AI visibility is not measured by a single ranking. It depends on which brands are selected, which questions they appear in, and which competitors are shown instead.
PickMeLabs uses a structured experiment process to separate random AI output from repeatable visibility patterns.
Generative AI is becoming part of how consumers decide what to buy — not just how they search.
Capgemini Research Institute's 2025 consumer trends research found that 58% of consumers prefer product recommendations from generative AI tools over traditional search engines. Adobe Digital Insights also reported that generative AI-driven traffic to U.S. retail sites rose 4,700% year-over-year in July 2025.
For ecommerce teams, this means AI visibility is becoming part of the same discovery system as SEO, content, PR, and onsite experience.
Sources: Capgemini Research Institute, What Matters to Today's Consumer 2025; Adobe Digital Insights, Generative AI-Powered Shopping Rises with Traffic to Retail Sites, 2025.
The same brand can appear in one response and disappear in another. A broad query may surface one set of competitors, while a more specific query may surface an entirely different group.
That means a single prompt is not enough. To understand AI visibility, brands need a structured way to test multiple buyer questions, compare outputs, and identify patterns across systems.
The framework focuses on three core questions: where you appear, where you don't, and what's driving the difference.
We measure how often your brand is included across a structured set of buyer-style queries.
We identify the query types where competitors are selected and your brand is absent.
We compare your brand against the brands being selected to identify differences in positioning, authority, language, and context fit.
Every experiment follows a consistent five-step process. The structure is the same across categories — what changes is the query set and the brands being tested.
We start by defining the category, key competitors, and the types of buying decisions customers are likely to ask AI about. This keeps the experiment focused on commercially relevant visibility, not generic brand mentions.
We create a structured set of buyer-style queries across broad, constrained, and comparison-based intent. These are designed to reflect how a real customer might ask for recommendations before searching or visiting a website. Query sets are customised by category and are not reused as generic templates.
Queries are run across relevant AI systems such as ChatGPT, Claude, Gemini, and Perplexity depending on the category and scope. Results are captured in a consistent format so outputs can be compared across platforms.
For each response, we track whether your brand appears, which competitors appear instead, and what role each brand plays in the answer. This helps separate meaningful recommendation from passing mention.
The final analysis connects visibility gaps to practical changes across positioning, content, authority signals, and technical structure. The goal is not just to show where your brand is missing, but what can be improved.
Three metrics anchor the analysis. Each one answers a different question about how your brand is being treated by AI recommendation systems.
How often your brand appears across the tested query set.
Why it mattersThis shows whether your brand is consistently being considered or only appearing occasionally.
Which types of queries your brand appears in — and which ones it misses.
Why it mattersThis shows whether your visibility is concentrated in narrow contexts or spread across the buying journey.
Which brands appear when yours does not.
Why it mattersThis shows who is capturing consideration in the AI-generated answers where your brand is absent.
A brand can appear in an AI response in different ways. It may be the primary recommendation, one option in a shortlist, a comparison point, or a passing mention.
In the report, we separate meaningful inclusion from weak mentions where possible. This makes the analysis more useful than a simple mention count.
The brand is positioned as a leading or best-fit option.
The brand appears as one of several recommended options.
The brand is referenced, but not clearly recommended.
The goal is to understand whether your brand is actually being considered, not just whether its name appears.
Running queries is only the collection step. The real value comes from comparing patterns across responses.
We look for differences between your brand and the brands that appear more often, including:
These patterns help explain why competitors may be selected more often — and where your brand has room to improve.
AI responses can change. Results may vary by platform, timing, location, and query phrasing. The experiment is not designed to predict every possible AI answer or guarantee a specific recommendation outcome.
It is designed to create a controlled, repeatable snapshot of how your brand is being selected across the queries that matter most.
If your team is investing in content, SEO, PR, or onsite experience, AI visibility is becoming part of the same discovery system. The first step is understanding where your brand is already being included — and where competitors are taking the slot.