Amplitude AI Visibility review (2026): free inside Amplitude, two engines only
Amplitude AI Visibility is included on every Amplitude plan, Free included, and tracks ChatGPT and Google AI Overview. Limits are unpublished. Where it fits and where Promptwatch goes further.
Amplitude AI Visibility is the AI search module inside Amplitude, the product and web analytics platform. It reports a visibility score, share of voice and competitor rankings across a set of prompts, shows the sources LLMs cite, flags sentiment and suggests page-level fixes.
It costs nothing extra. It is included on every Amplitude plan, including Free. The catch sits in Amplitude's own FAQ: it tracks ChatGPT and Google AI Overview only, and you cannot switch models.
What you get per plan
Read on amplitude.com/pricing, amplitude.com/ai-visibility and the AI Visibility FAQ in Amplitude's docs on 20 October 2026.
| Amplitude plan | Price | AI Visibility |
|---|---|---|
| Free | $0, 2M events a month, no card | Included, limited capabilities |
| Plus | Starts at $0, scales with event volume | Included, higher limits |
| Growth / Enterprise | Custom | Limits set by plan and account configuration |
| Non-customers | Free | Limited one-off report |
Amplitude publishes no AI Visibility limits. The FAQ says prompt capacity, refresh cadence and some capabilities depend on your plan and account configuration, and that the tool notifies you when you hit your prompt capacity. Seats are unlimited and prompt metrics export to CSV. An optional search volume per prompt is based on 12-month average Google keyword demand.
What works
The connection to the rest of Amplitude is the reason to use it. AI-referred visitors become a segment in analytics. You can watch their sessions in Session Replay, run a Web Experimentation test on the pages they land on, and sync them as cohorts. Amplitude also lets you chart the path from AI mention to purchase through its Global Agent or MCP. Standalone trackers rarely reach that far into behavior data.
The FAQ is candid. Prompts are generated by asking LLMs what is relevant to your brand, because OpenAI and Google do not share real search data. An identical prompt counts once across models. LLM referrals are identified by referrer or utm_source, and since Amplitude cannot see the prompt that sent a visitor, it points you to entry pages instead.
Prompts and responses can be configured in any language. The page-level recommendations (clearer headings, shorter paragraphs, FAQ blocks) are basic but concrete, and you can simulate content changes.
What to check before relying on it
Two engines. ChatGPT and Google AI Overview cover a lot of traffic, but Gemini, Claude, Perplexity, Copilot and AI Mode are out, and the FAQ says you cannot change the LLM.
Unpublished limits. You have to ask your account team for prompt capacity and refresh cadence.
Paused refreshes. Scheduled refreshes can pause when nobody opens a brand report for a while. A tool that stops measuring when you stop looking is hard to run as a program.
A page contradiction. The landing page's meta description mentions Claude while the FAQ on the same page lists only ChatGPT and Google AI Overview.
Generated prompts. Amplitude states plainly that prompts are LLM-generated, not drawn from real user queries.
Amplitude's FAQ also compares the module favorably with Profound, Evertune, Semrush and Ahrefs. Read that as vendor positioning.
One naming trap. Amplitude's pricing page has an "Agent Analytics" line. It covers AI agents inside your own product, not AI crawlers on your website.
Where it falls short of a full platform
The pages we read do not describe AI crawler logs, Reddit or YouTube citation views, or content publishing to a CMS. So Amplitude can tell you an AI visitor converted. It cannot tell you whether ChatGPT's crawler read the page before the answer cited it, or which Reddit thread the answer leaned on.
That makes it a strong outcome tool and a narrow measurement tool.
When Promptwatch is the better buy
Keep Amplitude for behavior. Add Promptwatch for measurement once two engines stop being enough.
Essential at $95 a month runs 50 prompts across all tracked LLMs with 6,000 responses and daily refresh: ChatGPT, Gemini, Claude, Perplexity, Grok, Llama, DeepSeek, Mistral and Copilot, plus Google AI Overviews and AI Mode. Limits are published, and refresh does not depend on someone opening a report. Prompt tracking includes search volumes, difficulty scores and query fan-outs, and you can tag prompts by persona and topic.
Professional at $245 a month adds the crawler layer, which is Promptwatch's own Agent Analytics: real-time logs for ChatGPTBot, ClaudeBot, PerplexityBot, GoogleOther and the Meta AI crawler, with a crawl-to-citation path and error tracking. Citation analytics covers page, domain, Reddit, YouTube and offsite mentions. Content Agents publish to Webflow, Framer or WordPress.
Promptwatch has its own visitor analytics with conversions, installed by script or GTM template. For an Amplitude shop that overlaps with what you already have; Amplitude's replay and experimentation go deeper on behavior. The reason to add Promptwatch is the measurement, not the funnel.
Verdict
Rank: easy yes for Amplitude customers, weak as a standalone program. Free, connected to revenue data, honest FAQ. Two engines, unpublished limits and pausing refreshes keep it from being the system of record. Run the measurement on Promptwatch Professional and keep the behavior analysis in Amplitude.
FAQ
Which AI engines does Amplitude AI Visibility track?
ChatGPT and Google AI Overview. A model filter on the Prompts tab switches between them, and you cannot change the LLM.
How much does it cost?
It is included on every Amplitude plan, Free included. Capacity and cadence depend on your plan and are not published.
Is Amplitude's Agent Analytics the same as Promptwatch's?
No. Amplitude's covers AI agents inside your own product. Promptwatch's Agent Analytics logs AI crawlers on your site. More at promptwatch.com.