| Entry price and pricing unit | Four flat monthly plans: Lite $29 (25 prompts), Starter $79 (100 prompts), Growth $199 (200 prompts), Pro $499 (500 prompts). The plan sets prompts and engines, and extra prompts are $89 per 100. Unlimited seats on every plan. An Agency and Enterprise plan covers many clients or brands, priced on your client list. See our pricing | Lite $29, Standard $189 and Premium $489 a month, with limits of 15, 100 and 400 prompts. Enterprise starts at 1,000 prompts. Extra prompts are sold in packs of 100. A 15% discount is listed for annual billing. Price scales with prompts, plus add-ons for extra engines. |
|---|
| Engine coverage | 5 engines: ChatGPT (web), Perplexity, Google AI Overviews, Google AI Mode, Gemini. Lite has 3, Starter 4, Growth and Pro 5. No engine add-ons. See our pricing | Every plan includes ChatGPT, Google AI Overviews, Perplexity and Microsoft Copilot. Google AI Mode, Gemini and Claude are paid add-ons on top of the plan price. |
|---|
| Free trial | A new account gets one free run: 10 prompts on ChatGPT, Gemini, Google AI Overviews and AI Mode, once. Plans bill monthly. See our pricing | Yes. Otterly offers a free trial with no credit card. Its own pages give 7 days in the FAQ and 14 days in the sign-up pop-up, so check Otterly for the length. |
|---|
| Geography and language | The United States in English by default. French- and Spanish-speaking countries on request. Ask about your markets | Otterly lists multi-country support for 50+ countries on all plans. |
|---|
| MCP | Built to run from the chat. Connect Claude, ChatGPT, Cursor or any other MCP client, and any member can set up a brand through a guided interview, read every result and change prompts, competitors and groups. 31 tools: 17 read your results, 14 set up and change your tracking. Every change waits for your approval, it starts only a brand's first run or one retry after a failed run, and nothing touches billing. How our MCP works | MCP server with OAuth that works with Claude and Cursor, and it includes write tools that create prompts, brand reports and tags. MCP is listed on Standard and above, at 2,000 calls a month on Standard and 5,000 on Premium. It is not listed on Lite. |
|---|
| API access | A read-only REST API on every paid plan. Any member creates keys in the app. It returns brands, results, prompts, runs and each prompt's stored answers in full. Read our API docs | REST API on Standard (2,000 requests a month) and Premium (5,000). It is not listed on Lite. |
|---|
| Reporting | Named and linked counts, rank and order of mention, sentiment of the answers that name you, head-to-head with each competitor, gaps (answers that name a competitor and not you), cited sources and source gaps, ChatGPT fan-out queries, Digital PR placements, prompt groups, and branded or unbranded views. Your own profiles and listings count as linked. See the product | Lists reports and export on every plan, GEO URL audits of 1,000 to 10,000 URLs a month by plan, and a Looker Studio connector from Standard. |
|---|
| Answer-level data and export | Full answers and sources are viewable in the app for 13 months, and counts and verdicts are kept for the life of the account. The API returns stored answers in full. CSV export covers prompts, engines, competitors and sources, without full answers. MCP returns snippets, and one answer's full text per call on request. Read our method | An MCP tool returns the raw model content, brand mentions and citations for each AI response. The pricing page lists reports and export on every plan, and a Looker Studio connector from Standard. |
|---|
| Methodology transparency | The method page is public. Each answer is scored by a model with a second model checking borderline cases. Percentages need at least 20 prompts, competitors named fewer than 10 times show as "fewer than 10", and a movement is reported only after three runs and a 5-point change. Single answers vary, and counts are reported as measured. Read our method | Otterly says it collects AI results the same way a real user would. We did not find a sample-size or runs-per-prompt statement. |
|---|