AI Visibility Report, where you stand today

You are barely visible to AI, in a category where buyers increasingly ask AI first.

Across the four assistants buyers use to find a plumber, Brightway is named in about 6% of relevant answers and ranks #11 of 14 tracked local competitors. The good news: the gaps are specific and fixable. This report measures the baseline, names the factors holding the brand back, and lays out the sequence to close them.

We ask AI the questions your buyers ask, and record who it names.

Visibility is measured by issuing realistic buyer prompts to each model and recording which businesses appear in the generated answers. We don't scrape rankings or read traditional search position, we observe what the assistant actually tells a prospective customer. Every prompt is run several times per model and location so a score reflects the share of runs that named the brand, which lets us attach a confidence interval to every number.

Ten signals AI models read before recommending a business.

Each dimension is scored from the audit, cross-model sentiment, and competitive benchmarking. Red is where the brand is materially behind the businesses AI names instead.

Asked about Brightway by name, the models hedge.

None of the assistants warn customers away, but all express the same underlying hesitation: not enough independently verifiable signals. These are the exact concerns the strategy has to resolve.

Where you sit against the businesses AI names instead.

Overall AI visibility across all tracked brands, with bootstrap 95% intervals. A handful of competitors own most of the attention; the middle of the field, where Brightway can realistically move first, is winnable.

Your visibility by model

Per-model figures are noisier than the overall, but they show where to concentrate. Perplexity already names Brightway occasionally; ChatGPT, where many buyers start, almost never does.

Visibility on the questions buyers actually ask

Branded questions show what AI says about you; these show whether it names you at all. Most buyers never type your name, they ask category, "near me" questions, and the model decides who to surface. Brightway rarely makes the list, and almost never on the highest-intent emergency query, the single most valuable moment to be named.

The off-site footprint AI cross-checks before it recommends you.

Models don't just read your site, they look for corroborating signals across review and directory platforms. Brightway is present on Google but thin or missing nearly everywhere else, while the brands AI names are listed and rated almost everywhere. Red marks a gap worth closing.

Which signals actually move the needle, ranked by correlation.

Across the businesses in this benchmark, a handful of signals do almost all the work. Eight features correlate strongly with AI visibility (coefficient ≥ 0.5); below that the link falls off a cliff and the remaining signals don't reliably move anything. The strong drivers are exactly the ones Brightway is missing, which is where the scorecard's red boxes come from. Correlation is association, not proof of cause, but it points the strategy straight at the signals worth fixing first.

Category importance

Each category scored by its strongest reliable signals. The leverage sits in five categories; the rest barely register.

Five gaps that compound each other.

No single weakness explains 6% visibility. Each gap makes the others harder to overcome, which is why the fixes are sequenced rather than parallel.

Four pillars, sequenced for compounding returns.

Sorted by impact-to-effort, so the cheap wins come first.

AI visibility moves on a predictable curve.

What we track every month from here.

Primary metrics: AI brand visibility per model (baseline 6%), review volume and freshness across platforms (baseline thin), and domain authority. Secondary: schema validation, indexed pages, and which pages and citations AI actually surfaces. Each recurring scan re-runs the same fixed prompt set so movement can be separated from run-to-run noise.

Prompts run in this baseline