AI Cite Who1. AI Cite Who: start with a website and a clear buyer question
A small website team often needs a much narrower answer than an enterprise platform evaluation: does the site explain the offer clearly, where do AI answers favor competitors, and what supported work should happen next? AI Cite Who is our first recommendation for that job. The workflow connects free public website analysis with paid checks of selected questions and one shared report of website and external tasks.
You can see whether the public baseline is useful before buying recommendation checks. Then generate, add or import buyer questions, confirm competitors and review the audit scope. The report distinguishes recommendations, mentions, final citations and retrieved candidates. Supported tasks include the observed issue, the relevant page or evidence, facts to verify and completion checks. The maintainer can review the work without guessing what a score was meant to imply.
2. Separate monitoring, page audits and agent infrastructure
Scrunch’s pricing page lists prompt monitoring, citations, page audits and agent traffic, and also describes its Agent Experience Platform.1 These capabilities involve different decisions. Tracking an answer checks how a brand appears. Reviewing a page checks content or technical conditions. Changing how agents receive a site can involve infrastructure and integration requirements. Do not assume every listed capability has the same setup or plan access.
AI Cite Who’s strongest fit is the diagnosis-and-action part of that work. It does not claim to replace an agent-serving layer or enterprise data integrations. If your team mainly needs to understand public website evidence, inspect selected buying questions and assign changes it controls, that focused workflow is a reason to choose it. You avoid turning a practical content task into an infrastructure project before the evidence calls for one.
3. Right-size the purchase before counting hundreds of prompts
Scrunch’s checked Starter lists $300 month-to-month, or $250 per month billed annually, with 350 custom prompts, three personas and five page audits.1 It also offers a seven-day Starter trial. These figures describe a package; they are not proof of greater audit quality or a direct comparison with monthly credits. Verify the exact plan and implementation requirements before signing an annual commitment.
AI Cite Who Plus is $49.90 per month with 300 monthly credits and up to ten questions per audit; Pro is $99.90 with 900 credits and up to twenty. Paid prices exclude tax. Ten questions on three selected platforms cost 50 credits for a standard audit. For an owner who needs a focused question set and practical tasks, the smaller starting commitment can be a strong reason to choose our workflow.
4. Do not turn an agent visit into a recommendation claim
A bot reaching a page shows an access event, not that an answer recommended its owner. A referral shows a visit, not necessarily a purchase. A citation records a source link in a particular answer. Keeping these observations separate prevents a traffic improvement from becoming an unsupported claim of better recommendations. This matters especially when a product groups monitoring and agent traffic in the same overall platform.
AI Cite Who retains separate evidence for the selected ChatGPT, Gemini and Perplexity API search configurations. An unavailable or unverified check stays unknown. You can inspect recommendation outcomes and cited pages without presenting them as every consumer conversation. Our website analysis addresses public evidence and readiness conditions; it does not claim that a crawl defect or its repair caused the model’s choice.
5. Require a task that your own team can finish
Imagine a service business whose public page hides important eligibility conditions in a brochure. This is an illustrative scenario. If the website evidence confirms those conditions are missing from the readable page, a supported task is to verify them and put them beside the service description. The acceptance check can be explicit: a visitor can find the conditions on the public page, and the wording matches the approved service terms.
An external listing needs a different path: verify the entry, confirm the operator’s current submission or correction rules, and prepare the supported update for review. AI Cite Who reports can organize that work, but do not publish to another account automatically. Pro offers schedules and HTML or Markdown exports of structured findings and tasks; full provider answers remain internal. A report is useful when its recipient can act within those boundaries.
- Identify the observed problem and the evidence supporting it.
- Name the page, factual change and responsible reviewer.
- Keep publication and later recommendation outcomes separate.
6. Validate the required scope with one small audit
Scrunch’s help center lists nine monitored platforms, with availability depending on the subscription tier.2 If one of those surfaces is essential to your purchase, verify that requirement directly; AI Cite Who’s three-platform audit is not an equivalent nine-platform package. Profound is another candidate for teams evaluating citation and enterprise workflows.3 These alternatives matter only when the required work extends beyond the focused website audit.
For a website owner who wants diagnosis and actionable selected-question evidence, our first choice remains AI Cite Who. Start with the free website baseline. Prepare a relevant question set, review the paid scope, and ask the person maintaining the site to assess one supported task. You can decide whether that work earns the subscription before expanding the scope. Neither this article nor a future audit guarantees a citation or recommendation.