Profound Agents are built from connected nodes, each running a single step on a single input. That's quick to build and run for one competitor page or one blog post, but applying the same steps to fifty inputs means duplicating every node fifty times, with each running sequentially. Build time and run time both grow with every input you add.
The Iteration node eliminates this scaling challenge.
Now, you can build your steps once, pass in a list, and the node runs those steps on every item simultaneously, up to 50 at a time. The output is a collected set of results that the rest of your Agent can reference, combine, or act on.
How the Iteration node works
The Iteration node encloses a group of steps inside your Agent. You build the steps once, pass in a list of items, and each item is simultaneously run through the enclosed steps. The node collects all outputs into a single array that downstream nodes can reference.
It accepts any array as input, whether that comes from an upstream LLM node generating a structured list, a Profound data node returning cited URLs, or a Sheets import. The output is an array of the same length, one result per input item.
How Profound customers can use this
Reddit narrative analysis at scale
Pull your top cited Reddit threads using Profound data, iterate over each one to scrape and summarize how your brand is discussed, then synthesize the per-thread results into a single narrative report. Your Agent can even send it to relevant teams via Slack.
Competitive pricing and packaging analysis
Upload a list of competitors and iterate over each one to crawl their pricing page and analyze structure, pricing, and product offerings. Feed the combined output into an LLM node that compares the results against your own pricing and packaging via your Knowledge Base, and recommends competitive positioning.
Media placement monitoring
Feed a list of top-performing third-party media URLs into an Iteration node. For each URL, crawl the article, check whether your brand was mentioned, extract the placement copy, and assess the tone. After the iteration, combine the results into a prioritized media target list with coverage gaps, key contacts, and drafted outreach tailored to each publication.
Using the Iteration node vs. Sheets
Both the Iteration node and Profound Sheets let you process multiple inputs through the same workflow. The difference is scope and context.
Sheets give each row its own independent Agent run, which makes them the right fit when you want to store and revisit individual results, run large batches over time, or when inputs don't depend on each other.
The Iteration node keeps everything inside a single Agent run. All items share context, so you can aggregate, compare, and synthesize across the full set of results before the Agent finishes. Use the Iteration node when the value comes from processing a list and doing something with the combined output, rather than storing each result on its own.
Get started
The Iteration node is available today for all Profound customers. If you are not yet a Profound customer and want to see this in action, we encourage you to get a demo.
