Home » Workflow Automation » Loops in Workflows

How to Loop Through Data in a Workflow

Loops let your workflow process a list of items one at a time, running the same set of steps on each item. When you need to send a personalized email to every contact in a list, update 50 database records, or process each line item in an order, a loop handles the repetition automatically so you do not need separate workflow steps for each item.

When You Need Loops

Most workflows process one item at a time: one form submission, one webhook event, one order. But some automations need to handle batches of data. A nightly report workflow might need to process every order from that day. A data cleanup workflow might need to check every contact in a list for invalid phone numbers. A content workflow might need to publish a post to every social media account.

Without loops, you would need to know the exact number of items in advance and create that many identical workflow steps. A loop handles any number of items with a single set of steps, whether the list has 3 items or 3,000.

How Loops Work in a Workflow

A loop node takes a list (an array of items) as input and runs the downstream steps once for each item in the list. On each iteration, the current item is available as a workflow variable. When all items have been processed, the workflow continues to the next step after the loop.

For example, if your loop input is a list of 10 customer records, the steps inside the loop run 10 times. On the first iteration, the current item variable contains the first customer record. On the second iteration, it contains the second record, and so on.

Step-by-Step Setup

Step 1: Get the list of items to process.
Before the loop, you need a step that produces a list. This might be a database query that returns multiple records, an API call that returns an array of results, or a CSV file that has been parsed into rows. The output should be an array where each element represents one item to process.

Step 2: Add a loop node to your workflow.
In the Chain Commands visual builder, add a loop node after the step that produces the list. Configure it to iterate over the list variable. The loop node will execute its child steps once per item in the list.

Step 3: Build the steps inside the loop.
Add the processing steps that should run for each item. These steps can reference the current item variable to access fields like name, email, amount, or whatever data each item contains. Keep the processing simple, as these steps execute for every single item.

Step 4: Handle errors within the loop.
If one item fails, you probably want the loop to continue processing the remaining items. Add error handling inside the loop so a failure on item 5 does not stop items 6 through 100 from being processed. Log failed items for later review.

Step 5: Add post-loop steps.
After the loop finishes all iterations, you might want a summary step. This could count how many items were processed successfully, how many failed, and send a completion report. Connect these steps after the loop node's "done" output.

Practical Loop Examples

Batch Email Sending

Query your contact list to get all subscribers who signed up in the past week. Loop through each contact and send a personalized welcome email using their name and signup source. The loop handles 10 or 10,000 contacts identically.

Data Enrichment

Pull a list of leads from your database that are missing company information. Loop through each lead, call an enrichment API with their email domain, and update the database record with the returned company data. Each iteration processes one lead independently.

Multi-Record Updates

After a pricing change, you need to update every product in your catalog. Query all products, loop through each one, calculate the new price, and write the updated price back to the database. One workflow handles the entire catalog regardless of size.

AI Processing of Multiple Items

You have a list of customer reviews to analyze. Loop through each review, send it to an AI decision step for sentiment analysis, and store the classification alongside the review. The AI processes each review individually with full context.

Performance note: Loops that call external APIs or AI models on each iteration use credits and time for every item. A loop of 100 items where each iteration calls an AI model at 3 credits per call costs 300 credits total. Plan your credit budget accordingly and consider whether batch processing through a single AI call might be more efficient for large lists.

Avoiding Infinite Loops

An infinite loop happens when the exit condition is never met. This is less of a risk with list-based loops (they end when the list is exhausted) but can happen with condition-based loops that repeat until a value changes. Always set a maximum iteration count as a safety limit. If your loop hits 1,000 iterations without completing, something is wrong, and the workflow should stop and alert you rather than running forever and consuming credits.

Process lists, batches, and collections automatically. Build loops into your workflows with no code.

Get Started Free

View the Chain Commands App