Know exactly who the AI recommends instead of you.
When someone asks ChatGPT or Perplexity for the best option in your category, a handful of brands get named — and the rest are invisible. cloro measures share of voice, ranks the domains every engine cites, and surfaces the precise questions where you don't appear.
Share of voice · Perplexity
Four numbers that decide whether the AI is on your side
Run one keyword set across every engine and cloro rolls the raw answers into the metrics that actually move a strategy.
Every brand mention, counted the same way across eight engines.
cloro reads the full answer, resolves brand names and aliases to a single entity, and tallies who gets recommended. No manual tagging, no engine-by-engine reformatting.
- Alias resolution folds "Acme", "Acme Inc." and "acme.com" into one brand
- First-position weighting separates "mentioned" from "recommended first"
- Per-engine breakdown shows where you're strong and where you leak
- Sentiment on each mention tells you if the naming even helps you
Overall share of voice
Aggregated across ChatGPT, Gemini, Perplexity, Copilot, Grok, AI Overview and AI Mode for a 120-prompt category set.
The domains the engines trust to answer your category
Every recommendation is backed by sources. cloro ranks the domains cited across your prompt set so you can see who's shaping the answer — and earn a place among them.
Counts are total citations across the prompt set. Illustrative structure — your scan returns your own domains and numbers.
The questions where rivals show up and you're absent
Aggregate share is useful, but the wins hide in individual prompts. cloro flags every question where competitors are named and your brand isn't — your optimization backlog, ranked by opportunity.
# score a category and rank the competition curl https://api.cloro.cloud/v1/competitors \ -H "Authorization: Bearer $CLORO_KEY" \ -d '{ "brand": "yourbrand.com", "engines": ["chatgpt","perplexity","gemini"], "prompts": ["best pm software for agencies", "cheapest tool for a 5-person team"], "format": "json" }' # → response { "share_of_voice": { "yourbrand.com": 0.29, "rivalalpha.com": 0.25, "rivalbeta.com": 0.18 }, "first_position_rate": 0.14, "top_cited_domains": [ { "domain": "g2.com", "citations": 218 }, { "domain": "reddit.com", "citations": 183 } ], "gaps": [ { "prompt": "cheapest tool for a 5-person team", "competitors_named": ["rivalalpha.com"], "you_present": false } ] }
A full competitive read from a single call.
Pass your brand, the engines you care about and a prompt set. cloro fans out, reads every answer, resolves the brands and hands back share of voice, citation rankings and gaps — already structured.
- Batch hundreds of prompts across engines in parallel
- Stable keys so week-over-week diffs are trivial
- JSON for dashboards, Markdown for reports and LLM grounding
- Billed per successful call — empty results aren't charged
How teams turn a competitor scan into moves
Define the battleground
List the prompts a real buyer would ask in your category and the rivals you want scored. cloro handles alias resolution, so you don't have to enumerate every spelling.
Fan out across engines
One call runs the set through every AI answer engine and Google. Rendering, rotation and retries happen upstream — you just receive parsed records.
Read share, ranks and gaps
Share of voice tells you the overall picture; the domain leaderboard shows who's shaping it; the gap list is your ranked backlog of prompts to win.
Re-run and diff
Schedule the same set weekly and diff on stable keys to prove whether your content and PR work is actually moving you up in the answers.
Win the prompts you're losing
Turn the gap list into a content and PR backlog, ordered by how often each question gets asked.
Earn the citations that matter
See which domains the engines lean on and pursue placements where a mention actually changes the answer.
For anyone keeping score against the competition in AI answers
- GEO & SEO teams — prove your share of voice is climbing, not your rivals'
- Brand & PR — track which competitors the engines name and cite instead of you
- Product marketing — sharpen positioning around the gaps rivals are filling
- Agencies — deliver a competitive AI read across a whole client roster from one key
Competitor analysis, answered
How does cloro decide who counts as a competitor?
You pass the brands you care about, and cloro also detects any brand the engines name across your prompt set — so rivals you hadn't listed still surface. Alias resolution folds different spellings, legal suffixes and domains into a single entity before anything is counted.
How is share of voice calculated?
For each prompt, cloro reads the full answer, resolves every brand mention, and tallies them. Share of voice is your mentions as a fraction of all branded mentions, reported per engine and overall. First-position rate is tracked separately so "mentioned somewhere" isn't confused with "recommended first".
Where do the cited domains come from?
Answer engines back their recommendations with sources. cloro extracts every citation across your prompt set and ranks the domains by how often they appear, so you can see which sites are shaping the answers in your category.
What exactly is an "answer gap"?
A prompt where one or more competitors are named and your brand isn't — or is mentioned last with no citation. Each gap includes the prompt, which rivals appeared and on which engines, so it reads as a ranked backlog of opportunities.
Can I track this over time?
Yes. Re-run the same prompt set on a schedule and diff on stable keys to watch share of voice, rankings and gaps shift week over week — the cleanest way to prove optimization work is moving you up in AI answers.
How is it billed?
Per successful call. Failed or empty results aren't charged, and your live usage counter reflects exactly what you've consumed.
See who the AI recommends — before your next campaign.
Spin up a key, scan your category across every engine, and turn share of voice, citations and gaps into a plan you can act on.