Documentation
MentionFlow docs
How MentionFlow works, for the people who use it and the people who build on it. Each feature page says what a screen does, how to use it, and exactly how it computes what it shows. The API reference is written by hand from the code that serves it.
QuickstartSet up a brand, pick prompts, and read your first answers. About five minutes.Core conceptsThe words we use: workspace, project, prompt, run, engine, window.API and MCPEvery number in your own tools. Auth, resources, tools, errors, limits.
Browse the docs
69 pages in 10 sections
Getting started
5 pagesWhat MentionFlow is, a five-minute start, the words we use, and how plans and quotas work.
What is MentionFlowMentionFlow shows you how AI assistants talk about your brand, and what to change. Here is what it does and how these docs are laid out.QuickstartGo from a new account to a project that is watching what AI assistants say about your brand, in about five minutes.Core conceptsThe words MentionFlow is built on. Workspace, brand, collection config, prompt, run, receipt, mention, citation, fan-out, carousel, and the engine classes.Date ranges and the top-bar pickerOne date control in the top bar sets the window for every data page. Presets, custom ranges, the compare toggle, and the pages that ignore it on purpose.Plans and quotasWhat each plan includes and how the quotas are enforced. Pooled prompts, engine slots, the seat cap, cadence, and the one action that frees a prompt slot.
Use cases
4 pagesStep-by-step playbooks for agencies, in-house teams, e-commerce brands, and content teams.
For agenciesRun AI visibility for many clients from one workspace. A portfolio rollup, white-label reports on a schedule, client-viewer seats, scoped access, and an audit trail.For in-house marketing teamsOwn your brand's AI answer share the way you own its search share — track the questions that matter, watch competitors take or lose ground, work a scored action board, and ground demand in your own Search Console data.For e-commerce & shopping brandsSee where your products land on the AI shelf. The carousels ChatGPT and Google AI answers show, matched against your own catalog, with imports from Shopify and Google Merchant Center, demand momentum, and honest engine coverage.For content & SEO teamsRun the whole loop. Find the questions AI assistants answer, draft against your own facts, publish, and track whether the page earns citations. Plus source intelligence, crawler analytics, and an llms.txt generator.
Feature guides
20 pagesEvery screen in the app. What it does, how to use it, and exactly how it computes what it shows.
Dashboard overviewThe dashboard home. Five metric cards, the you-vs-competitors chart with its ranking rail and full competitor matrix, an average-position bump chart, collection health, and top movers, all from one rolling window.PromptsThe tracked-prompt portfolio. Per-row window metrics, bulk CSV import, topic folders, tags and stages, and a per-prompt detail view with run diffs.Batch actions and searchWork a long prompt list like a spreadsheet. Search and filter down to a set, select it, change every prompt at once, and show or hide columns.Prompt locationsRun one tracked prompt as if it were asked from a specific country. Available on every plan, across a 241-country catalog, with an honest note on which engines honor it.Suggestion queueA lasting, demand-first queue of suggested prompts from research and discovery, with safe repeat enqueue, safe accept, and a restorable rejected list.Answers and receiptsA receipt is one completed answer run: what one engine said to one prompt on one date, with its ranked mentions, sentiment evidence, and citations.CompetitorsA Share-of-Voice leaderboard, trend, engine heatmap, and answer-gap matrix ranking your brand against the competitors you track, plus auto-discovery of rivals you don't.Sources and citationsThe domains and community threads engines cite when answering your prompts, classified and ranked, with "you appear" and "opportunity" flags and per-domain drill-down.Fan-outsThe real web searches an AI assistant ran while answering your prompts, lifted word for word from the stored payload, ranked by estimated demand and trackable in one click.Ads intelligenceWhich advertisers ChatGPT runs sponsored units for on your tracked prompts, who bids on prompts that name you, and the creatives behind each unit.ActionsA ranked to-do list built from your own metrics, on a board whose cards close themselves when the data says the gap is fixed.ReportsThe reports hub. The briefing that writes this window out in sentences, a print-ready client report (white-label on Agency plans), per-brand scheduled report emails, and the Looker Studio connector.PortfolioThe agency multi-client screen. One row per brand, biggest movers first, each brand measured over its own newest seven collected days.Page trackerWhich of your own pages earn citations in AI answers. Every page from your sitemap crawl plus any URL you track by hand, each with its citation record, Search Console clicks and impressions, DataForSEO authority ranks, and a page category.CrawlersWhich AI agents fetch your site and why, whether your robots.txt lets them in, and whether the pages they crawl go on to earn citations.Site healthYour own site through a crawler's eyes. Every page's status, speed, size, words, text ratio and links, plus the AI-readiness checks (robots.txt, llms.txt, bot walls) that decide whether AI can read you at all.ContentWhere your site covers the questions you track and where it doesn't, plus drafts and briefs written from your own facts and followed for the citations they win.ResearchDemand research that blends real search data with a language model into about 32 candidate prompts, plus a Search Console striking-distance block.Brand and knowledgeYour brand identity plus the knowledge base that grounds AI content and fact-checking: a structured fact sheet and a document library, both read at draft time.Matching rulesControl how answer text is credited to you and your competitors. Competitor aliases, safe custom patterns, per-brand chart colors, and a re-scan that applies new rules to your collected history.
Metrics
7 pagesOne page per metric: the formula in plain words, what it divides by, the window, and how to read it.
Metrics overviewThe six core metrics MentionFlow computes, the rules they all share, and where to read each one.VisibilityA 0 to 100 score for how prominently AI engines mention your brand, weighted so an earlier mention counts for more than a later one.Share of VoiceYour share of the mentions inside your tracked set, a zero-sum split across you and the competitors you track. Brands you don't track are not in the split.SentimentA 0 to 100 index of how positive engines are when they mention you, where 50 is neutral, and why unscored is never counted as neutral.Average positionThe mean rank of your first mention across the answers where you appear. Lower is better, and it is null until you have appeared at least once.Citation shareOf the answers that cite any source, the share that cite your own domain. A 0 to 1 measure of how often you are the receipt behind an AI answer.Presence rateThe share of sampled answers in which your brand appears at all. The simplest visibility metric, a 0 to 1 fraction with no position weighting.
Shopping
8 pagesYour product catalog, imports (Shopify, Google Merchant Center, CSV), the product page, and shopping analytics.
Shopping overviewHow MentionFlow tracks the product carousels AI engines show on your prompts, and matches them against your own catalog.Product catalogYour uploaded products, matched to the AI carousel placements engines showed, with per-product micro-bars, category filtering, and safe deletes.Product importsThree ways to add products to your catalog: CSV, a Shopify public products.json, or a Google Merchant Center feed. All preview first, and all are safe to re-run.Product detailOne product's shelf performance across the AI shopping carousels it appeared in. Win rate, visibility share, average slot, price versus shelf, and a trend.Shopping analyticsThe shopping home. Summary tiles, a trend, a top-30 product leaderboard, the most recent carousels, and a CSV export of your AI shelf presence.Product attributesHow AI shopping answers describe your product, in their own words, next to the brands winning the most shelf space. Extraction is rolling out.Shopping demandThe searches assistants run while building product carousels, ranked Top, Trending, Losing, and New so you can see which shopping demand is rising or fading.Engine coverageWhich AI engines put products on your shelf today, which are wired and waiting on vendor data, and how to read the coverage strip.
Data honesty
1 pageHow MentionFlow handles missing data: the dash rule, sample labels, and confidence tiers.
Integrations
6 pagesConnect Google Search Console, Shopify, Google Merchant Center, Looker Studio, and CSV feeds.
Google Search ConsoleConnect Search Console to back MentionFlow's demand estimates with your real search data, and to see which of your own pages AI assistants cite.Shopify importPull your Shopify catalog into MentionFlow so you can see whether AI shopping assistants recommend your actual products. No app install, no OAuth.Google Merchant CenterLoad your product catalog from a Google Merchant Center feed, so AI shopping placements can be matched to your real products. You preview before anything is written.CSV importsLoad your questions or your products from a spreadsheet. Two CSV importers that preview every row, re-check it on the server, and are safe to re-run.Looker StudioBuild your own dashboards on top of your MentionFlow data with the Looker Studio community connector, linked from the Reports hub.Crawler ingestGet your AI-crawler logs into MentionFlow so you can see which AI bots visit your site. A one-click Cloudflare Worker, a log paste or upload, or a continuous feed.
API & MCP
9 pagesThe v1 REST API and the MCP server: auth, conventions, every resource and tool, errors, and limits.
API and MCP overviewMentionFlow has a read-only v1 REST API and an MCP server over the same data your dashboard shows. Here is what they cover and what they promise.AuthenticationEvery request carries a workspace API key as a bearer token. Here are the key format, the kinds of key, the two scopes, and how scoping works.ConventionsThe rules every response follows. Scales, null handling, note fields, the fixed window, engine blending, and the contract-freeze policy.LimitsThe per-key request rate limit (120/min, X-RateLimit-* headers, 429 + Retry-After), the response caps, the MCP batch cap, and the ingest body cap.ErrorsThe JSON error shapes and status codes for the REST read API, the crawler-ingest endpoint, and the MCP JSON-RPC transport.Resource referenceEvery v1 REST resource. Request contract, response fields, types, and scales for brands, overview, prompts, competitors, sources, urls, answers, shopping, fact-check, ads, and crawler ingest.MCP serverA Model Context Protocol endpoint that exposes MentionFlow's reads as tools an AI agent can call. The setup, the protocol, and all sixteen tools.Webhooks and signature verificationThe cycle-completion webhook, the X-MentionFlow-Signature envelope, and copy-paste receiver code to verify deliveries and reject replays with your workspace signing secret.ChangelogThe API shipped in four steps. What each one added, newest first.
Troubleshooting
2 pagesWhy a monitor shows no data yet, engine coverage, share-of-voice numbers that differ, import and invite errors, API 401 and 429, demo vs live, and a fast FAQ.
Account & settings
7 pagesWorkspace and members, password and email verification, two-factor auth, and API keys.
Workspace and membersSettings in two tabs. Project Settings holds the collection config, notifications, Search Console and tracking. Workspace holds members and roles, API keys, the audit log, and the Danger Zone.Password and email verificationHow password reset by email works, the verification banner and resend, what an unverified account can and can't do, and which accounts are grandfathered.Two-factor authenticationAdd TOTP two-factor authentication with one-time backup codes to your account, available on every plan and verified before it ever turns on.API keysCreate, scope, expire, rotate, and IP-restrict the API keys that authenticate the v1 API, the MCP server, and crawler ingest, shown in plaintext exactly once at creation.Billing and plansManage your subscription through Stripe and see real usage meters that count exactly the way the plan quota guards enforce.Refer and earnShare your workspace's referral code to earn a share of a referred customer's payments, with first-touch attribution that never blocks a signup.Sessions and security policyHow long sign-ins last, how to see and revoke your active sessions, and how a workspace can require two-factor authentication of every member.