Limits and defaults

This page collects the limits and defaults you meet in Testmode, such as how long a test may run, how long an invitation lasts, how many addresses a schedule can email and how large a file the Assistant accepts. Each table links to the page that explains the feature.

For what kinds of apps and flows Testmode can test at all, see What Testmode can and cannot test.

Limit or default	Value
Test time limit	15 minutes by default. Set 1 to 60 whole minutes per test case with Maximum duration (minutes), or for the whole project with Default test duration (minutes) under Settings > General.
Which time limit applies	The test case’s own value, else the project default, else 15 minutes
A test that reaches its time limit	Stops and fails with “Timed out: the test did not finish within N minutes …”
Test cases with no steps and no Expected Outcome	Left out of runs and not charged
Parallel browser sessions per credential	Unlimited by default, or 1 to 50. Running tests and explorations with the credential count toward it. Tests beyond the limit wait in Queued.
Sign-ins per credential at the same time	One. Other tests and explorations with that credential wait, then reuse the saved login.

See Write a test case, Credentials and automatic login and Troubleshoot failed tests.

Limit or default	Value
Invitation validity	7 days after it is sent. Resend invitation starts a new 7 days and stops the old link.
Roles an invitation can grant	Viewer, Editor or Admin, never Owner, and never above the role of the person inviting
Default role in Invite Member	Viewer
Open invitations per email address	One per organization. Inviting the same address again replaces the pending invitation.
Invitations you can send	Limited per person and per organization; resends count. Over the limit, Testmode shows “Too many invitations sent recently. Try again later.”

See Manage your organization and team.

Limit or default	Value
Frequencies	Hourly, Daily, Weekly or Monthly
Days on a Weekly schedule	At least one
Hourly intervals	Every 1, 2, 3, 4, 6, 8 or 12 hours
Time of day	Any hour, at minute :00, :15, :30 or :45
Day of month	1 to 28
Timezones	13 to choose from: UTC, Eastern Time (US), Central Time (US), Mountain Time (US), Pacific Time (US), London (UK), Paris (CET), Berlin (CET), Amsterdam (CET), Tokyo (Japan), Shanghai (China), Singapore and Sydney (Australia)
Report recipients per schedule	Up to 20 addresses, including people outside your organization
Report options	Only when a test fails (default), After every run or Never
Reports per run	At most one email
A new schedule starts with	Daily at 09:00 in your browser’s timezone, enabled, reports Only when a test fails to your own address, and the project’s oldest environment
Runs I start on your profile	Don’t email me by default

See Schedule test plans and Email run reports.

Limit or default	Value
Rows per page on Runs, Test Cases, Test Plans and Explore	15
Test Cases grouped by tag or plan	Not split into pages. Every matching test case is shown.
Recent runs on Overview	The 5 most recent runs
Pass rate on Overview	Tests in runs started in the last 7 days
Screenshot viewer zoom	50% to 300%

See Read run results.

Limit or default	Value
Messages per conversation	Up to 60. Click New chat to start over.
Characters per message	Up to 20,000
Characters in a whole conversation	Up to 120,000, after which Testmode asks you to start a new chat
Steps in a test case the Assistant creates or changes	1 to 60. Longer tests must be split into several test cases. A test case the Assistant creates also needs an Expected Outcome.
Attached files	CSV or Excel (.xlsx or .xls), up to 500 rows and 100 columns. For Excel files, only the first sheet is read.
What the Assistant receives from a file	The column headers, the first 5 rows and the total row count
Chat history	Kept per project, in the current browser tab only
Dictation language	US English

See Create tests with the Assistant.

Limit or default	Value
Page budget	10, 25 (default), 50 or 100 pages
Active explorations per project	One at a time
Credit balance needed to start an exploration	More than zero
Interact mode	Tries interactions on up to 5 pages. Not available on production environments.
Production environments	Always explored in Observe mode

The page budget is a maximum. An exploration can finish having visited fewer pages than its budget. See Start an exploration.

Limit or default	Value
What a run costs	One credit per step of each test case in the run, at least one per test case, reserved when the run starts. Test cases with no steps and no Expected Outcome cost nothing.
Credits returned	For a test that ends before the agent’s first action, for example when login fails
Who can buy credits and open Invoices & receipts	Owners and Admins
Credit expiry	Credits come in batches that expire. The Billing tab shows the Next expiry.
Billing history	The 50 most recent entries, with no export

See How credits work and Credits and billing.

Testmode also limits how many turns the agent can take in a test, how long an exploration and a single login may take, how many console and network entries a test keeps, and how often it retries a test that could not be set up. It also ends a test early when the agent stops making progress. These numbers are not published, because they can change. The number of turns grows with a test’s time limit.

What they mean for you: keep each test short and focused, split long journeys into several test cases, and raise a test’s Maximum duration only when a journey really needs longer. See Write effective steps and Expected Outcomes.