Changelog & roadmap

We build in public. Everything below shipped on the date shown, and the roadmap is what we're genuinely building next, not vapour.

Shipped

• 14 AugThe MCP server grows (fourteen tools today): your AI can now pull the get_fix_plan priority list, get_citation_sources (the domains engines actually cite in your category, and whether you appear), and trigger a capped request_rescan without leaving the chat · histories now state their full span (measured N times since date, named on K) so partial windows can't mislead · read-only annotations on every read tool
• 14 AugMCP gains a sixth tool: get_share_of_voice, so your AI can pull your slice of AI-answer mentions against the rivals in your space, and how it has moved, in the same conversation as everything else · one-click connection test in your console's Connect your AI card: check the live endpoint and see exactly what your AI would get back, before you wire it up
• 13 AugYour console is now a live MCP server, out of private testing: add one URL to Claude, Cursor or any MCP client and your AI can query your real measurements mid-conversation, five read-only tools, 2-minute setup · where you sit inside the answer: reports and consoles now read position from the held verbatim answers, how deep your first mention lands and whether you're named ahead of rivals in the same answer · the market map adds the new free open-source BYOK corner (Lettertrace) with an honest read on what US$0 tracking includes
• 7 AugScoring v8: readiness now moves only when your site moves (engine answers can no longer inflate it), every report leads with the results and one next step, and reports reconcile the free scan's 2-engine spot check against the full 7-engine audit in plain words · new research: the AI-visibility market, mapped, published entry-tier pricing across the category, every cell verifiable at the source
• 23 JulDNS-AID: every scan now checks for the emerging agent-discovery DNS record (_index._agents, IETF draft): it renders in your agent-readiness section, and when it's missing your fix plan carries the exact record to add · new free DNS-AID checker and 5-minute setup guide · our own agent-skill card is live at /.well-known/agent-skills, we're discoverable the same way we score others · new research: who wins the moment buyers compare you (1,001 high-intent SERPs, brands lose 67% of their own 'vs' searches) and JS-injected schema is invisible to AI · comparison-page playbook + a new deep-scan fix that names the comparison moments you're surrendering
• 22 JulOne-tap AI handoff: buttons in every monitoring email open ChatGPT, Claude, Gemini or Perplexity with your live numbers, tracked questions, mentions and this week's action already loaded in the chat; nothing to copy or download · AI context pack: one-click .md download of your live data, built for ChatGPT/Claude projects, always fresh at the same link · every weekly email now leads with one recommended action derived from your own data · paid reports rebuilt for clarity: answer up front, tap-to-open chapters, cream default with a dark toggle
• 16 JulRival share of voice (Growth), you vs the rivals AI names, head to head in real AI answers · Launch Radar: arm daily measurement for two weeks around a launch and see, day by day, whether AI noticed · Re-ask any question live from your report · Name-collision detection: when AI confuses you with a namesake, we show exactly who and how to fix it · every result now shows the model, market and timestamp behind it · Markdown for AI agents at /llms-full.txt, we serve clean Markdown to AI, the practice we recommend · Rescan now: trigger a fresh deep scan from your console the moment you ship a change, instead of waiting for the weekly one (capped so it stays a spot-check)
• 14 JulLaunched on Product Hunt · Growth tier: AI share of voice, 25 tracked questions, a weekly AI Strategy Brief written from your own data
• 12 JulThe 🇺🇸 American AI Visibility Index opens: vets, dentists and plumbers live, growing nightly (🇬🇧 UK forming) · The monitoring console: every subscriber gets a living dashboard: score trend, question-by-question heat grid, verified business identity, a ⚙ technical view, daily web-mention sweep with same-day alerts when a new site cites you, and AI traffic on one private link. Weekly re-scans upgraded to full deep scans with measured citation sources. We're customer #1: our own live console · French INSEE Sirene verification joins the AU + UK registry checks · engine-outage failover (a down engine can no longer degrade your scan) · share-your-score on every free result
• 11 JulMeasured citation sources in every paid report, which domains ChatGPT actually cites in your category, from real conversations · AI-traffic beacon: a 1-line snippet that counts the real visitors AI sends you, with a live dashboard · UK Companies House verification joins the Australian Business Register check · 5-findings-or-refund guarantee on the full report · reviews open on Trustpilot
• 10 JulGlobal USD pricing · honest free-report gate (your 2 tested questions shown with real results, the paid dozen clearly locked) · every buyer question multi-sampled with majority voting
• 9 JulFeatured in The Shift, Packt's AI newsletter · dataset passes 500 businesses
• 8 JulSearch-demand snapshot in reports (real Google volumes per buyer question) · Reddit-presence signal · Google Search Console reality-check (private beta)
• 7 JulThe AI Visibility Index goes live, industry leaderboards that grow nightly · demand-weighted visibility scoring
• 5 Jul#1 AI-visibility tool on There's An AI For That, #3 trending across all categories
• 1 JulThe GEO framework published · 90-second demo video
• 28 JunLaunch: free scanner across ChatGPT, Claude, Perplexity and Google's AI, with Google AI Overviews checks

Next

• soon🇬🇧 UK edition of the AI Visibility Index (AU + US live) · public "connect your Google Search Console" (pending Google's verification review; pilot subscribers already see their data live)
• exploringAgency & white-label: agencies and consultants are already among our heaviest scanners, checking their clients. We're designing a partner tier: white-labeled reports (your logo, no upsells), a multi-client console, and wholesale Snapshot pricing so you can resell measurement under your own brand. Register interest →

Want something on this list? Tell us, the fastest way to shape a small tool is to talk to the people building it. Or run a free scan and reply to your results email.