One AI Credit equals one assistant action in Workspace CMS, charged against your plan’s monthly allowance. If you’re consistently bumping into your allowance we’ll reach out about moving you to a plan that fits — you can also set a monthly cap so AI actions pause instead of accruing more charges. No surprise overage on your invoice.
Two separate meters. AI-visibility prompts (the per-plan allowance the AI Visibility Tester draws against, used to check whether your pages appear in answers from ChatGPT, Claude, Perplexity, and Gemini) and AI Credits (the assistant-action pool scheduled below) are independent monthly pools. AI Visibility tracks 5 / 15 / 30 prompts per tier on a monthly / bi-weekly / weekly cadence; AI Credits give you 200 / 1000 / 2,500 assistant actions per month. Visibility checks never draw down the credits you spend on blog drafts, meta rewrites, audits, or any of the assistant actions in the schedule below.
Per-action cost
| Assistant action | Credits |
|---|---|
| Blog post draft generation | 2 |
| Meta title + description rewrite | 1 |
| Image alt-tag suggestions (per batch of 10 images) | 1 |
| Content rewrite (lengthen / shorten / paraphrase) | 1 |
| Content audit (full-site run) | 5 |
| Brand-voice training run | 10 |
Monthly allowances by plan
| Plan | AI Credits / mo | Approx. blog drafts* |
|---|---|---|
| Essentials | 200 | ~100 |
| Managed | 1000 | ~500 |
| White-Glove | 2,500 | ~1250 |
*Pure-blog-draft equivalent. Real-world usage is a mix — drafts, meta rewrites, alt-tag sweeps, audits — so the practical capacity is closer to twice the blog-only figure.
Allowances & caps
- Credits reset each billing cycle and do not roll over.
- You’re notified at 80% and again at 100% of allowance.
- If you’re consistently bumping into your allowance, our account team will reach out about moving you to a plan that fits your usage — no surprise overage line items on your invoice.
- You can set a per-month soft cap in Settings → Usage; once you hit it, AI actions pause until the next cycle or you raise the cap.
Why credits, not minutes? Frontier AI models vary in latency by orders of magnitude. Charging per credit (a fixed-cost action) keeps the unit economics predictable for you and for us — no “your audit took 47 minutes” surprise.