Account and Settings

The Account page is where you manage your Testzilla account: your credit balance and plan, billing, and your API key. This page also explains the Evaluation model that judges your tests. Open the Account page from the Account entry at the bottom of the left nav in the web app.

What’s on the page

• Email and balance — your sign-in email and current credit balance.
• Billing row — plan, credits, and billing controls sit together in a three-column row: your current subscription plan with a Change Plan expander that opens the plan options in place, credits remaining and renewal date, and Transaction History plus Manage Billing.
• API Key — your account API key for API and MCP access (see Authentication).
• Consent and sign-out — your marketing and communication consent choices, followed by Log Out, live at the bottom of the page.
• Delete Account — permanently removes your account (your balance must be zero first).

The Evaluation model that decides each test’s verdict is covered below.

Subscriptions and the run gate

Running tests is subscribe-first: an active subscription is required before Testzilla will start a billable run. You can create projects, channels, and tests, and explore the app, without a plan — but the first run is gated.

When you try to run without an active subscription (or with your free credits used up), Testzilla shows a run-gate dialog rather than a dead-end error:

• Not subscribed — the dialog is titled “Subscribe to run tests” and leads with Choose a plan. Add-credits is not offered here; a subscription comes first.
• Subscribed but out of included credits — the dialog is titled “Top up or upgrade” and offers Upgrade plan or Top up credits (a one-off credit purchase).

Either way you can dismiss it with Not now. Over the API, the same block surfaces as HTTP 402 and the response carries the gate’s reason, which begins Subscribe to start running tests. Subscriptions are managed in the web app only — there is no API or MCP call to start, upgrade, or cancel one.

Outbound phone tests are a paid-membership feature. Running an outbound test (your agent places the call to a dedicated Testzilla number) requires an active paid plan; inbound phone, chat, and web tests run on any active subscription with credits.

Top-up rates follow your plan. When you buy extra credits with Top up credits, the rate you pay per credit is set by your subscription tier. The current rate for your plan is shown in the top-up dialog.

Sharing with a team?

Credits and billing belong to the account you are working in. If you run inside a Workspace, the workspace’s own plan and balance apply, not your personal account’s.

Evaluation model

Every test you run is judged by an LLM that reads the transcript and returns a verdict — pass, fail, or uncertain — against your test’s criteria. The Evaluation model is the model that does that judging.

By default, Testzilla evaluates with Claude (from Anthropic). This is a platform-managed setting, not a per-account toggle on the Account page: the evaluator is configured by Testzilla and applies consistently across accounts. Changing it does not change your tests, criteria, or how results are displayed — only which model produces the verdict.

Synthetic-data-only

Testzilla runs against synthetic test data, so the default evaluator uses Anthropic’s general (US) Claude API. If your organization needs evaluation traffic to stay in-region — for example running Claude on AWS Bedrock in the EU (Zero-Data-Retention-capable) — that can be arranged with Testzilla for your environment rather than switched on self-serve. The Bedrock region is configurable per account, so we set the region that matches your data-residency needs (eu-west-1 is the EU default). An OpenAI evaluator (GPT-4o) is also available as an alternative provider. Talk to us if you have specific data-residency or compliance requirements before relying on a particular routing.

Storage and overage billing

Your plan includes a storage allotment for published test pages and files (see File Manager for usage). Storage is a soft quota: going over your allotment never blocks an upload outright. Instead, the excess is metered daily and billed in credits at a per-GB-per-month rate:

• Testzilla samples your storage once a day; each day you are over quota adds that day’s excess to the month’s tally.
• A grace period applies: a brief spike over quota costs nothing, and billing only starts once you have been continuously over quota past the grace window.
• At the end of the month, the accrued overage converts to a credit charge, and it appears as a storage_overage line in your Transaction History.
• The first time your account crosses into overage, Testzilla emails you once so you can trim files or upgrade before any billing starts.

Universal overage cap. All metered overage on your account, storage included, is bounded by a single overage cap: once your overdraft reaches the cap, further billable work pauses until you top up. The cap has a per-tier default, and we can adjust it for your account on request.