Plans and quotas
What each plan includes and how the quotas are enforced. Pooled prompts, engine slots, the seat cap, cadence, and the one action that frees a prompt slot.
Your plan sets how much you can track: how many brands, how many questions, how many teammates, and how often MentionFlow samples the engines.
Every limit is counted at the workspace level and pooled across all your brands. So a 150-prompt plan is 150 prompts to spread however you like, not 150 per project. For current prices, see the pricing page.
Limits by plan
The individual tiers, plus the free plan a workspace sits on before (and after) a subscription:
"Free" here is the plan a workspace holds when it has no subscription, and it never expires. It is not the 7-day trial, which is a real subscription costing $1 for its first week (see Billing and plans). A workspace on the trial is genuinely on Starter or Growth, with those limits, from the moment it subscribes.
| Limit | Free | Starter | Growth | Advanced |
|---|---|---|---|---|
| Brands | 1 | 1 | 2 | 5 |
| Prompts (pooled) | 50 | 50 | 150 | 350 |
| Team members | 3 | 2 | 2 | 5 |
| Content drafts / month | 5 | 5 | 15 | 30 |
| Engine slots | 4 | 4 | 4 | 4 |
| Collection cadence | 3 times/week | Daily | Daily | Daily |
| Agent (crawler) analytics | — | — | Yes | Yes |
| Site health | — | — | Yes | Yes |
| Claude engine (weekly) | — | — | Yes | Yes |
| Grok engine (weekly) | Yes | Yes | Yes | Yes |
The white-label agency tiers all come with unlimited brands plus pitch projects, daily cadence, agent analytics, site health, and the weekly Claude and Grok swaps:
| Limit | Agency Starter | Agency Growth | Agency Advanced |
|---|---|---|---|
| Prompts (pooled) | 80 | 200 | 400 |
| Team members | 5 | 10 | 10 |
| Content drafts / month | 5 | 15 | 30 |
| Engine slots | 4 | 4 | 4 |
How the prompt quota works
The prompt count is checked whenever a prompt would be created: manual add, CSV import, accepting a suggestion, promoting a fan-out, or research. Two states count against it, active and paused. Only one action frees a slot:
Archiving a prompt frees its slot. Pausing does not. A paused prompt still holds its slot. This is deliberate. If pausing freed slots you could pause 50, add 50, and resume to run 100 on a 50-prompt plan. Archive to make room. Archived prompts keep their history and can be restored or re-imported.
Bulk operations (CSV import, suggestion accept) take a per-workspace lock and re-check the quota inside the write transaction, so two imports at once can't both slip past the last slot. Imports are all-or-nothing. If the batch doesn't fit, nothing is written.
Engine slots
Every plan runs four engine slots. Your workspace picks a default set of four that every project inherits, and any project can customize its own four within those same slots, on Collection. Which engines you can swap in depends on your plan:
- Every plan starts with the four defaults (ChatGPT, Perplexity, Gemini, and Google AI Overviews) and can swap in AI Mode, Copilot, or Grok.
- Growth and above, including every agency plan, can swap in any supported engine. That adds DeepSeek, Mistral, and Claude.
- Claude and Grok are weekly engines. Swapped in, they collect on Mondays (UTC). Claude costs roughly 40 times as much per answer, which is why it stays weekly. There is no way to move either to daily collection.
- The slot count is workspace-wide and billed once. Tracking the same engine, or a different four, in another project never costs an extra slot. A project you haven't customized follows the workspace default, and a Reset clears its override so it inherits again.
Cadence
Cadence is how often projects collect:
- 3 times a week runs on Monday, Wednesday, and Friday (UTC).
- Daily collects every day.
First-run banners and empty states match your real cadence. You are never told to expect data on a day you won't collect.
Content drafts
The AI content brief and draft generator is metered per UTC calendar month across the workspace, and resets on the first of the month. Draft generation and brief regeneration draw from the same allowance. The meter is checked before the paid model call.
Seats
Team seats are capped per plan (see the table). Invites are single-use, expire after seven days, and never grant the owner role. The last owner of a workspace cannot be removed or demoted. Read-only client viewer seats let you share dashboards without edit rights. See Workspace and members.
What is not plan-gated
- The v1 API and MCP server are gated by API key scope, not plan. Any workspace with a full-scope key can read its data.
- The Actions board and content briefs are on every plan. Briefs are bounded only by the monthly draft meter.
- Every market is available on every plan. You can track up to five per project, from 241 countries. Each market a prompt is asked in uses one of your daily prompt checks.