AI visibility, measured
Your brand,
as AI sees it.
AmpliRank asks AI answer engines the questions your customers ask. See where you appear, who appears instead, which sources the engines trust, and what to change. Every number comes with its range.
- Engines
- Gemini and Google AI Mode, and more as they're enabled
- Per run
- Your questions, web search on
- Every rate
- With its 95% range
- Competitors
- Measured on the same answers
- Method
- Published, footnoted on every figure
Sample dataNorthwind is a fictional brand
Illustrative example data — not real customer results.
Run of 11 November
48 questions2 engines192 answersweb search on
Named, questions that don't mention you1
this run31%
25–38% range · 192 answers · previous 25% (19–32%)
Up 6 points since 4 Nov · not yet a clear change, the ranges overlap
By engine
95% rangeRanges that overlap don't show which engine names you more.
On the same answers
192 answersA gap is called only when its 95% range, paired over the same questions and engines, excludes zero.
What to do next
from this run- 01Answer “best waterproof jacket under $300”Searched in 64% of answers. Fieldstone is named in 9 of 12; your jacket pages never state a price band.Plan page
- 02Get reviewed on outdoor-reviews.exampleCited in 41% of the answers that leave you out.See answers
- 03Ask “how to reproof a jacket” directlyEngines search it on their own; asking it gives it a trend.Add question
1 Sample data for Northwind, a fictional brand. Each question is asked through the engine's API with web search on; answers can differ from what a signed-in person sees. 95% Wilson intervals.
Method as proof
Numbers you can check.
31% · 25–38%
Example, from the sample run
Every rate comes with its range and the number of answers behind it, so you can tell a real change from noise.
Same answers
Competitors are measured on exactly the answers you are, never on a separate sample.
Published method
How questions are asked, how ranges are computed, and what the numbers can't tell you.
Why it matters
Your customers ask AI before they ask you. Most brands find out what it says by accident.
Today
- Questions typed into chatbots one at a time
- Screenshots pasted into a deck
- One answer taken as the truth
- No idea which sources the engine read
With AmpliRank
- A fixed set of questions, asked on a schedule
- Every engine on the same footing
- Ranges that show when a change is real
- The sources, and your own pages, side by side
How it works
From question to page, in six steps.
Questions
48 in your sets4 setsdrafted from northwind.example
| Question | Set | Source | Intent |
|---|---|---|---|
| best waterproof hiking jacket under $300 | Price-band questions | Drafted from your site | Buying |
| packable rain shell women | Price-band questions | Drafted from your site | Buying |
| gore-tex vs eVent shell | Comparisons | Yours | Comparing |
| lightweight rain jacket for travel | Travel | Drafted from your site | Buying |
| how to reproof a gore-tex jacket | Care guides | Searched by engines | Learning |
Named, questions that don't mention you1
this run31%
25–38% range · 192 answers · previous 25% (19–32%)
Up 6 points since 4 Nov · not yet a clear change, the ranges overlap
By engine
95% rangeRanges that overlap don't show which engine names you more.
On the same answers
192 answersA gap is called only when its 95% range, paired over the same questions and engines, excludes zero.
| Question the engines searched | Searched in | You | Most named | Status |
|---|---|---|---|---|
| best waterproof hiking jacket under $300 | 64%123 of 192 | 3 of 12 | Fieldstone · 9 | Opportunity |
| packable rain shell women | 27%52 of 192 | 1 of 5 | Arcadia · 3 | Opportunity |
| how to reproof a gore-tex jacket | 22%42 of 192 | 0 of 4 | Fieldstone · 2 | No page answers it |
| gore-tex vs eVent shell | 16%31 of 192 | 0 of 3 | Lumen Gear · 2 | Opportunity |
| lightweight rain jacket for travel | 41%79 of 192 | 6 of 8 | Northwind · 6 | You lead |
Sources that leave you out
cited inWhat the engines searched
this run38 questions the engines searched on their own while answering yours. The ones searched most, where competitors are named and you aren't, are marked as opportunities.
| Question | Your best page | Coverage | Next |
|---|---|---|---|
| best waterproof hiking jacket under $300 | /jackets/storm-shell | Partly | States no price band |
| how to reproof a gore-tex jacket | None found | No page | Write a care guide |
| packable rain shell women | /womens/rain | Partly | Add packed size and weight |
| lightweight rain jacket for travel | /jackets/travel-lite | Answers it | Nothing to do |
Each step shows the real screen, on sample data for Northwind, a fictional brand.
Who it's for
For anyone whose customers ask AI.
Brand and marketing leaders
One honest number for how AI presents you, and what's behind it.
Content and SEO teams
The questions you lose, the sources that leave you out, the page to write.
Agencies
Every client in one place, and a brief per client each week.
Founders
Know whether AI recommends you before your customers decide.
Trust
Built to be trusted with your data.
Isolated workspaces
Each workspace's data is isolated, and every operation is tested against cross-workspace access.
Sign-in without passwords
A single-use email code or Google. Codes are stored hashed.
Scoped API keys
Every API key belongs to one workspace and is stored as a hash.
Hosted on Cloudflare
The app, its database and its email run on Cloudflare.
Transparent pricing
Priced in answers you can count.
One answer is one question asked to one engine. Plans list how many answers a month they include, and the calculator shows what a run of your size uses.
Free
$0
Your questions, the AI crawler check and the method. Runs start on a paid plan.
Starter
500
answers a month
$29 a month · pending product approval
Growth
2,000
answers a month
$79 a month · pending product approval
Agency
10,000
answers a month
$199 a month · pending product approval
Questions
Asked often.
- How does AmpliRank get its answers?
- Through each engine's API with web search switched on, the same way for every question. API samples are not a complete view of what every person sees: answers can differ from a signed-in session in the app, which is why every rate comes with a range.
- Which engines?
- Gemini and Google AI Mode today. Other supported engines switch on as they're enabled for your workspace.
- How accurate is it?
- Every rate is a sampled estimate, not a guarantee, so it comes with its 95% range and the number of answers behind it. Around 200 answers give a range of roughly 6 points either side of a 30% rate; more samples per question narrow it.
- What's free?
- Your questions, the AI crawler check and the method. Runs start on a paid plan, because every answer is a paid call to an engine.
- Do you guarantee better rankings?
- No. AmpliRank shows what changed after you act. It never claims the change was caused by your work.