One continuous cycle, not a one time scan.
BrandWater AI observes how AI describes your brand, explains why, and recommends the next step. Approval based execution and verification are coming in V3.
From your website to a first result
Ten steps at launch. You give a URL, review what we understood, and only then do we run anything. Search Console and GA4 are optional and come later.
- 1
Add your website
Enter a URL. It is normalised so the same site is never created twice, and country and language are detected but editable.
- 2
We analyse the site
Robots and sitemap are checked, then the site is crawled, up to 100 pages at launch, with live progress.
- 3
We show what we understood
Brand, products, services, audience, locations and use cases are extracted, each with its source page.
- 4
You review and confirm
Edit what is wrong and confirm the core facts. Confirmed context becomes the reference for everything after.
- 5
Topics we found
Eight to fifteen topics come from products, problems and use cases, not from your brand name, each with the reason.
- 6
You confirm topics
Keep, remove or add topics. Only confirmed topics feed prompts.
- 7
Your first discovery set
Fifteen unbranded prompts across core intents, with duplicates removed.
- 8
You activate prompts
Edit, remove or add. Saved, active and executed are separate, so you control what runs.
- 9
First audit
Fifteen prompts on three engines is forty five observations. Failures are explained, never hidden.
- 10
Your Performance view
Visibility, recommendations, citations, competitors and a short list of opportunities, with the evidence behind each.
Engines are observed through provider APIs, which may differ from what a person sees in each consumer app.
Seven steps across three areas
- 1AI Visibility
Observe
performanceYour activated prompts run on Gemini, Google AI Overview, Google AI Mode, ChatGPT and Perplexity, and the full answers are stored, with sources where the provider shows them.
- 2AI Answers
Measure
performanceAnswers become consistent metrics with defined denominators and visible sample sizes.
- 3Omissions
Diagnose
intelligenceAnswers are compared with your confirmed brand facts to explain why a number moved.
- 4Opportunities
Prioritize
agentGaps are ranked by business relevance so you work on the few that matter.
- 5Strategies
Act
agentA strategy is recommended with its evidence. At launch you make the change. Approval based execution is coming in V3.
- 6Verification
Verify
agentComing in V3: the changed page is checked and only the affected prompts are rerun. At launch, your next audit shows what changed.
- 7Experiments and Impact
Learn
intelligenceComing in V3: measured outcomes feed back into prioritisation so each cycle starts smarter.
From domain to first audit
No crawler to configure, and nothing to integrate before you see a result.
Add your website
BrandWater crawls your domain and drafts your Brand Intelligence automatically.
Confirm your prompts
Review the topics and questions customers would ask AI about you.
See your first audit
Visibility, mentions, competitors and gaps land in Performance, ready to inspect.
What stays in your control
- You review every fact in Brand Intelligence before it is used.
- You choose the competitors, topics, prompts and platforms.
- At launch the Agent only recommends. From V3, every action needs your approval.
- BrandWater never edits your website on its own.
- Each brand's data stays separate from every other brand's.
Honest by design. BrandWater observes the answers each engine returns through its provider API, which can differ from the consumer app. It has no private access to any model, and it reports association, not proof of cause.
Ready to see your first audit?
Start with a first audit. We show where you appear, who is named instead, and what to fix first.