Proofsource

Verified

Summary

Proofsource is an AI visibility platform that puts approved buyer questions to AI answer engines, stores every answer with its sources and a screenshot, and ranks the fixes behind each lost answer. Pricing is per brand and per question.

From: $9/month$7.50 billed annually, 17% less
Open source: No
API: Yes
Category: AI visibility monitoring

Captured 20 October 2026 · citedindex.com

Visit Proofsource Compare Find alternatives

What it does

Monitors AI answers: YesCore of the product: approved buyer questions are asked repeatedly and each answer is stored, scored for mention rate, share of voice and position per engine.Proofsource is an AI intelligence platform that shows whether ChatGPT, Perplexity, Claude and Google AI Overviews name your brand, who they name…proofsource.co
Optimises your content: YesIt ranks gaps behind lost answers by likely gain against effort, names the page or source to change, and runs technical, SEO and AI-readiness checks on existing pages.It ranks the gaps behind each lost answer by likely lift against effort, names the page or source to change, and drafts the fix from your own content…proofsource.co
Generates content: YesDrafted pages and answers are produced from the brand's own facts for human approval, and the feature table lists them as included on the free trial and every paid brand.Drafted contentPages and answers drafted from your facts, approved by a person.proofsource.co/pricing
Publishes to your site: PartlySelf-serve publishing reaches WordPress only, with preview and rollback; Webflow, Vercel and other platforms are shipped by the vendor's team on Managed instead.Publish a drafted fix to your WordPress site. You see the exact preview first, confirm it, and can roll it back.proofsource.co/integrations
Technical AEO: PartlyCrawler access, llms.txt and structured-data checks are measured on every plan, but robots, schema and rendering changes are only made for you on the Managed tier.AI search access (weight 10). Area score from the site audit. Whether AI crawlers may read the site and llms.txt exists.proofsource.co/methodology
Done-for-you service: PartlyA done-for-you Managed tier exists in which the vendor's own team runs tracking, fixes, outreach and reporting, priced by agreement on a demo call rather than on the price card.Our in-house team runs AI visibility for your brand end to end: the questions and tracking, the gap analysis, the content and page fixes written and…proofsource.co/pricing

Answer engines

ChatGPT: YesChatGPT is on the standard panel for the trial and the price card, read as a logged-out visitor on chatgpt.com with the OpenAI API as a fallback.As a logged-out visitor on chatgpt.com, read through a data provider. If that read fails, the OpenAI API answers instead, and the stored answer…proofsource.co/docs/engines-and-models
Perplexity: YesPerplexity is queried through its own API with web search, is on the four-engine standard panel, and is selectable per question at checkout.Perplexity's own API, which searches the web before it answers.proofsource.co/docs/engines-and-models
Gemini: PartlyGemini is not on the self-serve price card; the vendor says it is switched on only after a demo call or on the Managed tier, priced per question like the panel engines.Google AI Mode, Gemini, Microsoft Copilot, Grok, DeepSeek and Meta AI are enabled on a call (Book a demo), each priced per question like the other…proofsource.co/docs/engines-and-models
Google AI Overviews: YesGoogle AI Overviews is a standard-panel engine: the question is searched on Google and the Overview on the results page is read and stored.The question is searched on Google and the AI Overview on the results page is read, through a search data provider with a browser read as the…proofsource.co/docs/engines-and-models
Google AI Mode: PartlyGoogle AI Mode sits outside the four engines a buyer can pick at checkout and is enabled only on a sales call or on the Managed tier, at a per-question price.Google AI Mode, Gemini, Microsoft Copilot, Grok, DeepSeek and Meta AI are enabled on a call (Book a demo), each priced per question like the other…proofsource.co/docs/engines-and-models
Claude: YesClaude is on the standard panel, asked through Anthropic's API with web search enabled, and the vendor notes it costs more per question because it reads more.Anthropic's API with the web search tool switched on.proofsource.co/docs/engines-and-models
Microsoft Copilot: PartlyMicrosoft Copilot is one of the six engines outside the self-serve checkout; it requires a demo call or the Managed tier before it is tracked.Google AI Mode, Gemini, Microsoft Copilot, Grok, DeepSeek and Meta AI are enabled on a call (Book a demo), each priced per question like the other…proofsource.co/docs/engines-and-models
Grok: PartlyGrok is counted among the ten engines the vendor tracks but cannot be selected self-serve; it is turned on after a call or on Managed, priced per question.Google AI Mode, Gemini, Microsoft Copilot, Grok, DeepSeek and Meta AI are enabled on a call (Book a demo), each priced per question like the other…proofsource.co/docs/engines-and-models
DeepSeek: PartlyDeepSeek is supported but gated: the checkout picker offers only four engines, and DeepSeek is enabled on a sales call or on the Managed tier.Google AI Mode, Gemini, Microsoft Copilot, Grok, DeepSeek and Meta AI are enabled on a call (Book a demo), each priced per question like the other…proofsource.co/docs/engines-and-models
Meta AI: PartlyMeta AI appears in the vendor's ten-engine strip but is not on the price card; tracking it needs a demo call or the Managed tier.Google AI Mode, Gemini, Microsoft Copilot, Grok, DeepSeek and Meta AI are enabled on a call (Book a demo), each priced per question like the other…proofsource.co/docs/engines-and-models
Coverage in the vendor’s words: YesThe vendor states ten engines in total but a hard split: four on the trial and on self-serve checkout, the other six only after a call or on Managed.We track 10 AI engines. At checkout you pick from ChatGPT, Perplexity, Google AI Overview and Claude; at least one is always on. Each has its own…proofsource.co/pricing

What it reports

Names the cited sources: YesEvery cited domain and exact page is listed per question, flagged for whether it mentions the brand, with the answer and its sources stored as dated evidence.Citations and sourcesEvery page AI cites, and which of them mention you.proofsource.co/pricing
Competitor tracking: YesRival brands are counted from the answers themselves, including ones never added, with share of voice per engine and per question on one leaderboard.Competitor watchWho AI names instead of you, on every question and engine.proofsource.co/pricing
Brand sentiment: YesA sentiment view reports how each engine describes the brand next to competitors, and the vendor notes it is a three-class agreement figure rather than per-class precision.SentimentHow each engine describes you next to your competitors.proofsource.co/pricing
Recommendations: YesIt goes past reporting: gaps are ranked by expected gain against effort, drafts are produced, and fix check-ups re-measure at days 7, 14 and 28 on paid brands.Ranked fixesEvery gap ranked by likely gain against effort.proofsource.co/pricing
AI crawler analytics: YesAI bot visits are pulled from the customer's Cloudflare zone analytics per page, alongside a check of whether GPTBot and ClaudeBot are allowed to read the site.AI crawler trafficWhich AI bots visit which pages, through Cloudflare.proofsource.co/pricing
Wrong answers about you: YesBrand-named questions are read for wrong or stale claims about prices, features and identity, and each flag carries the page likely to have caused it.Wrong factsWhat AI gets wrong or out of date about you: prices, features, claims.proofsource.co/pricing

Plans and signing up

Prompt or credit allowance: YesQuestions are the billed unit: the calculator runs from 5 to 500 per brand, with volume above 50 discounted, while the trial covers 25 questions and 200 answers.Each brand pays $28 a month for daily checks or $7 for weekly, plus a small price for each question on each AI engine you choose, from $9 a month in…proofsource.co/pricing
Projects or brands: YesSeveral brands are supported and each is priced separately on the same card; the agency workspace adds unlimited client brands with a switcher and per-client logins.Unlimited clients and team members, in one free workspace. Each client brand is priced on the same card.proofsource.co/pricing
Countries: YesThe buyer country and answer language are both chosen by the customer, and each report states which country the answers on screen were asked from.On Check what we found, Asking AI as a buyer in sets the country the questions are asked from, and Answer language sets the language.proofsource.co/docs/engines-and-models
How often it checks: YesCadence is a priced choice per brand: daily or weekly on the card (monthly also selectable), against two fixed scans during the two-day trial.You choose per plan: daily or weekly. AI answers shift from day to day, so daily catches the day a competitor enters your shortlist; weekly and…proofsource.co
Card required to start: NoNo payment card is asked for to start the two-day trial, which the vendor repeats across the site; it is limited instead to one trial per company domain.One free trial per company domain, no card.proofsource.co/pricing
Self-serve signup: YesSign-up, question approval, the price calculator and checkout are all self-serve with no sales call, though six of the ten engines still need a demo call.200 answers. 25 Prompts/Question. Top AI engines. 2 days. Free. No card · no sales callproofsource.co/features/visibility

Access and integrations

API: YesA REST API returning visibility, citations and sentiment as JSON with customer-held keys is listed as included on every column, free trial included.REST APIVisibility, citations and sentiment as JSON, with your own keys.proofsource.co/pricing
MCP server: YesAn MCP endpoint lets an external agent read visibility, competitors, citations and fix status; keys are read-only unless write access is explicitly granted.Let your own AI agent read your visibility, competitors, citations and fix status over the Model Context Protocol. Keys are read-only unless you…proofsource.co/integrations
BI export: PartlyCSV and JSON exports plus the REST API are on every plan, but delivery into Looker Studio or a BI tool is marked not included outside the Managed tier.Your data where you workLooker Studio or BI feeds, and reports in your format.proofsource.co/pricing

Checked 13 October 2026 · How we check · Editorial policy

Plans and pricing

Price and terms

Free to start: PartlyPartly
The terms in full
Two-day trial: 2 scans (today and tomorrow), 25 questions on all four standard-panel engines, 200 answers, plus 100 trial credits for drafts, chat and agents. No card, one trial per company domain; scanning then pauses and stored answers remain.
Card to start: YesNot required
Sign up without sales: YesYes