Resources · n8n integration

n8n Merge nodeConfigure Merge in n8n.

The n8n Merge node is where two branches of a workflow become one again. It waits until every connected input has run, then joins the items with a single parameter, Mode, and its 4 options. Version 3 handles 2 to 10 inputs, matching fields, a SQL query, or one branch kept exactly as it arrived.

Try the Merge node on n8nSee the 1 parameters

Verified Trustpilot reviews · AI, automation & growth agency

Why automate

What does the n8n Merge node actually do?

Merge brings several branches of the same workflow back into one stream. It sits after a split, after two lookups, or after two paths that did different work, waits until every input it is wired to has finished, and sends out one list of items. An item is one record travelling through the workflow, and a branch is one of the paths that record can take. Nothing to install, nothing to authenticate: the node ships with n8n.

The first scenario is the one most people arrive with. A workflow reads customers from one source and orders from Google Sheets, and the two lists need to become one. Mode set to combine, Combine By set to Matching Fields, and the field names typed into Fields to Match tell n8n which values have to be equal on both sides. Output Type then decides who survives: Keep Matches for the intersection, Enrich Input 1 when the customer list has to stay complete even where no order was found.

The second scenario is simple stacking. Two paths each produce alerts, one from a form and one from a monitoring call, and both should end up in the same Slack message or the same Gmail draft. Mode set to append outputs the items of input 1, then those of input 2, in that order. Number of Inputs raises the node to as many as 10 connection points, so three or four paths converge on one node instead of a chain of them.

The third scenario is for people who already think in tables. Mode set to combineBySql turns the inputs into tables named input1, input2 and so on, and Query holds the statement. A left join written by hand does what a chain of nodes would take five steps to express, and Query Parameters keeps values out of the query text.

When another node is the better answer: a straight API call that no dedicated node covers belongs in HTTP Request, not here, and reshaping a single stream is a job for the Code node. Merge only earns its place when two or more streams have to wait for each other. On an older n8n, the Code node is also the fallback, since SQL Query and more than two inputs arrived in 1.49.0.

The limits are worth knowing before the first run. Items passed into Input 1 take precedence: five items on input 1 against ten on input 2 means five items processed and the rest dropped. Fuzzy Compare is off by default, so the text "3" and the number 3 are not the same value until it is turned on. And Merge never starts a workflow on its own, a trigger always runs before it. The n8n review goes into where that model helps and where it gets in the way.

Parameters

Which parameter controls the Merge node?

The Merge node has one parameter. For each one: the node as you configure it in n8n, what the parameter changes, and our field notes.

01

Mode

mode

What you see in n8n

Notes & use cases

Mode is the node's only top-level parameter.

Key parameters

  • Mode: append stacks items, combine merges matches, combineBySql runs a query, chooseBranch passes a branch untouched.
  • Number of Inputs: 2 to 10 connection points.
  • Combine By: Matching Fields, Position or All Possible Combinations.
  • Fields to Match: typed as id, name, plus Fields To Match Have Different Names if labels differ.
  • Output Type: Keep Matches, Keep Non-Matches, Keep Everything, Enrich Input 1, Enrich Input 2.
  • Query: required on SQL Query; inputs become tables input1, input2.
  • Output and Use Data of Input: Choose Branch sends one input, or an empty item.
Use cases
enrich a customer list with orders from a second branch. Output Data From can send input 1 alone.
Need help

Need help automating Merge with n8n?

A person reads every message.

FAQ

Questions people ask about Merge

01Is the n8n Merge node free on n8n Cloud and self-hosted?
Yes. Merge is a core node shipped with n8n, so it is there the moment a workflow is created. There is nothing to install, no marketplace to visit, and no extra cost on the n8n side. It behaves the same on n8n Cloud, the hosted offer run by n8n, and on a self-hosted instance installed with Docker or npm under the Community Edition and its Sustainable Use license. A workflow built on one runs on the other without changes. The only thing that differs between instances is the node version: a workflow created long ago can still show an older Merge with fewer options, and a node never changes version on its own.
02What credentials does the Merge node need?
None, and that is the short answer. Merge has no credential and no Authentication selector, because it never calls an outside service. It works only on the items already travelling inside the workflow, so there is nothing to configure on the account side and no key to rotate. What it does need is wiring: at least two branches connected to its inputs, and a trigger somewhere upstream, since Merge does not start a workflow by itself. If the branches it waits for never run, the node never fires either. Everything else happens in the Mode parameter and the fields that appear once a mode is chosen.
03What are the limits of the Merge node?
Three of them matter in practice. Input 1 takes precedence when the streams have uneven numbers of items: five items on input 1 against ten on input 2 gives five processed items, and the remaining five are not processed. Type differences are treated as real differences unless Fuzzy Compare is added, so a value stored as text and the same value stored as a number will not match. And the parameter set depends on the version: more than two inputs and the SQL Query mode arrived later, so an older instance shows neither. Beyond that, Merge joins data, it does not transform it, and it cannot call an API.
04When should you use Merge instead of the Code node?
Use Merge whenever two or more branches have to wait for each other before anything else happens. That waiting behavior is the real value: the node holds until every connected input has been executed, and the join itself is four options in a dropdown instead of a script to maintain. Use the Code node when a single stream needs reshaping, when the join logic is conditional in a way no mode covers, or when the instance is too old for SQL Query and more than two inputs. A reasonable rule: if the operation can be described as append, match, query or keep one branch, Merge does it with no code at all.
05n8n or Make for merging two data streams?
It depends on three things, and price is only one of them. Make runs on its own hosted platform with no self-hosting option and bills by operation, so a workflow that merges large lists item by item becomes a volume question. n8n runs on n8n Cloud or on a server you control, which matters when the merged data is customer or financial data that should not leave your own infrastructure. The third criterion is how you think about the join. Merge exposes it as modes, matching fields and even SQL, which suits people comfortable with joins. Both tools draw workflows visually, so the choice rarely comes down to the canvas.
Hack'celeration Lab

Get our weekly integration tips.

No spam. Unsubscribe anytime.

Same platform

Other docs you might need on this stack.

Browse all integrations