Testim
Also known as: Tricentis Testim, Testim.io
Testim is Tricentis software for automating application tests across web, native mobile and Salesforce. Its Salesforce agent turns a tester's instructions into a runnable test, works against the connected test environment, and validates the result. Testim combines that authoring flow with a visual editor, reusable steps, smart locators, a test execution grid and release pipelines so quality teams can run the same checks again after an application changes.
Testim is a Tricentis product, bought in 2022, for creating and running end to end tests against web applications, native mobile apps and Salesforce. The web and mobile editors combine recorded interactions with an editable sequence of steps. A tester can insert a validation, add JavaScript to an individual step or group a sequence for reuse across a project.
Smart Locators weigh several characteristics of an element to keep the same step working when a single identifier changes, and Salesforce locators use the connected org's metadata to find dynamic Lightning elements. When a run fails, root cause analysis groups the errors and pairs each failing step with comparison screenshots and logs. The mobile service runs tests on iOS and Android, with real devices, simulators and emulators available across different grid configurations.
Agentic Test Automation for Salesforce, launched in October 2025, starts from a description of the behavior a tester wants to verify. A tester can ask it to create a lead, move it through conversion and check the resulting account, contact and opportunity. Several specialist agents work together behind the prompt, using Salesforce actions to build the test in the connected test environment.
The agent asks the tester to approve some steps while it works and asks again before running the finished sequence. The tester can switch to the visual editor to change the sequence or run it manually. When the agent runs the test, it validates the finished sequence, tries to resolve issues it finds and offers a summary artifact that the tester can attach to the test.
Testim quotes the paid plans for Web, Mobile and Salesforce. The Web plan adds Turbo mode, accessibility testing and activity audit logs, and the Salesforce plan covers testing on Salesforce preview releases and codeless steps for Flows, permissions and document validation. A trial rolls into the free Community plan, limited to one per organization, which keeps unlimited local mobile runs on one parallel test. Remote mobile capacity is purchased separately from local mobile testing, and the Mobile package starts with one parallel test.
The Salesforce agent has its own two month trial and then draws on Tricentis AI Credits shared across Tricentis tools. Testim holds SOC 2 Type II and Professional projects support single sign on through Okta, OneLogin and Microsoft Entra ID. The browser grid can run in the public cloud or in a private configuration, including a connection to a customer's own Selenium grid. The agent's prompt processing takes place in the US through Microsoft Azure OpenAI Service.
Application quality teams use Testim when they need a test asset that a tester can review and run repeatedly as a release check. Microsoft's cybersecurity engineering team went from no end to end tests to 80 stable tests in one six hour session. JFrog built 150 new tests in two weeks after a platform release, and Outbrain cut UI test authoring time by 95% to 20 minutes and its time on test automation, maintenance included, by about 66%.
Globality increased test coverage 30% in its first 30 days, and NewsCred runs its suite on every pull request, catching two to three bugs a month before production. A Salesforce release team can keep agent generated checks for the metadata and permissions in its own Lightning instance, and parallel execution lets the same suite cover several browsers or devices within a release window.
Vendor details
Canonical URL
https://www.testim.io
Category
SRE / DevOps agent
Subcategory
Agent assisted application and release testing
Company status
acquired
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
The Testim CLI connects test runs to Jenkins, CircleCI, TeamCity, Travis CI, Codeship and GitHub pipelines. Jira, Slack and TestRail receive defects, alerts and results, and Applitools adds visual validation. Tests execute on Testim's own grid or on Selenium Grid, BrowserStack, Sauce Labs and LambdaTest. SeaLights, another Tricentis product, maps tests to code changes so a pipeline runs only the affected ones. The Salesforce agent works through Lightning and Salesforce API steps in the connected org.
In practice
A Salesforce team asks the agent to convert a lead in its test org, approve the generated steps and validate the resulting account and opportunity.
A web quality team attaches a regression suite to its pull request pipeline and runs the same scenarios across its browser grid.
A mobile engineer records an onboarding flow on a connected phone and runs it across a virtual device grid before release.
Sources & related URLs
Research sources
Agentic Index coverage score
10.0 / 14 capabilities · 71%
| Integrations & Tool Calling | Partial |
|---|---|
|
Salesforce actions run against the connected test environment with the selected persona. The agent can create and edit records, advance opportunities and verify related records. Salesforce steps reach both Lightning and Salesforce API operations. The web editor also accepts custom API action steps. Those extensions are configured in the test editor, while the Salesforce agent works inside Salesforce. Sourcedocs.tricentis.com/testim/content/salesforce-testing/salesforce-steps/index.htmread 2026-11-20 |
|
| Workflow Orchestration | Full |
|
Agentic Test Automation coordinates specialist workers to turn a natural language request into an executable test. For a Salesforce lead conversion, the authored workflow can create a lead, move through conversion and verify the new objects in order. The resulting test can contain reusable groups, conditional steps and validations, so its execution follows the sequence and constraints the tester saves. Sourcetestim.io/airead 2026-11-20 |
|
| Knowledge Grounding & RAG | Partial |
|
Salesforce steps use fields and record types from the environments connected to the project. That context lets a test target an organization's own picklists, related records and quick actions, including custom choices that are specific to its test environment. The agent uses the environment's metadata and the tester's instructions to assemble an executable sequence for the task. Sourcedocs.tricentis.com/testim/content/salesforce-testing/salesforce-steps/index.htmread 2026-11-20 |
|
| Human Oversight & Guardrails | Full |
|
Step confirmations let a tester accept or decline proposed actions as the Salesforce agent assembles a test. Before it runs the complete sequence, the agent asks for a separate confirmation. At that point the tester can move the test into the standard editor to review individual steps, make changes, or run it personally. A user opts in before opening the agent for the first time. Sourcedocs.tricentis.com/testim/content/salesforce-testing/create-a-salesforce-test/use-agentic-test-automation-for-salesforce.htmread 2026-11-20 |
|
| Security, Identity & Governance | Full |
|
Testim holds SOC 2 Type II for security, availability and confidentiality controls. Professional projects support single sign on through Okta, OneLogin and Microsoft Entra ID, with access tied to project invitations. Project and company owners have access to the audit log. The service encrypts traffic and stored disks and runs application and external penetration assessments. Sourcetestim.io/pricingread 2026-11-20 |
|
| Observability & Auditability | Partial |
|
The execution screen records the branch, browser or device, start time, duration, concurrency and pass or fail outcome per run. A Salesforce agent session can also save its closing summary as a test artifact. Professional project and company owners can export account events from the audit log. The run details center on test execution and do not expose the worker's reasoning for each choice. Sourcedocs.tricentis.com/testim/content/results/execution-runs-screen.htmread 2026-11-20 |
|
| Memory & State Persistence | Partial |
|
A Salesforce test authoring session keeps the prompt and chat history together while the tester asks for revisions and reviews confirmations. Session data has a unique identifier, and prompt data may be kept for up to 24 hours to sustain that conversation. A finished test can store the agent's summary as a project artifact. The session context does not become a buyer managed memory for later sessions. Sourcedocs.tricentis.com/testim/content/overview/testim-overview/use-ai-in-with-testim/ai-data-usage-policy.htmread 2026-11-20 |
|
| Deployment & Data Residency | Full |
|
The web authoring service runs in Testim's hosted application. Web and Salesforce browser testing can use the public grid or a private grid reached through VPN, and the platform can connect to a customer's Selenium grid. Mobile execution has separate local and remote device options. Azure OpenAI processes the Salesforce agent's prompt data in the US even when a test uses a private grid. Sourcetestim.io/pricingread 2026-11-20 |
|
| Prebuilt Agents / Templates / Packs | Partial |
|
The Salesforce step library supplies operations for logging in as a persona, looking up a record, validating a related list and checking permissions. The tester or agent combines these building blocks into a test for a particular environment. Reusable groups can carry common sequences across the team's own tests. The library packages individual actions for authoring rather than complete agents for separate jobs. Sourcedocs.tricentis.com/testim/content/salesforce-testing/salesforce-steps/index.htmread 2026-11-20 |
|
| Triggers & Channel Coverage | Partial |
|
A tester starts a new Salesforce agent session from the testing account. Once a test exists, scheduled runs can select a suite, label or test plan and execute it against different data sets. Web test suites can also run from continuous integration. Those recurring runs execute saved tests, while the authoring session starts with a person opening it and submitting a prompt. Sourcedocs.tricentis.com/testim/content/running-tests/scheduler.htmread 2026-11-20 |
|
| Model Flexibility & Routing | Not documented |
|
Microsoft Azure OpenAI Service powers Agentic Test Automation for Salesforce and the Testim coding and help assistants. The service adds context for the selected feature before it sends a request, then returns the result to that tool. Company owners allocate access to the coding assistant and individuals opt in to the Salesforce agent. Model selection stays with Testim. Sourcedocs.tricentis.com/testim/content/overview/testim-overview/use-ai-in-with-testim/ai-data-usage-policy.htmread 2026-11-20 |
|
| APIs / SDKs / MCP Extensibility | Full |
|
The Professional REST API can initiate a run for a test, suite, label or test plan, return execution results and manage project branches. Project settings issue API keys that administrators can revoke separately. The Testim CLI can also connect the test runner to a build pipeline and select which tests to execute. These interfaces operate on the Testim test portfolio and its execution history. Sourcedocs.tricentis.com/testim/content/administration/api-access.htmread 2026-11-20 |
|
| Testing, Debugging & Optimization | Full |
|
The visual editor lets a tester define expected results, inspect failing steps and repeat a test with different CSV, Excel or JSON inputs. Comparison screenshots, DOM information and console output help isolate the cause of a failure. Breakpoints let an engineer stop during a local run and inspect the current step. The platform can repeat a suite across device or browser configurations. Sourcedocs.tricentis.com/testim/content/overview/testim-overview/testim-automate.htmread 2026-11-20 |
|
| Browser / Computer-use | Full |
|
The Salesforce agent runs tests against the connected application through the Testim browser extension and the supported test environment. A generated sequence can move a test opportunity through Lightning, then check the fields on the resulting records. Testim can execute saved web tests on its own browser grid and covers native mobile interfaces through physical devices and simulators. Sourcedocs.tricentis.com/testim/content/salesforce-testing/create-a-salesforce-test/use-agentic-test-automation-for-salesforce.htmread 2026-11-20 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Quote only
Parallel test capacity and Salesforce AI Credits
Included quota
Testim Mobile starts with one parallel test. Local mobile runs on the Web plan include one parallel test.
What is public
Testim quotes its Salesforce, Web and Mobile plans. Any trial converts to the free Community plan, with one Community plan per organization, unlimited local mobile runs on one parallel test and support through documentation and the community Slack channel. The Salesforce test agent has a separate two month trial for each user who opts in.
Billing mechanics
Test runs are unlimited. Mobile starts with one parallel test, and additional parallel tests and remote real devices are sold separately. Tricentis AI Credits are shared across products and the difficulty of a Salesforce agent request determines how many it consumes.
Variable cost rationale
Parallel capacity, remote devices and the compute used by Salesforce agent requests can all increase the cost of the package.
Sales call required
Yes, required for paid access
Free / trial
Community plan and product trial. Salesforce agent trial of two months.
Commercial notes
Paid plans include online chat support 24 hours a day on weekdays. Professional customers can export tests to Puppeteer, Selenium or Playwright. Payment is by check or wire transfer, and plan changes and limit increases run through the sales representative. Tests remain in the service for up to 30 days after cancellation.
Missing data
The paid package prices, AI Credit bundles and additional grid capacity are quoted.
Related vendors
- AlertD — AI agents for AWS operations that run read only in the customer's…
- Anyshift — AI SRE built on a versioned knowledge graph of infrastructure, apps…
- Avesha — Obliq, Avesha's autonomous AI SRE for Kubernetes and agentic…
- AWS DevOps Agent — AWS DevOps Agent investigates incidents and proposes mitigations for…
- Better Stack — Better Stack offers incident management with built in on call,…
- BigPanda — BigPanda is an agentic IT operations platform whose Biggy agents…
Alternatives to Testim
Other products that do the work Testim does, and what each one is.
- Grafana k6
- Tricentis
- BluebricksContext and control layer
- Chamber
- HoneycombObservability platform for traces, logs and metrics
- Massdriver