Six interconnected product systems measure observable evidence, prepare governed work, and retain delivery and outcome states separately. External engines decide whether any source is crawled, indexed, ranked, mentioned, recommended, or cited.
A five-family AI Readiness audit of observable setup evidence, with prioritized findings. It does not report exactly how an external AI engine sees, ranks, recommends, or cites the business.
Weighted scoring across five evidence families: Off-site Authority (28%), On-page GEO (26%), Schema Engine (18%), Entity / E-E-A-T (18%) and Crawler Access (10%). These are starting estimates, not weights fitted to customer outcomes. SEO Health and measured AI Presence are separate scores, and insufficient readiness evidence remains unavailable rather than being converted to zero.
Thirty-nine industry-specific suggestion libraries plus one general fallback. Selection uses the submitted site's confirmed business type. Templates are not fact sources: persisted markup is limited to facts admitted by the shared customer authority record.
The scan requests current Google PageSpeed evidence and records the metrics the provider returns. Missing or failed PageSpeed evidence remains unavailable rather than being replaced with an invented measurement.
Scans robots.txt for directives that name GPTBot, ClaudeBot, PerplexityBot, and Google-Extended. A declared allow or block is reported as policy evidence; it does not prove a crawler arrived or that an engine indexed the site.
Run the same available-evidence scan against a submitted competitor URL and compare the five AI Readiness families side by side. Missing evidence remains unavailable, and the comparison does not establish why an external engine selected either business.
The current five family weights are starting estimates and have not been fitted to customer outcomes. The score summarizes observable setup evidence; it does not estimate citation eligibility or a causal share of external outcomes.
ClickRadius connects measured findings to prepared eligible corrections. It can apply a supported site effect only with current authorization and a connection proven capable of that effect; otherwise the work remains for review or customer implementation. Provider acceptance is recorded separately from an independent public-origin reread, and reversal is available only where the connector declares and binds a supported revert path.
Prepares 120–160 character meta-description revisions with the configured topic and editorial call to action. Generation does not promise a search snippet, click-through change, or AI citation.
Prepares title-tag revisions within the configured editorial range for review. A generated title does not promise click-through improvement, parsing behavior, ranking, or citation.
Prepares descriptive alt-text suggestions for eligible observed images. Coverage and factual fit must be reviewed; generation alone does not establish accessibility compliance, image-search visibility, or external-engine comprehension.
Relevance scoring can analyze up to 100 observed pages and prepare internal-link suggestions. A suggestion does not prove topical authority, crawler behavior, ranking, or citation.
Structured-data preparation from 40 schema template keys: 39 industry-specific keys plus a general fallback, selected through the confirmed business type. Generated markup can be validated before review; applying an approved exact revision requires current authorization and a connection proven capable of that effect.
WordPress REST API, JavaScript snippet on a compatible site, or Cloudflare Worker. Capability is proven per connection and effect: snippet changes are client-side and are not verifiable in the raw public-origin response.
After an approved exact revision proceeds with current authorization through a connection proven capable of the effect, ClickRadius retains the provider receipt separately from the applicable readback. A JavaScript-snippet effect is client-side and cannot be verified in raw origin HTML.
Each eligible effect record retains its exact revision and available prior state. Effects deployed through a connector that declares a bound revert path — including supported WordPress metadata and publication operations — can be reverted from the portal. Where a connector declares no reversal capability, the change is recorded and recovery is operator-owned.
Observed CMS and plugin signals can guide setup. They do not prove write scope or rendering behavior; each requested effect still requires a live, site-bound capability probe.
AI-assisted content drafts designed around citation-relevant inputs. Each draft passes source and factual review controls and can include supported schema. Eligible publication requires current authorization and a destination connection proven capable of publication; otherwise it is held for review or customer implementation. External engines decide whether to cite it.
Each standard site includes 8 authority-grade blog posts and 10 standalone knowledge articles per monthly billing period, plus a one-time 50-article Knowledge Foundation during the first 60 days. The Knowledge Coverage Engine selects distinct, useful topics from measured audience and corpus gaps rather than generating pages merely to reach a count.
Where the subject and sources support them, drafts can use quotations, statistics and source citations associated with gains in the cited Princeton-led experiment. That experiment does not prove a citation outcome for a customer article.
Governed drafts inventory visible factual units, fetch and hash cited sources, check quotation and numeric support, and fail closed when required evidence or authority is incomplete. These controls reduce unsupported claims; they do not certify zero errors or predict an engine response.
An admitted exact revision can publish only with current automatic-execution authorization for that revision and a WordPress connection proven capable of publication. Otherwise it remains for review; a provider receipt and public-origin observation retain separate states.
Where the active connector and configuration support discovery notification, ClickRadius can send a bound provider request after publication. A request or receipt is not indexing proof; each search engine decides whether and when to crawl or index.
Topic selection uses measured audience and corpus gaps to prepare distinct customer-question candidates. A candidate is not represented as observed AI demand, and the business is not inserted as an answer without supporting authority.
Record what sampled AI responses say about your business at a known time. ClickRadius queries five supported AI engines on a monthly schedule, classifies observed sentiment, detects grounded URL citations, and records confidence without treating the sample as every user response.
Monthly OpenAI-provider sample of observed mentions, recommendations, and grounded URLs
Supported pathMonthly Anthropic-provider sample of observed brand mentions and grounded URLs
Supported pathMonthly Google-provider sample of observed mentions and grounded URLs
Supported pathMonthly Perplexity-provider sample retaining returned source URLs
Supported pathMonthly xAI-provider sample of observed brand responses and grounded URLs
Supported pathEach observed mention is classified as positive, neutral or negative. Month-over-month records show changes in the sampled responses; they do not establish a model-wide perception or predict a recommendation.
Records when a retained provider response includes a grounded URL for the site, together with the exact query and page. The observation does not establish what caused the response or represent every user answer.
Each month, ClickRadius runs the governed query set across ChatGPT, Claude, Gemini, Perplexity and Grok. Historical samples retain their query and engine context so changes can be reported without inventing a weekly frequency.
Website optimization and off-site corroboration serve different evidence roles. ClickRadius evaluates business presence across connected and observable platforms while keeping external actions, provider receipts, and measured AI outcomes distinct.
Site-bound OAuth can support approved post distribution; other listing or response work remains guided unless its exact provider path and authority are proven.
Structured-data observation and bounded exact-name Wikidata candidate search; a candidate or publisher declaration is not proof of graph inclusion or entity recognition.
Site-bound OAuth can queue an adapted post for human approval and provider execution. Configuration, provider receipt, and external outcome remain separate.
Adapted draft and manual export only. Public Medium publishing is held until a site-bound destination identity is available.
Assisted manual research and customer-claimed permalink readback only; automated off-site conversation work is held.
YouTube and Vimeo analysis can inspect available metadata, transcript evidence, and prepare VideoObject markup; it does not publish or promise discovery.
Provider and customer authority are not established for public launch; any press-release work remains unavailable until its exact connection and effect are proven.
Directory evidence can be assessed and guided. No listing distribution is represented as complete without site-bound provider authority and a receipt for the exact destination effect.
Keep dated traffic, referral, conversion, cost, and revenue observations beside delivered-work evidence without automatically assigning causation. Coverage, attribution source, and unavailable states remain visible; before-and-after timing is not proof.
Display dated organic-session observations by available page, keyword-cluster, and time-period dimensions. A change after ClickRadius work is not automatically attributed to that work.
Classify observed visits carrying supported AI-referrer evidence and record their landing pages. Missing or stripped referral data remains unavailable rather than being inferred.
Where both populations and conversion evidence are available, compare observed AI-referrer and organic cohorts with their coverage. The comparison does not by itself value AI visibility or prove incremental revenue.
Compare dated AI Readiness evidence across all five families and show the observed score difference. The comparison does not assign causation or convert an internal score change into an external result.
Reports can place ClickRadius cost beside revenue records that carry their own attribution evidence. Score movement, organic growth and AI referrals are not automatically assigned causal credit.
White-label monthly reports can be generated on schedule with delivered-work records, measured score changes, and bounded AI Presence observations. Agency review and human accessibility/policy acceptance remain required before launch use.
ClickRadius evaluates eligible work on named schedules. An exact external effect proceeds only with current authorization for that revision, a destination connection proven capable of the effect, and the applicable runtime policy; otherwise it is held or remains for review. Citation monitoring runs monthly.
Behind the core loop, ClickRadius has specialized analysis and workflow modules. Each reports its own evidence and execution state; none promises an external-engine outcome.
Audit the content and entity signals that prepare your site for Google’s answer surfaces. Observed Google AI Overview tracking activates only after its live collector passes validation.
A lightweight beacon identifies visitors arriving from ChatGPT, Perplexity and Gemini, so referral traffic from AI engines is measured rather than estimated. These are observed arrivals reported as evidence; ClickRadius does not attribute them to any single optimization.
Compare observable five-family readiness evidence and separately retained AI Presence samples. The comparison does not identify a guaranteed action, causal gap, or path to overtake another business.
ClickRadius’s public library contains 1,650 published Q&A pages across 25 industries, counted September 4, 2026. That count measures publication inventory, not whether an external engine cited a page.
Analyze available YouTube or Vimeo metadata and transcript evidence, prepare supported VideoObject markup, and report a bounded content score without promising retrieval or citation.
Configured profile-derived views for ChatGPT, Gemini, Perplexity, Claude and Grok. The profile weights are hand-assigned, not fitted to provider behavior or customer outcomes.
ClickRadius provides dashboards, connection workflows, and governed automation while keeping customer approval, provider execution, public readback, and external outcomes distinct.
Interactive visualizations for scores, citation trends, content performance, and outcome metrics. See your progress without needing to understand the tech.
Approved agencies set and collect their own end-client price and remain the client support owner. ClickRadius wholesale terms, objective scope adjustments, and agency expenses remain separate from that retail-price authority.
WordPress and Shopify connector code supports different operation sets. Credentials alone are not capability proof: each site and requested effect requires the applicable live scope, plugin, provider, and public-origin checks.
Step-by-step setup can collect site connection details, confirmed business facts, and an initial AI Readiness scan. Completion time depends on current authority and provider checks; no fixed activation time is promised.
The implemented retail lifecycle uses a required payment method for a 14-day trial followed by month-to-month billing. No annual retail offer is approved, and public payment activation remains closed until test-payment and webhook proof is accepted.
Try ClickRadius with a free AI Readiness scan before committing. See your score, review your issues, then decide if you want the full platform.
See how ClickRadius separates an audit, prepared work, current authorization, effect-specific delivery evidence, and later observations.