AI visibility tracking

Know exactly how the answer engines talk about your brand.

Rankings told you where you sat in ten blue links. But when ChatGPT, Gemini and Perplexity write the answer, the only question that matters is whether they name you at all. cloro measures your mention rate, citation rate and share of voice across every engine — and tracks how they move week over week.

● 5 answer engines measured ● Mention · citation · share of voice ● Tracked over time

Share of voice — "project management software"

last 7 days
Your brand31.4% ▲ 4.2
Competitor A26.8% ▼ 1.1
Competitor B21.0% ▲ 0.4
Competitor C20.8% ▼ 3.5
Prompts sampled 240 Engines 5 Your mention rate 68%
ChatGPT Gemini Perplexity Copilot Grok AI Overview AI Mode
What we measure

Three numbers that define AI visibility

Being visible in an answer engine isn't one thing. cloro breaks it into three measurable signals, computed the same way across every engine so the numbers actually compare.

Signal 01

Mention rate

Out of every prompt where your category comes up, how often does the engine name your brand at all? This is the base layer of presence — the answer to "do they even know I exist?"

mentions ÷ prompts sampled
Signal 02

Citation rate

When the engine backs its answer with sources, how often is one of them your domain? A mention is reputation; a citation is a click-path and a trust signal the engine is passing along.

answers citing you ÷ answers with sources
Signal 03

Share of voice

Across the whole competitive set, what fraction of all brand mentions are yours? Share of voice turns raw counts into a single, rank-able position against the rivals the engines actually recommend.

your mentions ÷ all brand mentions
Why it matters

The answer is the new front page.

A growing share of buying research never touches a results page. People ask an assistant, read one synthesized paragraph, and act. If your brand isn't inside that paragraph, you were never in the room — and no rank-tracking dashboard would ever have told you.

  • Mentions and citations live inside prose, not in a ranked list — you can't parse them with a classic SERP tool
  • Every engine phrases the same answer differently, so a single spot-check tells you nothing
  • Answers drift weekly as models and their sources change; visibility is a moving target
  • Your competitors are being recommended in answers you never see
ChatGPT74%
Perplexity61%
Gemini55%
Copilot49%
Grok38%

Illustrative mention rate for one brand — the same prompt set, measured per engine. Real gaps this wide are common.

Multi-engine, side by side

One brand looks different in every model

A brand that dominates ChatGPT can be almost invisible in Grok. cloro runs the identical prompt set through every engine so you see the whole picture — and know precisely where to spend your optimization effort.

⚖️

Normalized, comparable scores

Mention, citation and share-of-voice are computed identically per engine, so a 60% in Gemini means the same thing as a 60% in Perplexity.

📉

Drift detection over time

Re-run your prompt set on a schedule and cloro stores each snapshot. Diff any two periods to see exactly which engine moved, and by how much.

🧭

Prompt-level breakdown

Drill from a headline share-of-voice number down to the individual prompts where you were named — or conspicuously left out.

POST /v1/visibility
# measure a brand across every answer engine
curl https://api.cloro.cloud/v1/visibility \
  -H "Authorization: Bearer $CLORO_KEY" \
  -d '{ "brand": "Northwind",
        "prompts": ["best project management software",
                    "tools for remote team planning"],
        "engines": ["chatgpt","gemini","perplexity","grok"],
        "period": "7d" }'

# → response
{
  "brand": "Northwind",
  "share_of_voice": 0.314,
  "by_engine": {
    "chatgpt":    { "mention_rate": 0.74, "citation_rate": 0.41 },
    "gemini":     { "mention_rate": 0.55, "citation_rate": 0.29 },
    "perplexity": { "mention_rate": 0.61, "citation_rate": 0.52 },
    "grok":       { "mention_rate": 0.38, "citation_rate": 0.17 }
  },
  "competitors": [
    { "name": "Competitor A", "share_of_voice": 0.268 },
    { "name": "Competitor B", "share_of_voice": 0.210 }
  ],
  "delta_vs_prev": 0.042
}
One call, every engine

Visibility as a single JSON object

Pass a brand, a prompt set and the engines you care about. cloro runs the answers, resolves brand mentions and citations, and returns computed visibility metrics — ready to store, chart or alert on. No parsing prose yourself.

  • Per-engine mention and citation rates in one response
  • Competitor share of voice resolved from the same answers
  • Period-over-period delta so change is one field away
  • Markdown output too, when you want the raw answers as evidence
3visibility signals per brand
5answer engines compared
1API call per snapshot
snapshots kept for trends
A tracking loop

How a visibility program runs on cloro

Step 01 · once

Define your brand set and prompts

List your brand, your competitors and the buyer questions where you want to show up. This prompt set becomes the fixed yardstick every snapshot is measured against.

Step 02 · each run

cloro queries every engine

On demand or on a schedule, the same prompts fan out to ChatGPT, Gemini, Perplexity, Copilot and Grok. Rendering, rotation and retries are handled upstream.

Step 03 · each run

Mentions and citations get resolved

Every answer is scanned for brand names and source domains, then rolled up into mention rate, citation rate and share of voice — normalized per engine.

Step 04 · over time

Snapshots stack into a trend

Each run is stored with its timestamp. Diff any two periods to see which engine moved you up, which dropped you, and whether your optimization work actually landed.

Questions

AI visibility, explained

What's the difference between mention rate and citation rate?

Mention rate is how often an engine names your brand anywhere in its answer — pure presence. Citation rate is how often the engine links your domain as a source when it shows its references. You can be mentioned without being cited, and occasionally cited without being named in the prose, so cloro tracks both separately.

How is share of voice calculated?

For a given prompt set and competitive set, share of voice is your total brand mentions divided by all brand mentions across the group. It turns raw counts into a single percentage you can rank and trend, so "we're at 31%, up 4 points" is a sentence a stakeholder understands.

Which engines can you measure?

ChatGPT, Gemini, Perplexity, Copilot and Grok on the answer-engine side, plus Google AI Overview and AI Mode. The identical prompt set runs through each one, so per-engine scores are directly comparable.

How do I track visibility over time?

Re-run the same prompt set on a schedule. cloro stores every run as a timestamped snapshot, and the API returns a period-over-period delta out of the box, so charting a trend or firing an alert when your share of voice drops takes one field, not a data pipeline.

Do I need my own model accounts or proxies?

No. You authenticate with a single cloro API key. Fetching the answers, rendering, rotation and retries all happen on our side, and you're billed only for successful calls.

Find out what the AI actually says about you.

Point one API at every answer engine and turn mentions, citations and share of voice into numbers you can track from week to week.