Changelog

What's shipped on Yogoo. New entries land with every customer-visible release.

v0.8.3Your answers, counted

  • improvedThe Overview now shows your answers as counts by engine, in place of the Yogoo Diamond.

v0.8.2Claude answers now come from Claude

  • fixedBetween October 14 and October 22, answers we showed as Claude’s were produced by a different AI model. A routing fault in our sampling setup sent the Claude questions to a Google model. We have removed those answers from every number and view; where they were, you will see Claude’s previous answer until a new one arrives. From today every Claude answer is checked to come from Anthropic before we record it. No published market score used the affected answers.

v0.8.1ChatGPT now searches the web before every answer

  • improvedFrom October 20, ChatGPT has to search the web before it answers the questions we ask it. Before, it chose for itself, and on a few questions it answered from memory with no sources every time. Our August 29 note below said we record those answers as they come. That held until mid-October, when we started leaving them out instead, because an answer with no sources can’t show whether ChatGPT links to you, and we didn’t say so here at the time. Requiring the search fills those gaps. In a test on the same questions with and without the requirement, answers kept about the same length and the same number of sources. The test was too small to rule out small changes in how often a particular business appears, so we’re checking each engine line over the next week. Answers from before October 20 stay in your 28-day window.

v0.8.0What counts as a mention, and two things that no longer exist

  • improvedYour market number now counts an AI answer that names your brand in prose, not only one that links to your domain. Until today the scorer matched your domain and nothing else, so an answer saying “Fox & Jane is a reliable choice” — a recommendation, by name, with no URL — scored as no appearance at all. That is most of how AI answers talk about brands, so almost every market number was lower than the truth. In our own test markets the correction moved one brand from 31.7% to 37.5% and another from 1.7% to 4.2%, and raised answers-naming-the-brand from 8 of 120 to 37 of 120. The same change stops the opposite error: a domain like seoexpertdubai.ae yields phrases such as “SEO Expert in Dubai”, and those were being treated as that firm’s name, so an answer recommending “an SEO expert in Dubai” counted as a mention of one specific company. Generic phrases no longer count; a name has to identify a company in that market. Figures from before and after today are not comparable — we would rather say that plainly than quietly restate your history.
  • removedScores no longer carry a confidence band, and we no longer publish interval maths in the methodology. We shipped bands at launch (v0.1.0, below) and removed the underlying calculation some time ago; the claim outlived the feature on our own methodology page, which is our mistake. What we report instead is the rate together with the counts behind it — how many AI answers we looked at, and how many named or linked you — so you can see the measurement rather than a margin around it.
  • removedThe methodology page described a Sentiment view as available on every tier. It was never built. The section is gone. We measure whether AI names or links you, not how warmly it talks about you.
  • fixedThe free check now lives at yogoo.ai/app/onboarding/new-tracker, and yogoo.ai/audit redirects there. v0.1.0 below announced it at /audit; that URL had been returning a 404 rather than redirecting, which we have fixed. Seeing your market number now needs a free account — no card.

v0.7.0Live web search on ChatGPT, Gemini, and Claude — phased rollout

  • improvedChatGPT, Gemini, and Claude measurements now run with provider-native web search in the following categories: e-commerce platforms, legal SaaS, CRM, marketing automation, and knowledge base & docs. In these categories, all five engines run with live web search enabled — measuring what AI finds when it can actually search, instead of model training memory.
  • improvedPhased rollout: remaining categories continue to run memory-based on these three engines and are scored accordingly (samples measured under the web-search configuration are never mixed with samples measured without it). Categories switch to retrieval as they are enabled; each switch is dated here.
  • addedHonesty note: ChatGPT invokes its web search at its own discretion — when it answers from memory instead of searching, that is recorded as-is, not forced. Perplexity and Google AI Overviews have always measured live retrieval and are unchanged.

v0.6.0The Yogoo blog is live

  • addedNew blog at yogoo.ai/blog with our first three articles — on AEO, improving AI visibility with GEO, and a comparison of AI citation tracking platforms.

v0.5.2Faster first recommendations

  • improvedOn Pro and up, your first recommendations now appear shortly after you add a tracker — no long wait for your initial set of moves to show up.

v0.5.0Content — draft and tune to your voice

  • addedArticle drafting (Pro and up) — generate a full long-form article draft from any brief, then refine it in the editor. It stays a draft you review and publish — nothing is posted for you.
  • addedBrand voice (Premium) — teach Yogoo your voice from a few of your own writing samples. Your articles and posts then match how your brand actually writes.

v0.4.0Workflow — from citation gaps to drafts

  • addedRecommendations (Pro and up) — a new dashboard workspace that turns your citation gaps into prioritized opportunities, article briefs, and content drafts you review before publishing.
  • addedShort-form posts (Pro and up) — generate X and Reddit post drafts from an opportunity. Drafts only — you review and publish; nothing posts automatically.

v0.3.0Citation Position Score + custom prompt monitoring

  • addedCitation Position Score — new Diamond axis measuring how prominently you're placed when AI cites you (replaces Top Source Rate). Position 1 = 100%, smooth decay to position 10.
  • addedCustom prompt monitoring — add your own buyer-intent prompts and track how AI responds. Similar-prompt suggestions help you find what's already in the system.
  • addedStage-aware prompt CTAs — empty buyer-journey stages now surface 'Add a prompt' actions so you can fill coverage gaps directly.
  • addedPhrase highlighting — brand names and competitor names are highlighted in AI response text across Activity and Prompt detail views.
  • addedActivity feed pagination — browse the full history of sampled AI responses, not just the latest batch.
  • improvedYogoo Score now uses system prompts only — consistent benchmark across all users. Custom prompts are tracked separately in Monitoring.
  • improvedDomain redirect resolution — misspelled domains (e.g. saleforce.com) now resolve to the canonical target automatically during onboarding.
  • improvedBranded email chrome — all notification emails now carry the Yogoo header, footer, and one-click unsubscribe.
  • fixedDiamond chart no longer lopsided — Citation Position produces values in the 40–90% range vs. the old 2–5% Top Source Rate.

v0.2.0Dashboard, tiers, and onboarding

  • addedYogoo Dashboard — Pillar 1 monitoring surfaces: Overview with Yogoo Diamond, Monitoring with competitor Share of Voice, per-prompt citation detail, Mentions feed, and Settings.
  • addedYogoo Diamond — 4-axis radar chart visualizing Citation Rate, Citation Position, Prompt Coverage, and Share of Voice. Overlay your brand against competitors.
  • addedVisibility over time — trend chart tracking your Yogoo Score and each Diamond dimension over 30/60/90-day windows.
  • addedTiered plans — Free, Basic ($29/mo), Pro ($99/mo), Premium ($399/mo), and Enterprise. 14-day Basic trial, no card required.
  • addedOnboarding wizard — enter your domain, we crawl and classify it, suggest a category, and set up your tracker in 4 steps.
  • addedDynamic category suggestions — if your category isn't in our taxonomy, propose one. We validate it against the B2B SaaS landscape and auto-create if viable.
  • addedCompetitor pinning — pick the competitors you care about. The Diamond, leaderboard, and Share of Voice all reference your comparison set.
  • addedStripe billing — upgrade, downgrade, and manage your subscription from the dashboard. Annual plans save 2 months.
  • addedBand-change email alerts — get notified when your visibility band shifts (e.g. Emerging → Established).
  • addedLeaderboard anchored mode — see where you rank in the full universe, with your picked competitors badged.
  • addedFAQ sections with structured data on /pricing and /methodology for better AI engine indexing.
  • improved50-prompt seed corpus — polished buyer-intent prompts across 5 launch categories, up from the T1 minimal set.
  • improvedFreemium dashboard — Free users see their score and a masked preview of the full dashboard to drive conversion.

v0.1.0T1 launch — public AI citation audit

  • addedFree AI citation audit at yogoo.ai/audit — no login, instant brand visibility check.
  • added30-category coverage across CRM, project management, observability, marketing automation, and 26 more B2B SaaS categories.
  • added5-engine measurement — ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews.
  • addedDecomposable Yogoo Score with Citation Hit Rate, Share of Voice, Prompt Coverage, and confidence band components.
  • addedConfidence bands on every score — variance-derived intervals so a single-run estimate doesn't masquerade as precision.
  • addedEmail signup for uncovered domains — get notified when your category is added.
  • addedOpen methodology page at yogoo.ai/methodology with every metric defined and worked examples.