AI visibility audit (preview)
What does ChatGPT say about your brand?
Find out in about 30 seconds. BotCite scores 10 buyer-intent prompts for Visibility, Position, Sentiment and emails you the report. No credit card, no sales call. This preview runs the clearly-labeled demo engine; live ChatGPT and Perplexity runs begin when the engine milestone ships.
- Demo engine (ChatGPT + Perplexity at launch)
Preview mode: runs execute on the clearly-labeled demo engine. Two live engines at launch — stated plainly, never padded.
The product is the proof
This is the report you get
A full audit of a fictional CRM brand, so you can judge the report before you run your own. Synthetic data, labeled as such — real public-brand reports replace it as audits come in.
One weighted number over the three category-standard metrics — layered on top of them, never replacing them.
Share of voice — "best CRM" prompt set
Top cited sources across all answers
- g2.com/categories/crm9 citations
- reddit.com/r/sales — "best CRM for a 10-person team?"6 citations
- hubspot.com/blog/crm-comparison5 citations
- capterra.com/crm-roundup2 citations
- pipedrive.com/docs2 citations
The vocabulary you already use
Visibility. Position. Sentiment.
The category-standard metrics, measured the standard way — so your numbers are comparable to whatever you track today.
Visibility
The share of answers that mention your brand at all. If your buyers' questions come back without you, nothing else matters.
measured: mentions / total answersPosition
Where you appear when you do — first recommendation or a trailing "also consider." Engines rank; the top slots get read.
measured: avg. first-mention rankSentiment
How the answer frames you: recommended, merely listed, or warned about. Judged in context, per mention — not keyword matching.
measured: LLM-judged, per mentionBotCite Score
One weighted 0–100 number over the three metrics above, so you can track a trend and report a single figure upward. It layers on top of the standard vocabulary — it never replaces it, and every report shows the components.
AI search was a black box
Your buyers ask ChatGPT which tools to shortlist, and until now you found out from lost deals. Measured weekly, it becomes ground you can take and hold.
- You don't know which prompts your buyers actually ask
- You don't know if engines recommend you — or a competitor
- You don't know which sources engines trust for your category
- You can't tell whether last quarter's content moved anything
- A buyer-intent prompt set for your vertical, tracked on a schedule
- Every mention with its Position and Sentiment, per engine
- The exact cited sources — your next content targets, ranked
- Share of voice vs. named competitors, trended week over week
Audit. Track. Fix.
Start with a free snapshot. Keep it measured. The fixing part is in development — and we'll say so until it ships.
Audit
Type your brand name. About 30 seconds later, 10 buyer-intent prompts have run through ChatGPT and Perplexity and your scored report is on its way to your inbox.
Track
Turn the audit into a baseline: 50–150 prompts refreshed weekly (daily on Pro), with a digest email that tells you what moved and which competitor moved it.
Fix In development
Gap analysis that drafts the FAQ, schema, and content changes most likely to earn citations — then re-measures the delta. Coming after launch; not a live feature today.
Radical honesty, in table form
How we compare — including where we don't win
Peec and Otterly are good tools. Here's the honest matrix, so you can decide with the same data we'd want.
| Feature | BotCite | Peec AI | Otterly.AI |
|---|---|---|---|
| See your numbers before paying | Free audit, ~30 secondsemail-gated report, no sales call | Trial | Free trial |
| Entry price | $79/mo | €70/mo | $29/mo |
| Prompts at entry | 50 | 50 | 15 |
| Engines at entry | 2 — ChatGPT + Perplexityfewest in this table; Gemini + Claude arrive on Pro | 3 models | 4 engines |
| Refresh at entry | Weeklydaily on Pro — they're ahead here today | Daily | Daily |
| Unlimited seats | Every paid tier | Yes | Yes |
| Prompt-volume dataset | No — deliberatewe meter the prompts you choose, not volume estimates | Yes | Yes |
| Multi-market / language | Not yet | Yes | Yes, 50+ |
Competitor figures pulled from public pricing pages, August 2026. Where a competitor is better, this table says so — that's the point.
Pricing that starts with proof
Run the free audit first. Pay only after you've seen your own numbers.
Free
See your own numbers first. The audit is the demo.
one audit, no card
- 1 full AI visibility audit
- 10 buyer-intent prompts × 2 engines
- Site findings — why AI can't read your site
- Visibility, Position, Sentiment + BotCite Score
- Shareable report link
Starter
Your audit becomes a baseline you track.
launch pricing — metered on prompts, not people
- 50 tracked prompts, managed from your dashboard
- 2 engines — ChatGPT + Perplexity
- Weekly tracking + delta digest email
- Competitor share of voice over time
- History and trends from day one
- Unlimited seats — bring the whole team
Pro
For teams and agencies that need daily reads.
launch pricing — metered on prompts, not people
- 150 tracked prompts, managed from your dashboard
- 4 engines — adds Gemini + Claude
- Daily refresh + weekly delta digests across engines
- Competitor share of voice on every engine
- CSV export + priority support
- Unlimited seats — clients and colleagues included
Unlimited seats on every paid tier. Pricing meters tracked prompts — never teammates. Every plan starts with the free audit; upgrade from your own report, self-serve via Stripe, cancel anytime.
Questions senior marketers ask us
How is this different from SEO rank tracking?
Rankings don't exist inside an AI answer — there's one response, a few named brands, and a handful of cited sources. BotCite measures that surface directly: whether you're mentioned (Visibility), where (Position), how you're framed (Sentiment), and which URLs the engine trusted. Your rank tracker can't see any of it.
Which engines do you cover?
ChatGPT and Perplexity on Free and Starter. Pro adds Gemini and Claude. That's the whole list — we publish coverage plainly instead of padding a logo grid, and every adapter reads the engine's native citation metadata rather than scraping.
Do you track brand visibility or source visibility?
Both, separately. Brand visibility is whether the answer names you. Source visibility is whether your domain is among the cited URLs. The gap between the two — talked about, but described from other people's pages — is usually the most actionable finding in the report.
AI answers change run to run. How is this reliable?
It isn't — per single answer. That's why we track a prompt set on a schedule and report rates and trends, not one-off responses. A brand mentioned in 55% of answers this week and 70% next month is a real signal even though any individual answer varies.
Where do the prompts come from?
From a buyer-intent battery for your vertical — the comparison, "best X for Y," and alternative-seeking questions your buyers actually ask. On paid tiers you can edit the set and add your own; prompt count is the only thing we meter.
What's the catch on the free audit?
We ask for your work email to deliver the report — that's the entire gate. No credit card, no trial clock, no sales call, one audit per brand. We're betting that seeing your own numbers is a better pitch than anything we could write here.
Your buyers are already asking.
Ten prompts, two engines, one scored report — in the time it took to read this page.
Preview free auditFree · no credit card · report by email