Pricing
Pick the engines you actually care about and the price moves with them. Every plan measures the same way, with the same accuracy: the cheaper tiers track fewer prompts, they do not track them less honestly.
Priority slot packs
+10 fast-lane prompts · $25/mo per pack
Tap the engine marks on any card · prices update everywhere
Every tracked prompt is measured on every engine you select, checked weekly, and promoted to daily for a week the moment your brand enters or leaves an answer. Priority prompts skip the wait: they run at each engine’s real change speed all month, ChatGPT daily. Same measurement, same honest ranges on both. Priority buys speed, never truth.
Starter
One business, the full instrument.
Billed monthly
Start free- 30 tracked prompts · 5 priority
- 1 business
- 3 seats · Pulse 20/mo
Includes
- Site audits and fix drafts
- Alerts by email and Slack
- Every engine, same accuracy
Growth
The hero tier. Their $125 buys the API surface; this buys the real one.
Billed monthly
Start free- 100 tracked prompts · 8 priority
- 2 businesses
- 10 seats · Pulse 150/mo
Includes
- PROVE receipts, CI-gated
- Revenue attribution (GA4 + GSC)
- Ranked fix queue, drafted for approval
- CRM push (HubSpot, Salesforce)
Scale
Portfolio and franchise variance.
Billed monthly
Start free- 300 tracked prompts · 25 priority
- 5 businesses
- 25 seats · Pulse 500/mo
Includes
- Franchise variance rollup
- White-label reports
- Competitive report
Pro
The platform tier.
Billed monthly
Start free- 600 tracked prompts · 60 priority
- 10 businesses
- 50 seats · Pulse 1000/mo
Includes
- Longest history
- Priority support
- Bulk export
- Grok, measured on native xAI Live Search
- Daily cadence on every prompt, priced to contract
- Custom prompt volumes and caps
- Dedicated support and onboarding
Compare every plan
Accuracy is identical on every row of the first group. What you buy as you climb is scope and the outcome layer, never a better measurement.
| Capability | Starter $49 | Growth $119 | Scale $349 | Pro $699 | Custom Custom |
|---|---|---|---|---|---|
| Measurement & accuracy | |||||
| Engines included | Base 3 + selector | Base 3 + selector | Base 3 + selector | Base 3 + selector | All, incl. Grok |
| ChatGPT measured on chatgpt.com | |||||
| Priority prompts (per-engine cadence, ChatGPT daily) | 5 | 8 | 25 | 60 | Custom |
| Watch lane (weekly + change burst) | |||||
| Honest ranges on every rate | |||||
| Hard spend guard (a runaway month is impossible) | |||||
| Pulse copilot questions / mo | 20 | 150 | 500 | 1,000 | Custom |
| Coverage | |||||
| Tracked prompts (pooled, org-wide) | 30 | 100 | 300 | 600 | Custom |
| Businesses tracked | 1 | 2 | 5 | 10 | Custom |
| Countries & regions (a localized prompt uses one slot) | |||||
| Team seats | 3 | 10 | 25 | 50 | Custom |
| Read-only logins for clients (free, never a seat) | |||||
| Local & franchise | |||||
| Per-location AI answers | |||||
| Local Maps grid (pin-by-pin) | — | ||||
| JSON-LD schema generator | |||||
| Franchise variance rollup | |||||
| Outcomes & actions | |||||
| Get Cited targets + contacts | |||||
| Ranked fix queue | |||||
| Revenue attribution (GA4 + GSC) | — | ||||
| PROVE receipts (CI-gated) | |||||
| White-label reports | — | — | |||
| Platform & support | |||||
| Data exports (CSV) | |||||
| API + MCP access | — | ||||
| SLA + dedicated CSM | — | — | — | — | |
| Support | Priority | Dedicated | |||
Every plan measures the surfaces buyers actually read, multi-run with honest ranges, the source graph, the copilot, drafted fixes, and CI-gated PROVE that ties a citation to the revenue it moved. Unlimited countries & regions on every plan: a localized prompt uses one pool slot. Spend is bounded by a hard per-tier engine allowance, so a runaway month is impossible by construction.
What every plan includes, at any price
- Three rates per prompt and engine (named, cited, supported), each with the number of runs behind it and the range it could honestly be. No plan buys a surer number than the data supports.
- Developer APIs measured and reported on their own, never averaged into a headline with the consumer apps your buyers actually use.
- Movement reported only when the ranges stop overlapping. A difference that could be noise is labelled as normal variation on every tier.
- Anything unmeasured renders as a dash, never as a zero.
The rules behind those are on the methodology page, and the enforcement page names the test that fails if we break one.