Per seat, per conversation, or flat rate: which live chat pricing model is safest?

The short version

• Per-seat pricing is predictable when conversation volume changes but team size stays small.
• Per-conversation or per-ticket pricing can suit a larger team with controlled volume.
• Workspace pricing can be predictable, but plan-level seat, contact, and feature limits still matter.
• AI-resolution pricing is only valuable when the resolution and customer outcome can be verified.
• Always model overages and a peak month before choosing.

Per-seat pricing

The bill rises when more teammates need access. This is easy to forecast for a founder or tiny support team, but it can discourage inviting specialists who occasionally need context or approval.

Check whether observers, administrators, mobile users, or occasional responders consume a paid seat.

Per conversation or ticket

The bill rises with customer contact. Read what creates the charge: a human reply, automated rule, AI answer, reopened thread, or channel message. Gorgias and Tidio publish different definitions, which is why the headline allowance is not directly comparable.

This model can become less predictable during promotions, delivery incidents, or seasonal peaks, exactly when support is busiest.

Workspace and AI pricing

Workspace plans can include several seats for one flat subscription while placing limits on contacts, integrations, retention, or advanced features. Crisp currently describes this approach on its pricing page.

For AI, distinguish a generated response from a resolved request. Count the extra ticket fee, AI fee, human handoff, and whether the promised outcome was actually completed.

Model three months before buying

Quiet month: current seats and low conversation volume
Normal month: expected team and contact volume
Peak month: promotion, incident, or seasonal surge
For each: base + seats + usage + AI + integrations + overages

FAQ

Recommended next steps

FAQ