Scorecard measures where you stand. Search Tracker watches the prompts that matter four times a day. AI Ranking pulls your real GSC queries and runs them through four AI models with a 5-layer brand-safety classifier. Autopilot ships the technical fixes through an approval workflow with auto-rollback. Content Engine generates quality-gated pages. Strategy Advisor synthesizes the signal. Visibility Trends proves what worked. Everything you need to win the top slot in AI search.
The AI Visibility Scorecard runs 8 buyer-intent queries across ChatGPT, Gemini, Claude, and Perplexity. Every query is generic to your category. None of them mention your brand. The result is an unbiased view of whether AI actually recommends you when buyers ask real questions.
Most brands do not know where they stand in AI search until they run their first scorecard. The scorecard gives you the answer in 60 seconds: your visibility category (Not Visible to Dominant), your exact score across 4 AI models, and the specific prompts where you lose to competitors.
Every paid plan includes unlimited scorecards. Free users get one scorecard to see their visibility category.
The Scorecard is a snapshot. AI Search Tracker is the movie. Add up to 25 custom buyer-intent prompts per brand, and TopSlot queries all 4 AI models four times a day at 00:30, 06:30, 12:30, and 18:30 UTC. You see which prompts you win, which you lose, when competitors overtake you, and which fixes moved your visibility.
Three variance layers run on top of the daily snapshots. ST-2 generates rephrased prompt variations weekly to surface rephrasing-sensitivity. ST-3 computes a stability score per prompt by comparing the four daily snapshots, so you know which trend lines are signal and which are model-load noise. ST-4 runs proxy-rotated US/UK/IN snapshots on Agency to expose regional competitor differences. Tracker data is consumed in parallel by Strategy Advisor, Content Engine, and Visibility Trends — no engine drives another's decisions.
Pre-built prompts are guesses. Real GSC queries are evidence. AI Ranking pulls every query that drove an organic impression in the last 28 days, removes anything containing your brand name through a 5-layer classifier, ranks the survivors by impact, and fans the top 30 across ChatGPT, Claude, Gemini, and Perplexity.
Every result is labeled with one of four citation states: cited_with_url (your domain appears as a link), name_only (verbatim brand mention without URL), descriptive_only (semantic match against your brand vector without naming you), or not_mentioned. The descriptive_only layer catches recoverable visibility most other tools miss.
Autopilot is the execution layer that separates TopSlot from every other AI visibility tool. One JavaScript pixel plus an NPM server adapter cover both client-rendered and server-rendered fixes. Adapters ship for Next.js, Express, Remix, and Nuxt.
Six locked sub-features keep deploys honest. F-1 catches AI bot crawls server-side. F-2 ships fixes through the hybrid pixel + NPM model. F-3 routes every fix through a deploy-approval workflow with tier-gated quotas (1/1/5 per day) and a 200-entry JSONB audit log. F-4 verifies every deployed fix every 10 minutes against your rendered HTML and auto-rolls back regressions with a 24-hour alert throttle. F-5 tracks which AI bots crawl which pages (GPTBot, ClaudeBot, PerplexityBot, Google-Extended). F-6 is the Agency-only cross-platform deployment dashboard.
<!-- TopSlot AI Autopilot --> <script src="https://cdn.topslot.ai/pixel.js" data-site-id="SITE_ID" async></script>
Not a blank-canvas writer. A six-stage content operations system. CE-1 classifies every page on your site as winning, opportunity, underperforming, dormant, or neutral. CE-2 recommends optimize, create, or retire with locked dedup precedence (retire > optimize > create) so you don't get conflicting suggestions on the same topic. CE-3 generates 9-section briefs with an approval workflow.
CE-4 is where most other tools fail. The LLM generates HTML + JSON-LD from an approved brief, and the output must score ≥60 on an eight-dimension quality gate (citation likelihood, intent match, schema validity, keyword coverage, internal links, length, readability, originality). Drafts that fail return HTTP 422 with a focus-dimension hint — you never receive bad content. CE-5 DBSCAN-clusters your full portfolio into 11 content types and computes the internal link map. CE-6 baselines every published piece and snapshots performance at day 7, 14, and 30, calibrating brief weights monthly with a ±3pp cap.
Data without interpretation is noise. The Strategy Advisor ships five distinct intelligence layers. ADV-1 generates a weekly Wins/Alerts/Opportunities briefing every Monday morning. ADV-2 runs daily forecasts on every tracked prompt with locked confidence levels (high ≥0.85, medium-high 0.70–0.85, medium 0.55–0.70, low <0.55). ADV-3 fires anomaly alerts only when volume + rank + citation drops happen together in the same window, severity 2/3/4, capped at 5 active per brand to prevent alert fatigue.
ADV-4 is the counterfactual layer. Ask the Advisor what would happen if you publish three FAQ pages next week and it projects impact across 1, 7, 30, and 90-day windows using your tracking history as the baseline. ADV-5 bundles all of this into a monthly white-labeled PDF per client brand on Agency, with cross-tenant isolation enforced at the data layer. The Advisor describes what your data is doing — Autopilot and the Content Engine ship the fixes. (ADR-001: synthesizes, never prescribes.)
Single-point data is vanity. Trend data is strategy. Six sub-features stitch together every visibility-relevant event in your brand's life. VT-1 unifies AI mentions, schema fixes, content publishes, rank changes, external industry events, and calibration corrections into one timeline. VT-2 overlays external events (algorithm updates, news cycles) so you can see when industry shifts moved your score. VT-4 attributes each visibility delta to specific actions across 7/14/30-day windows against a 28-day baseline. VT-5 tracks calibration drift between forecasted and actual values daily.
VT-6 rolls everything into a daily categorized brand health composite (Strong / Moderate / Weak / Critical) across five dimensions (mentions, citations, prominence, sentiment, freshness) and ships a weekly executive-brief PDF every Friday at 11 AM UTC for Agency tier. (ADR-001: Visibility Trends DESCRIBES, never PRESCRIBES — same locked rule as the Advisor.)
Every feature above is free to preview on the scorecard. Run yours in 60 seconds, see the data for your brand, decide from there.