Pay for what your team actually uses.

Yes. Every new account starts with a 14-day free trial (or 100 credits, whichever comes first) — no credit card required. You get full access to the Team plan features so you can run real knowledge capture sessions and see what Clara can do for your team. At the end of the trial, pick a plan to keep going or simply let it lapse.

Yes. Committing to an annual plan gives you 10% off the monthly platform fee across all tiers. Your full year of included credits is provisioned upfront as a single bundle, so you can consume credits freely across months rather than being capped each calendar month. Heavy processing months and lighter months balance out automatically within the same subscription.

Your plan's monthly fee includes a bundle of active users and credits. Two things draw from that pool:

• Active users: 10 credits per active user per month.
• Document processing: ~0.06 credits per page (varies by complexity).

The bundle is flexible: if you use fewer active users than your max, the unused credits roll into your page budget. For example, on the Team plan, 5 active users consume 50 credits, leaving 25 credits for roughly 400 additional pages.

When you exceed your included credits, overage is metered at your tier's single rate. One tier, one rate, no surprises.

A typical project has three kinds of active users:

• Owners: Work directly with Clara, assign and manage interviews, edit artifacts and knowledge outputs. Usually 1 to 3 people per project.
• Contributors: Take interviews with Clara and contribute to knowledge capture. The most common role for domain experts and team members.
• Users: Call the MCP to leverage captured knowledge inside their own agents or workflows. Typically the largest group: developers, analysts, or anyone consuming the knowledge downstream.

Processing cost depends on document complexity. Parsing is fixed; extraction and tagging vary. Indicative ranges:

Indicative ranges. Your real usage is metered per-job in the platform with a per-document breakdown. You can cap, alert, or pause runs from inside the app.

No. When external AI clients (Claude, ChatGPT, Copilot, etc.) access Clarifeye via MCP, no credits are charged. Those interactions count only as active users. You pay those providers directly for their token usage. Credits only apply when AI usage runs through the Clarifeye harness: Clara, Playground, or agentic workflows we host.

If you stop your subscription, you can export everything you built on Clarifeye: knowledge artifacts (briefs, playbooks, mental maps), source data, structured extractions, and version history, in open formats. No proprietary lock-in. It's part of what the platform fee buys.

A Forward Knowledge Engineer is a senior Clarifeye engineer who embeds with your team. They co-design the first knowledge capture sessions with your subject-matter experts, set up the right processing pipelines, and stay close to your stack until you're self-sufficient. Available on the Enterprise plan.

It's a ballpark, not a quote. The estimate uses average processing ratios. Real usage is metered live in the platform, you see per-document, per-run costs and can set budgets and alerts. We'd be happy to put together a tailored quote based on a sample of your actual documents.

No. Your knowledge is never used to train models. Ever. Your IP stays your IP, on every plan.