Questions about PlayHero

PlayHero is WebUX intelligence for GTM teams: specialist agents monitor your site, surface high- and critical-severity issues as Rallycards, and log every approve / dismiss / defer decision on a permanent Heroboard. Four launch agents — UX Sentinel, Drift Detector, Page Vitals Enforcer, and Lumen — cover conversion UX, metric drift, Core Web Vitals, and AI discoverability. Humans approve every action; nothing executes on your site without you.

WebUX stands for Web User Experience as it relates to conversion. PlayHero measures WebUX across seven scored dimensions: Load Intelligence, Signal Hierarchy, Messaging Clarity, Content Clarity, CTA Clarity, Conversion Friction, and Agent Visibility. A WebUX score is a 0–100 composite that reflects how well your website converts visitors into leads — not how it looks, but how well it performs at its commercial job.

Enter your URL on the PlayHero homepage and run a free scan. In about 60 seconds you get a WebUX score, seven-pillar breakdown, and up to three Rallycards — no account or card. Create a free account when you want to save results, open a basic Heroboard, and run Compare. Upgrade to Brand ($24 / site / month, 1 to 5 sites) for daily monitoring, history and alerts.

Yes. Public scans stay free: no account required, no card. Fair-use rate limits apply (per IP and per org) to keep the service healthy — not unlimited blast scanning.

A PlayHero scan completes in 60 seconds or less. The scan runs four detection layers simultaneously: DOM structure analysis, LLM messaging interpretation, Lighthouse performance measurement, and viewport simulation across common screen sizes. Results are available immediately in the browser — no waiting for an email report.

A WebUX score is a 0–100 composite score measuring your website's conversion health. Scores below 60 indicate conversion-critical issues. Scores between 60 and 75 indicate meaningful opportunity. Scores above 75 indicate a well-optimized site with incremental improvement available. The score is calculated from seven pillar scores — each weighted by its impact on conversion — and recalculated on every scan so you can track improvement over time.

Yes — this is one of the fastest-growing use cases for PlayHero. When a prospect uses an AI assistant to research solutions in your category, that assistant reads and parses your website rather than browsing it visually. PlayHero scans for signals that affect both human visitors and AI parsers: structured data presence, page content machine-readability, pricing page clarity, and whether key conversion content is accessible without JavaScript execution. The Agent Readiness findings surface as Rallycards the same way human UX findings do — prioritized by severity and routed to a human for approval.

A Lumen scan checks whether your website is structured in a way that AI systems can correctly parse, cite, and reason about. This includes: presence of an llms.txt file at your domain root, schema.org structured data markup, OpenGraph tag completeness, machine-readable pricing content, robots.txt directives for AI crawlers, and whether key page content is accessible without JavaScript rendering. These signals determine whether your site appears correctly when an AI assistant researches your product category on behalf of a buyer. PlayHero monitors these signals on the same cadence as human UX signals — so regressions are caught before they affect your AI search visibility.

llms.txt is a plain-text file placed at the root of your domain (e.g. yourdomain.com/llms.txt) that tells AI systems what your product does, how to cite it, and where to find structured information about your pricing and features. It is to AI crawlers what robots.txt is to search engines. When a buyer's AI assistant researches your category, sites with an llms.txt file are more likely to be correctly identified, cited, and recommended. PlayHero's Lumen agent checks for llms.txt presence and quality on every scan. When llms.txt is missing, a Generate button appears in the scan results — PlayHero can synthesize a ready-to-place file from your scan data.

Yes. Email a URL to agent@playhero.ai and PlayHero runs a scan and replies with the score, top Rallycards and a results link. There is also a read-only MCP server at /api/agent — tools list_rallycards and get_rallycard return agent-safe cards (page URL, DOM selector, fix hints, impact rank) so a coding agent can propose fixes for human review. Bearer-token auth; no writes.

Yes, and this is the exact profile we built for. Teams using Clay, HubSpot AI, Gong, or agentic outbound tools are typically the ones who notice first that their pipeline attribution data doesn't match their site conversion data. PlayHero monitors the conversion layer that all of those tools depend on — the pages your buyers land on, your pricing page, your demo request flow. If those pages degrade after a deploy, an update, or a content change, PlayHero surfaces it before it becomes a quarter-end number conversation.

Three things have the highest impact: (1) Add an llms.txt file to your domain root — PlayHero can generate one from your scan results. (2) Ensure your pricing page has a machine-readable version — a plain markdown file at /pricing.md that AI agents can parse without rendering JavaScript. (3) Add schema.org structured data markup to your key pages. PlayHero checks for all three as part of Lumen scanning and surfaces missing or incorrect signals as Rallycards with specific remediation steps.

This is exactly the shift PlayHero was built for. The buyer journey now has two phases: an AI research phase (where agents parse your site, compare pricing, and shortlist vendors) and a human decision phase (where the actual purchase is made). PlayHero monitors both surfaces simultaneously — scanning for human UX signals that affect conversion and for agent-readiness signals that affect whether you appear in the AI research phase at all. No other monitoring platform watches both layers continuously.

A Rallycard is a structured decision card generated at high and critical severity, plus lower-severity findings with strong supporting evidence. Each card contains the detecting agent, severity, a plain-language finding, estimated pipeline impact, the detection method used, and a recommended action. Rallycards require an explicit human decision — approve, dismiss, or defer — before anything happens.

The Heroboard is the permanent, searchable record of every Rallycard and every team decision made in response to it. It is a governance layer, not a dashboard or task manager. Every card that passes through the Heroboard — approved, dismissed, or deferred — creates a permanent log entry with the actor's name and timestamp. The Heroboard is institutional memory: a complete, auditable history of every conversion decision your team has made.

Approving a Rallycard moves it to 'In Progress' and creates a permanent audit log entry with your name and timestamp. It is a coordination signal for your team — it does not automatically execute any change on your website. Your team executes the fix. Agents continue monitoring and update the card status when the next scan detects the issue is resolved. Nothing executes without your approval. That is not a limitation — it is the product.

Yes. Dismissing a card archives it permanently with a required reason selection — the reason is logged and searchable. Deferring a card snoozes it for a duration you specify; it returns to the pending queue automatically when the snooze expires. Both dismissed and deferred cards remain in the Heroboard audit trail. No card is ever deleted — the permanent record is a feature, not a limitation.

Findings surface at high and critical severity, plus lower-severity findings with strong supporting evidence. PlayHero does not publish a finding-confidence percentage for that rule — there is no measured one to publish. The exception is Lumen's framing-accuracy check, which compares an AI excerpt to your own headline and value props and surfaces only when that secondary pass scores 70 or above. Pipeline impact stays a separate benchmark-based estimate, always labeled as such.

Pipeline impact is a benchmark-based estimate modeled from your WebUX pillar scores mapped to B2B SaaS conversion sensitivity research. It is never sourced from your CRM, your actual traffic data, or your real revenue numbers at the anonymous scan tier. The estimation methodology is disclosed on every Rallycard — not in a footnote, inline on the card. After signup, you can edit the underlying assumptions (traffic volume, average deal value, CVR baseline) to calibrate estimates to your actual numbers.

No. Once a Rallycard is approved, the original finding is locked and cannot be edited or deleted. Status changes (approve, dismiss, defer, resolve) are logged with actor and timestamp, but the underlying finding data is immutable. This is a deliberate trust architecture decision — the audit trail requires immutability to be trustworthy as institutional memory.

PlayHero launches with four specialist agents: UX Sentinel (conversion friction and technical UX signals), Drift Detector (WebUX score / pillar and AI-signal drift across observations), Page Vitals Enforcer (Core Web Vitals and performance thresholds), and Lumen (agent visibility and AI buyer discoverability). Each agent has a defined domain and a specific detection methodology. A Rallycard surfaces at high and critical severity, plus lower-severity findings with strong supporting evidence. No two agents overlap in mandate.

UX Sentinel monitors conversion friction and technical UX signals. It watches CTA visibility across mobile and desktop viewport sizes, readability thresholds (line-height, font-size, and related layout heuristics), heading hierarchy, conversion friction (motion-aware primary-form field thresholds and conversational-interface detection), and visual scannability. It surfaces a Rallycard at high and critical severity, plus lower-severity findings with strong supporting evidence. UX Sentinel never executes changes, and it never conflates finding validity with revenue certainty. It is not a WCAG audit.

Drift Detector watches for material changes in WebUX pillar / score observations and AI-signal snapshots for a monitored URL — catching regressions before they become a quiet quarter. It compares against prior observations rather than CRM metrics. It does not ingest CTR, MQL velocity, form-completion rates, or pricing-page engagement analytics. Drift findings today surface as observations and alerts; they are not always minted as Rallycards.

Page Vitals Enforcer monitors Load Intelligence signals (LCP, FCP, CLS, TBT) against Google's conversion thresholds — aligned with Core Web Vitals methodology where applicable: LCP ≤2.5s, CLS ≤0.1, TBT ≤200ms. It fires a Rallycard when a threshold is breached on a scheduled observation. It does not currently detect deploy events — monitoring is cadence-based (daily / weekly / monthly), not deploy-triggered. Page Vitals findings are objective lab measurements when PageSpeed data is available — an LCP of 3.4s is a fact, not an interpretation.

Monitoring requires Brand or Agency. Brand is $24 / site / month, 1 to 5 sites, no minimum, on a daily cadence with all four launch agents. Agency is $24/site with a 9-site minimum ($216 floor). Free includes on-demand scans, Compare, and Heroboard — not continuous monitoring.

No. PlayHero is read-only on your site: agents detect and surface; they never deploy or edit your pages. Approving a Rallycard is a team coordination signal — your stack, your change. Any future automation will still require explicit human approval.

Yes. PlayHero supports competitor URL scanning in observation mode. When you scan a URL you do not own, the full seven-pillar analysis runs but pipeline dollar estimates are removed — revenue impact requires your actual traffic data to be credible. You receive relative UX scores, qualitative findings, and competitive positioning without speculative revenue projections. Competitor URLs you add for Compare don't count toward your sites.

Signals that are neither high nor critical, and that lack strong supporting evidence, do not become Rallycards. They stay with the detecting agent and are re-evaluated on the next run. You do not get a card for every detection — only high and critical severity, plus lower-severity findings with strong supporting evidence.

Yes. PlayHero's agents evaluate your pages against the signals AI systems use to assess and recommend sites — structural clarity, conversion path legibility, Core Web Vitals, and messaging consistency. A page that converts human visitors but confuses agent evaluators is still a pipeline risk. PlayHero surfaces both.

Rallycards are triggered by what PlayHero's agents detect during scans — not by session type. If an issue affects the pages an AI agent would evaluate on behalf of a buyer (pricing, demo request, signup), it will surface as a Rallycard with severity and pipeline impact context.

Three public tiers: Free ($0) — 1 site, 3 scans a month. Brand — $24 / site / month, 1 to 5 sites, no minimum. Six or more sites is Agency. Agency — $24/site/mo with a 9-site minimum ($216 floor); portfolio view, client workspaces, resale rights. Pay annually and get two months free.

The top 3 ranked experiments for any URL are free with no account. The full Testplan — 5–8 experiments with complete ICE scoring and every evidence source — requires a signed-in workspace. Test plans copy as Jira/Linear markup.

From the same scan that powers your free WebUX report. PlayHero derives test candidates from detected findings and ranks them by confidence and ease. The full plan adds AI-generated hypotheses grounded in your Rallycards, competitor gaps, and journey data.

Pipeline estimates only appear for domains you own. Free anonymous scans run in observation mode — you'll see rankings and rationale, never invented dollars.

No. PlayHero does not require access to your CRM, analytics platform, or internal revenue data to scan your website or surface Rallycards. Scans run against the publicly accessible DOM of your website. Pipeline impact estimates use benchmark data from B2B SaaS conversion research — not your actual traffic, leads, or revenue numbers. After signup, you can optionally input your own traffic and deal value assumptions to calibrate estimates to your real numbers.

PlayHero uses four detection methods across its agents: DOM structure analysis (objective measurement of page elements — a CTA below the fold is a measurable fact), LLM messaging interpretation (probabilistic AI assessment of copy quality and ICP alignment), Lighthouse integration (Google's open-source Core Web Vitals measurement tool), and viewport simulation (testing CTA and layout visibility across common device screen sizes). Every Rallycard discloses which detection method produced the finding.

No. PlayHero never writes to your site or your repo. It is read-only monitoring and analysis. Human approval records intent on the Heroboard; your team ships the fix.

Yes. Approved, dismissed, and deferred Rallycard decisions are permanently logged with the actor's name and timestamp. Deletion of decisions is not permitted — this is a deliberate architectural choice. The Heroboard functions as institutional memory, and institutional memory requires permanence to be trustworthy. Full decision history is exportable as CSV on Brand and Agency plans.

Pipeline impact ranges are benchmark-based estimates, not guarantees. They are modeled from B2B SaaS conversion sensitivity research and are disclosed as estimates on every Rallycard — inline on the card, not in a footnote. Finding Confidence (the likelihood the issue is real) and pipeline impact are always stated separately and never conflated. PlayHero will never present a revenue estimate as a revenue guarantee.

Free accounts do not include continuous monitoring. Brand monitors 1 to 5 sites at $24 / site / month, with daily scans. Six or more sites is Agency — $24/site with a 9-site minimum. Competitor URLs you add for Compare don't count toward your sites.

Free includes rate-limited free scans (with Rallycards per scan), Compare, and a basic Heroboard. Continuous monitoring — UX Sentinel, Drift Detector, Page Vitals Enforcer, and Lumen on a scheduled cadence — requires Brand or Agency.

Yes — paste an incoming-webhook URL under Settings → Notifications; new Rallycards at your chosen severity and score drops post there, and go by email to your team.

Yes. Brand and Agency plan accounts can export their full Heroboard decision history as a CSV file. The export includes every card, every decision, the actor, the timestamp, the agent, the finding, and the outcome. Free plan accounts can view their history inside PlayHero but cannot export it.

Yes. Scan results and Resolution Briefs are shareable via a read-only link. The recipient does not need a PlayHero account to view the scan result or the brief content.

A read-only MCP server is live at /api/agent. The tools list_rallycards and get_rallycard return agent-safe Rallycards (page URL, DOM selector, fix hints, impact rank) under bearer-token auth with scope agent:rallycards:read; no writes. A public REST API is not shipped. Until it is, agent@playhero.ai accepts scan requests via email and returns structured results.

A Resolution Brief is a full implementation specification generated by the detecting agent after a Rallycard is approved. It includes a technical implementation spec for the fix, structured verification criteria that the next automated scan will check to confirm resolution, and an effort estimate. The brief is shareable via a read-only link — no account required for the developer or recipient who receives it.