The Merge Node
Combines two branches of a workflow back into one.
What it actually does
Once a workflow has split — via an If or Switch node, or by calling two different APIs in parallel — you often need to bring those separate branches back together again. The Merge node does exactly that, combining two inputs into a single output, with a few different modes depending on what “combining” should actually mean for your data.
Append simply stacks the items from both inputs one after another — useful when both branches produce the same kind of data and you just want them in one list. Combine matches items from each input by a shared field (like an ID) and merges their fields together into single, richer items — useful when one branch fetched a customer’s details and the other fetched their order history, and you want one item per customer with both. Choose Branch picks the output from just one of the inputs, useful as a way to pick a “winner” between two possible paths.
Merge always needs exactly two inputs connected before it’ll run properly — a common early confusion is connecting only one and wondering why the node won’t execute as expected.
A worked example
A workflow splits with a Switch node — VIP customers down one branch, everyone else down another — and each branch applies different logic (a bigger discount calculation for VIPs, a standard one for everyone else). Afterward, a Merge node set to Append brings both branches back into a single list, so the rest of the workflow (writing to a spreadsheet, sending confirmation emails) only needs to be built once instead of duplicated across both branches.
A Combine example: one branch calls an HTTP Request node to fetch a customer’s profile, another branch fetches that same customer’s recent orders from a different API. A Merge node in Combine mode, matched on customer ID, produces one item per customer containing both their profile and their order data together — ready for a single downstream step instead of juggling two separate datasets.
The mistake almost everyone makes first
Using Append when Combine was actually needed, ending up with two separate lists of items stacked together rather than genuinely merged records. If your goal is “one item per customer with data from both branches,” Append won’t get you there — it just concatenates the two inputs as-is, leaving you with twice as many items rather than fuller ones. Combine, matched on a shared field, is what actually merges the fields of related items together.