Clear answers about crawler-readable HTML, setup, AI crawlers, and pricing.

Answers about website Health, AI Visibility, content drafts, the Agent, AI Ads, crawler rendering and account limits. Sign in for the setup and workflows available to your account.

What does Prerender Buddy do?

PB connects website Health and recorded crawler activity with AI Visibility monitoring and evidence-based next steps. Eligible accounts can prepare content drafts and AI ad creatives. Managed prerendering helps when public pages have a confirmed JavaScript delivery gap. Features depend on account access and configuration.

Who is Prerender Buddy for?

Founders, marketers, developers and agencies who want to understand website access, monitor relevant AI answers and act on the evidence. A site does not need broken JavaScript rendering to benefit from AI Visibility or page monitoring.

Does Prerender Buddy replace SEO strategy?

No. It helps collect and interpret evidence and supports specific workflows. Product fit, useful content, credible sources and technical delivery still matter. Monitoring does not guarantee rankings, recommendations or revenue.

Does PB guarantee rankings, traffic or AI mentions?

No. It records checks and provider answers within a defined scope. Search and AI systems control their own crawling, indexing and responses. A passed page check or creative proof is not a performance guarantee.

Why do JavaScript sites sometimes have crawler visibility issues?

Many modern sites send a small HTML shell first, then load the real content in the browser after JavaScript runs.

That can be fine for human visitors, but some bots may only receive the first HTML response. If that response lacks headings, copy, links, and metadata, the page can be harder to understand.

What do crawlers usually need to see?

They need a useful title, meta description, canonical URL when appropriate, visible page copy, headings, internal links, and enough context to understand what the page is about.

For JavaScript-heavy sites, those signals may exist visually in the browser but be missing from the raw HTML response.

How does prerendering help?

Prerendering loads the page in a browser-like environment, waits for the JavaScript-rendered content, and returns readable HTML to crawler-style requests.

The goal is not to redesign the site. The goal is to make the content that humans already see available in a crawler-readable response.

Which crawlers matter?

Important crawlers can include Googlebot, Bingbot, Applebot, GPTBot, ChatGPT-User, ClaudeBot, PerplexityBot, and link preview bots.

Different bots behave differently, so the practical test is to compare raw HTML, rendered HTML, and crawler-style requests for the actual page.

What setup options are available?

Prerender Buddy supports two main paths: managed proxy setup for no-code and AI-built sites, and developer setup for teams that can add middleware, edge logic, or server-side routing.

The right path depends on where the site is hosted, who controls DNS, and whether the team can safely add server-side code.

What is the no-code setup?

The no-code setup routes a public hostname through Prerender Buddy so crawler requests can receive rendered HTML while visitors continue to reach the original website.

This can fit hosted builder sites where editing server middleware is not practical. Test first: newer Lovable-hosted projects may already use server rendering, while older hosted projects may receive Lovable-managed crawler rendering.

What is the developer setup?

The developer setup uses middleware, edge functions, or server-side routing to detect crawler requests and request rendered HTML from Prerender Buddy.

This is usually a better fit for Vercel, Netlify, Cloudflare, custom Node apps, and teams that can safely keep API keys server-side.

Do normal visitors still see the original site?

Yes. The intended behavior is that normal visitors keep using the live app. Prerender Buddy focuses on crawler-style page requests.

Assets, API routes, non-GET requests, and normal visitor traffic should not be routed through the rendering path.

Can I test before changing DNS or middleware?

Yes. Start with the crawler visibility checker or bot view checker to compare what the browser shows with what crawler-style requests receive.

Testing first helps avoid changing DNS or middleware before you know whether the site actually has a crawler-readability problem.

Is there a free plan?

Yes. The Free plan supports one website with basic visibility, JavaScript, rendered-content, robots.txt, llms.txt, Health, and recent Search & AI Visits checks. Ongoing monitoring and AI answer tracking require an eligible paid plan. See Pricing for current limits.

What counts as a render?

A render usually means Prerender Buddy has to create or refresh rendered HTML for a page request instead of serving an already cached result.

Cached responses reduce repeated rendering work. The dashboard usage view is the best place to monitor current render usage.

What happens when I reach a render limit?

Behavior depends on the integration and whether cached content is available. Review the failure-behavior reference and your active account limits.

Do you offer custom or enterprise plans?

Yes. Custom plans are available for larger websites, agencies, or teams that need more control, higher capacity, priority support, and flexible rendering limits.

Contact support with the number of sites, estimated traffic, crawler needs, and setup type so the right plan can be mapped.

Does this help AI crawlers?

It can help with the technical readability layer. If an AI crawler or agent requests a page and only receives a thin JavaScript shell, it may miss the actual content.

Serving readable HTML gives crawlers a better chance to evaluate headings, copy, metadata, and links. It does not guarantee that an AI system will mention, cite, or recommend the page.

Why does readable HTML matter for AI agents?

AI agents and crawlers still need source content they can retrieve and parse. If the first response hides important content behind JavaScript, the page may be harder to summarize or understand.

Readable HTML is a foundation. Clear content, accurate metadata, useful pages, and crawlable links still matter.

Should I create llms.txt too?

An llms.txt file can help describe important pages for AI systems that choose to read it, but it does not replace readable page HTML.

Use it when a maintained guidance file helps; it is optional and does not replace crawler-readable content.

Why is HTTPS pending?

HTTPS validation can depend on DNS records, certificate validation records, CNAME targets, proxy status, and propagation timing.

If DNS is active but HTTPS remains pending, check whether the exact TXT or validation records are present at the authoritative DNS provider, then refresh status after propagation has had time to settle.

Why does DNS validation take time?

DNS changes can take time to propagate across resolvers. The authoritative provider, TTL, registrar settings, and previous records can all affect timing.

If a domain was recently moved from one DNS provider to another, make sure records are being added at the current authoritative DNS provider, not the old one.

What if my root domain and www behave differently?

Root and www are separate hostnames. One can be configured correctly while the other still points elsewhere or redirects incorrectly.

Check both versions. Decide which hostname should be canonical, then either redirect the other hostname or add both as separate managed configurations. A managed root/www pair shares one website slot, although each direct hostname still needs its own DNS, HTTPS, and verification steps. Other subdomains count separately.

What if my site is built with Lovable, Bolt, Vite, React, or v0?

Crawler readability depends on the exact framework, project generation, hosting environment, and production response. Newer Lovable-hosted projects may use server rendering, while legacy hosted projects may receive platform-managed crawler rendering; exported Lovable projects do not necessarily keep that behavior.

Use the bot view checker or raw-vs-rendered comparison first. If the production response is thin and the rendered version contains the real page content, add SSR, static generation, or a prerendering layer that fits the deployment.

Start with the docs or run a free crawler visibility check.

Read setup guide Check bot view

What does AI Visibility monitoring measure?

It records answers to selected questions through configured provider APIs, including brand appearances, recommendations, competitors and source relationships. It does not read people's private chats or measure every answer a consumer app might produce.

Why does a manual ChatGPT answer differ?

Wording, earlier chat context, model/search settings, location, timing and ordinary answer variation can differ. Compare exact questions and dated evidence before treating the difference as an error.

Is every returned source a citation?

No. Explicit answer citations, returned sources and searched sources are different relationships. A source can be returned without being cited in the final answer.

Why is a competitor shown when I did not add it?

The competitor view distinguishes tracked brands from possible competitors detected in the available evidence. Review a candidate before tracking it.

Is page monitoring the same as AI prompt tracking?

No. Page monitoring checks configured public URLs and records health findings. AI tracking collects answers to selected questions. Their schedules and allowances are separate.

Does a crawler visit mean my page was indexed or cited?

No. A recorded request and response describe delivery evidence. They do not establish indexing, training use or an AI answer citation.

Does a suggested Health fix update my website?

A suggestion or generated discovery file is a draft. Review it, publish the approved change on the appropriate host and rerun checks.

Does scheduling publish an article?

No. A calendar date organizes the draft. Review and publish through your editor or CMS, then verify the live page. A PB status label alone is not proof of publication.

Can I send a complete agency report from the checker?

The shareable checker report is a dated public technical summary with an expiry. A complete agency report needs broader question, competitor, source, action and business context. See the client reporting framework.

Can the Agent change my settings or publish work?

The current Agent helps interpret product knowledge and scoped evidence. Do not treat its advice as confirmation of a setting change or publication. Complete and verify the action in the relevant workflow.

What happens to usage when I delete a chat?

The conversation leaves your history, but already-used Agent messages are not restored. Deletion does not reset the usage window or remove the underlying site evidence.

Can my external assistant access workspace evidence?

Eligible accounts can configure a scoped key in the local PB MCP process. The tools return bounded read-only summaries; public diagnostics work without a key.

Why is a feature locked?

Access can depend on the effective plan, trial allowance, add-on, provider configuration or release eligibility. Read the reason shown for the account. A locked feature is not automatically a broken integration.

Why are my trial limits different from the plan offer?

An eligible current-catalog trial can have its own allowance before the selected paid plan starts. Compare the active account limits with the trial terms and checkout offer. Ask support about a mismatch rather than assuming the plan's full allowance applies during the trial.

Does my legacy subscription include the new AI features?

Not automatically. A familiar plan name can refer to a different catalog and entitlement set. Existing subscribed terms and applicable overrides must be checked for the account.

When does an add-on become available?

An eligible add-on activates after successful payment and account confirmation. Review its active status and displayed allowance. Buying one capacity type does not increase every product limit.

Does PB publish ads?

The inspected PB workflow prepares creatives, checks and handoff/export. Campaign settings, review and activation remain in the advertising provider's tools. A saved creative or export is not a live ad. See the public AI Ads workflow overview, or sign in for the step-by-step ChatGPT Ads setup guide.

Does the AI Ads add-on grant access to ChatGPT Ads?

No. OpenAI controls advertiser access and verification. PB's add-on provides eligible PB functionality; it cannot grant provider account access.

Are ad spend and credits the same?

No. PB credits pay for applicable PB operations such as generation and proof. The advertising provider bills media spend separately.

Does a Passed proof mean an ad is approved?

No. It means the completed PB checks passed for that creative version. It does not establish legal clearance, provider approval or performance.

Does zero in Ads conversions mean nobody converted?

Not necessarily. The inspected importer does not currently populate conversion reporting. Treat an unsupported field as unavailable and verify conversions in the provider's measurement system.