives future analysis something dependable to work from. Without it, every new result can restart the argument about what was measured.

A practical review rhythm

The baseline should enter an agreed review rhythm rather than become a source of daily alarm. A named owner can check whether the tracking has run, whether the configuration remains stable and whether any result deserves escalation. Leadership does not need a commentary on every movement.

A periodic commercial review can then focus on the question groups that matter. The team should bring the underlying answers for any material change, describe the signal accurately and compare it with relevant competitors and answer engines. If the pattern is isolated, it returns to observation. If it repeats, the business can decide whether deeper interpretation is justified.

This approach prevents measurement from creating its own activity burden. The dashboard remains part of a management conversation about buyer consideration, not another feed that people feel compelled to react to.

It also protects the later evaluation of any action. When the baseline, configuration changes and decision dates are recorded, leadership can see what evidence existed before work began and what changed afterwards. That does not prove causation, but it creates a far more credible learning record.

What should be recorded

Keep a simple measurement record beside the dashboard. It should state the brand and offer in scope, the primary ICP, the selected competitors, the question-set version, the answer engines observed and the date the baseline was captured. The purpose is not bureaucracy. It is to preserve the context needed for a fair comparison.

When the configuration changes, record what changed and why. A new competitor, different buyer group or revised question can improve relevance, but it can also break direct comparison with the earlier baseline. Leadership should know whether a movement reflects the market, the answer engine or the measurement design.

This record becomes particularly useful if deeper analysis is commissioned later. The diagnostic can begin from a known evidence set rather than reconstructing how the dashboard came to contain its current numbers.

Questions to test the measurement

• Can we state the one brand and offer being measured in a single sentence?
• Is the primary ICP specific enough to shape real buying questions?
• Would a buyer genuinely compare the selected competitors?
• Do the questions cover different decision moments without mixing unrelated offers?
• Can leadership explain what the baseline includes and what it cannot conclude?

Frequently asked questions

What does the CONVRG AI Visibility Setup include?

It includes one brand, one primary ICP, a clear product or service focus, three to five selected competitors, 25 structured buyer questions and an initial baseline.

Why use one primary ICP?

It keeps the questions coherent around one main buying context. Other audiences can be considered later without weakening the starting measurement.

Why limit competitors to three to five?

A focused group creates useful context while keeping the baseline centred on plausible buyer alternatives.

Does the setup provide a strategy?

No. It establishes measurement. Diagnosis, prioritisation and action planning are separate jobs.

Is the initial baseline a final conclusion?

No. It is the starting position for observing patterns over time and deciding whether deeper investigation is justified.

A reliable starting point comes before confident action

AI visibility measurement is most useful when it creates discipline rather than urgency. A clear setup gives leadership a defined starting position, makes future comparison possible and prevents one striking answer from carrying more weight than the evidence supports.

Create a structured AI visibility baseline

The CONVRG AI Visibility Setup costs £750 + VAT for selected existing HubSpot clients. It configures the agreed tracking context and captures an initial baseline, without turning measurement into unsupported diagnosis.