How to review a failed Android run

Collect the device and scenario context, distinguish status from outcome and inspect partial completion before repeating an Android task.

About the demoExplore the use case

Editorial illustration: a magnifying glass over a result tile beside a phone

AI-generated editorial illustration; not a product screenshot.

1. Keep the run connected to its context

Begin with the target device, scenario, approximate start time, relevant application version and the result you expected. Keep the identifier or reference available for that run with your notes. Without this context, an error message from one attempt can easily be confused with another.

Taptain provides execution status and run diagnostics. Record the information available for the specific attempt before changing the task or trying again. Keep your description of what you observed separate from your explanation of why it may have happened.

2. Read the status without assuming the outcome

First determine whether the task is still waiting, executing or has a reported result, using the states actually shown for that run. A request being recorded or accepted does not by itself establish that the intended device work finished. Also check whether you are looking at a dry run or device execution.

Compare the reported result with the outcome agreed for the task. If these do not match, write down the difference precisely. “The run reports completion, but the expected device state is absent” is more useful for investigation than a broad statement that the automation does not work.

3. Find the last confirmed step

Use the available diagnostics to identify the last step whose result you can confirm and the first point where expectations differ. If the records do not show this level of detail, mark the boundary as unknown. Do not fill a gap in the evidence with an assumed sequence of events.

Compare the device's current state with the planned starting and ending states for that step. Note whether an earlier action could already have completed. A failed run can leave useful work, partial changes or an uncertain result; understanding which case applies is part of deciding what to do next.

4. Check one explanation at a time

Organise the open questions around the connection, permissions, device and application state, task inputs and the planned sequence. Treat these as investigation categories, not a diagnosis supplied by the product. Use the actual error and observations to decide which question to check first.

Make a bounded change only when it tests a specific explanation, and record what changed. If several settings, devices and task details change together, a successful next attempt may still leave you unable to explain the original failure or repeat the success reliably.

5. Decide whether a retry can repeat completed work

Before a retry, inspect the current device state and identify actions that may already have taken effect. Decide whether the agreed starting conditions can be restored and who is authorised to make that change. Do not assume automatic rollback or duplicate prevention unless it has been established for the particular operation.

Keep the next attempt limited to the device and task needed to test your explanation. Review other planned repetitions before allowing the same uncertain task to continue across a fleet. The practical question is what you expect this new attempt to prove and how you will inspect its outcome.

6. Close the review with facts and open questions

A useful review note contains the expected result, observed result, device and scenario context, available diagnostic details, the last confirmed step, the change tested and the next outcome. Label an explanation as a hypothesis until the checks support it. Leave unresolved points visible for the next operator.

When discussing Taptain in a demo, ask to see how execution status and diagnostics relate to one agreed scenario. Discuss how the operator would handle partial completion in your environment. Use a non-sensitive description of the task; do not send production logs, credentials or customer records through the website form.

Get to know Taptain

Start with the product capabilities and an automation example.

Personal demos are coming

Demo booking is not open yet. Explore Taptain’s capabilities, run the example scenario on this site and read the guides to preparing your Android fleet.

Explore the product

Read the guides