Two Audience Journeys, One AI Visibility Problem

The same company is now discovered in very different ways by very different people - and each group is asking AI systems their own questions.

In practice, we see two distinct approaches to AI visibility work:

• Within-funnel visibility: how customers and prospects find, evaluate and trust your answers when they’re actively trying to solve a problem.
• Above-funnel visibility: how investors, analysts, partners, procurement risk teams and other stakeholders form a view of your company and your market before any buying decision even exists.

They overlap, but they’re not the same journey. Treat them as one and the work becomes vague and hard to act on. Separate them and both become clearer.

Within the funnel: how customers research

This starts with the people you want as customers: decision-makers, influencers and users in your target accounts.

Their questions are practical and comparative:

• “How do I solve this in my organisation?”
• “What’s a credible approach?”
• “What should I ask vendors before I buy?”
• “How does X compare to Y?”
• “What does good look like in delivery?”

This is research with an intent to move forward - shortlist, justify, implement. The content that wins here tends to be grounded, specific and useful: guides, checklists, comparisons, implementation detail, proof.

Above the funnel: how the market forms a view

A different set of audiences starts with a different kind of question.

They’re not asking “How do I implement X?”. They’re asking things like:

• “Is this sector attractive or risky?”
• “Which companies matter in this space?”
• “Who looks credible, resilient and well-run?”
• “What are the structural risks and tail events here?”
• “Who is trusted in this market - and why?”

These are outside-in questions. They sit above any single buying decision and often come from people who may never become a customer - but who can still shape outcomes: capital allocation, analyst narrative, partnership decisions, supplier risk, and reputational context.

Why the approach has to differ by audience

The trap is trying to answer everything with product-led content.

An investor scanning whether a sector is investable does not want a product brochure. A supplier risk team doesn’t want brand statements. An analyst doesn’t want marketing language. And AI systems will mirror that: if the available material isn’t credible for the question being asked, the model will default to whatever it can find - usually competitors, third-party commentary, or generic summaries.

So the work needs two different inputs and two different test plans:

• Product visibility focuses on buyer journeys, evaluation steps and the questions that drive shortlists and decisions.
• Brand visibility focuses on stakeholder research journeys - macro context first, then drilling into named companies, evidence, track record, risk posture and credibility.

How Odyssiant handles both in one platform

Odyssiant is built as an AI visibility engine with two lenses, because the audiences behave differently.

1) Audience-specific prompt libraries - visible, editable, testable

Most tools hide the prompts and show a score. We do the opposite: we expose the prompt library and the answers, so you can see exactly what’s being tested, refine the questions, and understand why the outputs look the way they do.

2) Two audits, two outputs

We run separate visibility tests for the two journeys - one focused on product questions and one focused on brand/stakeholder questions - and produce outputs that reflect those different audiences, rather than forcing everything into one generic report.

3) Practical actions, not abstract metrics

The point isn’t a single “AI visibility score”. The point is knowing which questions matter, how you’re being described, where you’re absent or misrepresented, and what content changes will improve your position.

Why this matters now

AI-led research isn’t “coming” - it’s already how people start.

Customers are using AI to frame requirements and shortlist vendors. Investors and analysts are using it to scan markets and companies. Procurement and risk teams are using it to pressure-test credibility and delivery history.

If you only optimise for one journey, you leave the other to be shaped by whoever’s content the models can see and trust.

The questions have changed. The optimisation strategy has to reflect the audience asking them.