Why is my merged 'markets' array only capturing data from the last loop iteration in n8n?

Below is a few suggestions that might explain why only the last iteration’s data appears in your final (merged) output:

1. Merge Mode Selection:
   The Merge node in n8n offers different modes (for example, Append and Combine). If you’re using a mode where items are merged by position (or by fields) without appending, then only the data from the last iteration (the last run of the loop) might be kept. If your intent is to accumulate all market abbreviations, you’ll need to ensure that you use a mode that appends items from each iteration rather than overwriting the previous ones.

2. Loop Behavior and Data Accumulation:
   n8n nodes generally work item–by–item. When you loop over linked records and then use a subsequent Set or Merge node to add data to a field (for example, a “markets” array), it’s possible that the merging process is resetting on each loop run rather than accumulating. In many cases, you might want to perform the accumulation after the loop has finished by collecting all iteration results. For example, you could store the retrieved abbreviation per iteration into a temporary field (or output array) and then use either a Code node (with a reduce function) or an Append merge option to combine all the results into one array.

3. Changes in Node Behavior With Updates:
   Since you mentioned that the original code worked previously but now produces an empty array (or only the final iteration’s data), it’s possible that a recent update changed the way the Merge or Set node handles multiple runs in a loop. Double-check the node’s configuration (especially the merge mode and how you reference the data) to ensure that it aligns with the expected behavior in the updated version.

What can you try?

• Verify the merge mode in your Merge node. If you’re not already doing so, try using an “Append” option (if available) so that each loop iteration’s output is added to the existing array rather than replacing it.
• Consider accumulating the abbreviations outside the loop. For instance, after the loop node finishes, use a Code node to reduce all the values of the field (e.g. “Märkte”) into one array.
• Double-check your expressions—if you’re referencing nested fields (for instance when using dot notation) in a Set or Merge node, ensure they’re formatted as plain text without accidental evaluation errors.

If after these steps the problem persists, it might be worthwhile to test the loop with simple static data to see if the accumulation process works as expected. This can help narrow down whether the issue stems from the node’s configuration or code changes in the latest update.

These suggestions are based on best practices for merging and splitting data in n8n. Let me know if you need further help configuring your workflow!