AI Brand Perception API: how AI engines frame each brand, by theme
The AI Brand Perception API groups AI answers into themes, such as pricing or proofing workflow, and returns how each brand is characterised inside every theme. It shows where engines frame a brand as the leader, the cheap option or the hard one to use, theme by theme and engine by engine.
Updated
- ChatGPT
- Perplexity
- Google AI Mode
- Claude
- Microsoft Copilot
- Grok
Playground
Proofing workflow takes 31.4% of answers, and Loopwork is framed as its leader
The AI Brand Perception API clusters the answers in scope into themes and labels each brand inside every theme. In this sample on ChatGPT, proofing workflow takes 31.4% of answers and Loopwork leads it, while on pricing, 18.2% of answers, Loopwork reads as the premium option and Cadence as the cheap one. A single sentiment score would hide both.
POST /v1/themes/perception
models[]
brands in the table
- Themes in scope
- 4
- Largest theme
- proofing workflow, 31.4%
curl -X POST https://api.apiforgeo.com/v1/themes/perception \ -H "Authorization: Bearer $API_FOR_GEO_KEY" \ -H "x-project-id: $PROJECT_ID" \ -H "Content-Type: application/json" \ -d '{ "prompt": "best project management tool for creative agencies", "models": [ "chatgpt" ], "country": "US" }'
import os, requests API = "https://api.apiforgeo.com/v1" H = {"Authorization": f"Bearer {os.environ['API_FOR_GEO_KEY']}", "x-project-id": os.environ["PROJECT_ID"]} r = requests.post(f"{API}/themes/perception", headers=H, json={ "prompt": "best project management tool for creative agencies", "models": [ "chatgpt" ], "country": "US" }, timeout=30) print(r.json())
const API = "https://api.apiforgeo.com/v1"; const headers = { Authorization: `Bearer ${process.env.API_FOR_GEO_KEY}`, "x-project-id": process.env.PROJECT_ID, "Content-Type": "application/json", }; const post = (path, body) => fetch(API + path, { method: "POST", headers, body: JSON.stringify(body) }).then((r) => r.json()); console.log(await post("/themes/perception", { "prompt": "best project management tool for creative agencies", "models": [ "chatgpt" ], "country": "US" }));
| Theme | Share | Loopwork | Cadence |
|---|---|---|---|
| proofing workflow | 31.4% | leader | adequate |
| pricing | 18.2% | premium option | cheap option |
| client portal | 16.7% | adequate | leader |
| resource planning | 12.9% | hard to use | adequate |
{
"scope": {
"prompt": "best project management tool for creative agencies",
"models": [
"chatgpt"
],
"country": "US"
},
"results": [
{
"theme": "proofing workflow",
"share": 31.4,
"perception": {
"Loopwork": "leader",
"Cadence": "adequate",
"Northbeam": "hard to use"
}
},
{
"theme": "pricing",
"share": 18.2,
"perception": {
"Loopwork": "premium option",
"Cadence": "cheap option",
"Northbeam": "adequate"
}
},
{
"theme": "client portal",
"share": 16.7,
"perception": {
"Loopwork": "adequate",
"Cadence": "leader",
"Northbeam": "adequate"
}
},
{
"theme": "resource planning",
"share": 12.9,
"perception": {
"Loopwork": "hard to use",
"Cadence": "adequate",
"Northbeam": "leader"
}
}
]
} | Field | Type | Description |
|---|---|---|
theme | string | A topic the answers cluster around, such as pricing. |
share | float | Percentage of in-scope answers that fall into the theme. |
perception | object | How each brand is characterised inside the theme, keyed by brand. |
Get started
Four setup steps lead to the first theme map
The AI Brand Perception API reads the theme layer computed on every enriched run, so the first map arrives after four steps: create the project, generate a key in the app, import the prompts and run them. Reading the themes and labels back is not billed.
-
Step 1
Create a project
Create an organisation and a project, one project per brand or per client.
-
Step 2
Generate a key
Generate an API key in the app; keys are managed there, never through the API.
-
Step 3
Import your prompts
Import your prompts and tags with a single POST /v1/queries.
-
Step 4
Run and read
Trigger an execution, poll the job, then read any analytics endpoint.
Comparison
Themes and per-brand perception come as data in 1 of 4 approaches
The AI Brand Perception API returns theme clusters and per-brand perception on every stored answer. An AI scraping API returns neither, GEO dashboard tools rarely offer them, and an in-house build turns them into a research project: clustering, labelling and keeping the labels stable as engines change their answers.
| Capability | API for GEO | AI scraping API | GEO dashboard tool | In-house build |
|---|---|---|---|---|
| Theme clustering | Yes | No | Rare | Research project |
| Per-brand perception inside a theme | Yes | No | Rare | Research project |
| Sentiment per brand per response | Yes | No | Partly | LLM cost + eval |
Use cases
Perception data feeds 3 deliverables, one per audience
The AI Brand Perception API gives each audience a view a sentiment score cannot: platforms show customers how engines position them per topic, agencies bring positioning findings to client reviews, and brands learn whether they read as the premium or the cheap option in their category.
A positioning tab per topic
Show each customer the themes engines raise in their category and the label their brand gets in each one. A theme map explains a visibility score better than one number.
AgenciesPositioning findings for reviews
Bring the themes where a client reads as hard to use, or as the cheap option, to the quarterly review. Each finding comes with the answers behind it.
E-commerce & brandsPremium or cheap, theme by theme
Per-theme perception tells you whether you are the premium or the cheap option where it matters. Track how that label moves after a price change.
Pricing
AI Brand Perception API pricing runs from $0.013 to $0.007 per credit
The AI Brand Perception API reads themes computed during normal runs, so the theme map itself is not billed. Credits pay for the executions, one per prompt, per model, per locale, and theme clustering and per-brand perception come with that credit like every other layer. The tier follows the month's volume.
Price per credit, set by the month's total volume.
FAQ
AI brand perception, answered
What does one call return?
The AI Brand Perception API returns one object per theme in scope: the theme as the clustering named it, its share of the answers and a perception label for each brand inside it. Labels describe how the answers characterise the brand, such as leader, adequate, premium option or cheap option.
Can I compare perception across engines?
The AI Brand Perception API reads one model at a time, so running it once per model shows whether two engines frame the brand the same way inside a theme. Where they disagree, the citation layer shows which sources each engine relied on.
Can I track how competitors are framed?
The AI Brand Perception API labels every brand named inside a theme, competitors included, so one call shows where Loopwork leads and where Cadence or Northbeam holds the stronger position. The labels stay comparable across brands, so a competitor's strong theme reads the same way as your own.
Is the theme layer billed separately?
The AI Brand Perception API adds no charge to the runs. Theme clustering and per-brand perception are enrichment layers that every credit includes, and reading them afterwards is unmetered, so a monthly positioning review costs only the executions behind it.
Can I resell perception data?
The AI Brand Perception API allows resale and white-label use like every family. The payload carries no branding and each client can sit in an isolated project, so an agency can deliver positioning reviews under its own name and a platform can add a theme map to its product.