Skip to content
Kaarne Superstore⌕
AI & Digital Tools

The Automation Ran Twice. Which Customer Record Is Right?

A repeated automation run can create duplicate work. Decide how to identify one request, reconcile partial results and test a retry before routine use.

Two enquiry envelopes cast one shared house-shaped shadow in a photography studio.

Two rows appear in the booking list. Both contain the same property address and contact details. One colleague starts arranging a photographer; another sees the second row and starts arranging one too.

The immediate question is whether the records represent one enquiry or two genuinely different jobs. Fix that question before treating the second row as another commitment.

Decide what identifies the request

In this fictional photography-studio example, the same form was submitted twice. A stable request identifier can help connect later actions to the original request. A row number describes a position in a sheet; it should not be your only explanation of which enquiry the row represents.

Do not merge records just because the email address matches. The same customer may request two properties, two dates or a second service. Agree what makes a request distinct, and retain enough context for a person to review an uncertain match.

That identity rule belongs beside the workflow plan. If it exists only inside one filter or one person’s memory, the next maintainer may not know why an apparent duplicate was kept or rejected.

Check what happened before trying again

A later failure does not, by itself, tell you whether an earlier action completed. Perhaps a task was created but the source record was not marked as processed. Repeating the entire operation without checking could add another task.

Review the actual destination record and the original request together. Establish which action already happened, which step remains and which record will be authoritative. Preserve relevant human notes when reconciling the copies.

The exact recovery method depends on the tools and configuration. Make’s official error-handling documentation describes several possible behaviours; choose the behaviour for your scenario rather than assuming every error or retry follows the same route.

Test the second run deliberately

Create a fictional request with an identifier you recognise. Write the expected destination before running it. Then process the same request again and inspect the actual list, not only the run indicator.

Your check might require exactly one destination task, unchanged human notes and a visible explanation that the request was already handled. Then try a genuinely different request from the same customer. It must not disappear merely because the contact details match.

Also rehearse an uncertain match. If two records compete and neither is clearly authoritative, route the case to review. An automation that stops for a useful question can be easier to operate than one that quietly guesses.

Leave a short recovery note

Record the request identifier, what you found, the correction and the next check. A useful note lets a colleague understand the business decision without reconstructing your entire investigation.

The result you want is one trustworthy next action per real request. “It ran successfully twice” does not establish that.

Technical reference: Make’s official overview of error handling. Check the behaviour of the configuration you actually use.

Keep working on this question

Explore our turn a new enquiry into a clear next step guide, or read A New Enquiry Arrived. What Must Happen Before It Becomes a Task?.

For a structured learning option, see the Office Automation with Matze product overview.

Explore the course contents and current price. Affiliate link: Superstore may earn a commission if you buy. Review the format and requirements before purchasing.

0

Discussion 0

What did this article make you think about?

Join the conversation

Your email address will not be published. Required fields are marked *

Saved articles

Saved in this browser.