# Transaction Monitor: Catch a Broken Checkout Before Customers Do > The Transaction Monitor is the synthetic transaction monitoring agent in the Ai1 platform by MyZone AI: it replays journeys like login, sign-up and checkout in a safe sandboxed browser, stops before any payment and alerts you on its next scheduled check when a step breaks. Canonical page: https://myzone.ai/pages/agents/ai-transaction-monitoring-agent Part of Ai1, by MyZone AI. Book a Transactions walkthrough: https://calendly.com/d/ct6h-tcy-8qf/ai1-demo?a1=Transaction%20Monitor&utm_source=myzone.ai&utm_medium=agent-page&utm_content=ai-transaction-monitoring-agent-final Last updated: 2026-10-14. Reviewed by the MyZone AI team. Your checkout, tested like a real customer, with an alert when it breaks. ## At a glance - **You describe:** A journey in plain English, such as sign in, fill a cart and reach checkout - **Practice run:** Screenshots and a plain list of every site the journey loads from or sends data to - **You approve:** Nothing runs for real until you type the approval it shows you - **Replay:** In a sandboxed browser, stopping before any payment or order - **Alerts:** To your alert contacts when a step breaks, and again when it recovers - **What it never does:** Place an order, handle a card number or fix your site ## What the Transaction Monitor does - **Turns a journey into a test:** Tell it what a customer does, such as log in, add a product and reach checkout, and it writes that as a repeatable test. - **Replays it in a safe browser:** The test runs in a sandboxed browser that behaves like a real visitor, step by step, the same way every time. - **Alerts you when a step breaks:** When a journey fails, your alert contacts hear about it straight away, and again when it recovers. - **Explains the failure:** Each failure comes with the step that broke and structured evidence of why, so whoever fixes it can start right away. - **Spots stale tests and proposes repairs:** If a button was simply renamed, it proposes a small repair and shows you the difference before anything changes. - **Tracks reliability over time:** Run history, pass rate and flaky steps are shown on a Transactions panel, so you see trends as well as failures. ## How the Transaction Monitor differs from finding out when customers complain Waiting for a customer to email means a broken login or checkout can cost orders for days. The Transaction Monitor replays your key journeys in a safe sandboxed browser, stops before payment and alerts your contacts on its next scheduled check when a step breaks, with the step and the likely reason. It proposes repairs when your site simply changed and tracks pass rate over time. It never fixes your site itself. How this differs from the QA Orchestrator (https://myzone.ai/pages/agents/ai-qa-agent): For checking a change to your software before it goes out, rather than watching live customer journeys, use the QA Orchestrator. How this differs from the Uptime Monitor (https://myzone.ai/pages/agents/ai-uptime-monitoring-agent): the Uptime Monitor checks that your sites are up, while the Transaction Monitor walks through login, sign-up and checkout like a customer to prove each step still works. How this differs from the Tester (https://myzone.ai/pages/agents/ai-tester-agent): the Tester runs a browser check of your site when you ask and reports pass or fail, while the Transaction Monitor replays your key journeys on a schedule. ## How it works 1. **Describe the journey** (When you describe the journey): Tell it, in your own words, the path a customer takes, such as signing in, filling a cart and reaching checkout. 2. **Practice run** (Straight after the description): It writes the test and does a practice run, showing you screenshots and a plain list of every site the journey loads from or sends data to. 3. **You approve the test** (you approve) (Before any real run): Nothing runs for real until you type the approval it shows you. Repairs later need your approval in the same way. 4. **Replay and watch** (On every replay): The test replays the journey and stops before any payment or order, recording what passed and what failed. 5. **Alert and explain** (On the next scheduled check): If a step breaks, your contacts are told on the next scheduled check, with what failed and why, and told again when it recovers. ## When to use it - **A plugin update quietly broke your checkout:** The checkout test fails at the exact step, and you hear about it before customers start emailing. - **Customers say they cannot log in:** A login test shows whether the problem is real, which step fails and when it started. - **Your designer renamed a few buttons:** It notices the test is stale, proposes a small repair and waits for your approval. - **You want proof your sign-up form works:** A sign-up journey runs on request and keeps a history of passes and failures you can point to. - **A step fails now and then:** Pass rate and flakiness tracking show whether it is a one-off blip or a pattern worth fixing. ## What you get - Plain-English journey tests for login, sign-up, cart and checkout - A practice run with screenshots and a list of the sites each journey touches - Instant alerts when a step breaks and again when it recovers - A failure summary naming the broken step and the likely cause - Proposed repairs for stale tests, shown as a clear before and after - A Transactions panel with run history, pass rate and flakiness ## Example: weekly journey test report with a checkout alert Example with a fictional company. Names, people and figures are invented to show the agent's output. Any resemblance to a real company or person is unintended. What it was asked: Are login, seat picking, checkout and sign-up working for customers, what broke this week, and what needs my sign-off? This is an illustrative example of the weekly report Transaction Monitor is being built to produce: journey health, a failure alert with evidence, a gated repair proposal and the approved test itself. It was written by hand from the agent's design for an invented box office with invented results. No real website, payment, account, systems or people were involved, and nothing was sent or changed. Run date: 14 October 2026. ### What it found - Checkout broke for about 47 minutes on Saturday evening when the cart's continue button returned an error page; the alert reached three contacts within a minute and the journey passed again at 8:15 pm. - Across 2,688 test runs this week, 98.9% passed; login passed every one of its 672 runs and every journey stopped before the payment step, as designed. - The sign-up test went stale after a button was renamed to Join now; a one-step repair passed its practice run and is waiting for your typed approval. ## Guardrails - Your Ai1 team will tell you when the Transaction Monitor is ready for your account. - Every test stops before payment or placing an order. It never types, accepts or stores a card number. - No test runs and no repair is applied until you type the approval it shows you. - It never fixes your site. Breakages go to whoever looks after your website, only when you say so. - Passwords for test accounts go through a secure form, never into chat, and text on a web page is never treated as an instruction. ## Frequently asked questions ### Why monitor your checkout instead of waiting for complaints? A broken checkout loses orders quietly. The Transaction Monitor replays the cart-to-checkout journey in a sandboxed browser and alerts you when a step fails, so you hear before customers email. It stops before payment, so no real order is ever placed. ### How do you test a login or sign-up flow without writing scripts? Describe the journey in plain English. The Transaction Monitor writes the test, does a practice run and shows you what it touches, then replays it once you approve. Passwords for test accounts go through a secure form, never into chat. ### What should a broken checkout alert tell you? Which step failed, the likely reason and when it started, so whoever fixes it can begin straight away. The Transaction Monitor alerts your contacts on its next scheduled check, with structured evidence, and tells them again when the journey recovers. ### What is synthetic transaction monitoring? It means a robot customer walks through your key journeys, like login and checkout, on a regular basis. Transaction Monitor replays each journey in a sandboxed browser and tells you the moment a step stops working. ### Will it place real orders or charge a card? No. Every test stops before payment or order placement, and it never types, accepts or stores card numbers. Your sales figures and stock stay untouched. ### Do I need to know how to write tests? No. Describe the journey in plain English and it writes the test, does a practice run and shows you what it will touch before you approve it. ### What happens when my website changes? If a small change such as a renamed button breaks a test, it proposes a repair and shows you the difference. Nothing changes until you approve it. ### Can it fix the problem it finds? No. It tells you which step broke and why. With your go-ahead, the details go to whoever looks after your website so they can fix it. ### Is the Transaction Monitor available now? Not yet. Your Ai1 team will tell you when the Transaction Monitor is ready for your account. You can talk to us today about the journeys you would want it to test. ## More on this topic - [Your Website Is Getting Slower Every Day. You Just Don’t Know It Yet.](https://myzone.ai/pages/blog/ai-pagespeed-monitor): Sites get slower with every update. Daily automated PageSpeed monitoring catches regressions before they cost you rankings and conversions. ## About Ai1 Ai1 is the AI operations platform by MyZone AI, where each client runs on its own private server. The Transaction Monitor is not sold on its own: every Ai1 agent, including this one, is included on Developer Pro and every Fully Managed option, with no per-agent charge. Developer Core includes the development agents. Pricing: https://myzone.ai/pages/services/ai1-pricing. Security: https://myzone.ai/pages/security.