Describe the problem/error/question
I’m building a workflow in n8n to automatically generate SEO blog titles based on keyword data from DataForSEO. The setup currently works like this: I fetch multiple keywords through an HTTP Request, split them with a Split Out node, clean up the data with Edit Fields, and then send everything to an AI Message Assistant node.
The problem is that I don’t want to create a blog for each keyword. Instead, I want the AI to look at all keywords together, decide which one fits best (based on metrics like search volume and competition), and then create one single blog post around that keyword. The idea is that the AI first picks the best keyword (the primary keyword) and then uses the remaining ones as secondary keywords in the next AI node that writes the blog content.
Right now, the Message Assistant either receives multiple items (and tries to generate multiple outputs) or gets a full array but doesn’t know how to focus on one keyword.
What’s the best way to do this in n8n? Should I aggregate all Split Out items back into one array before the AI node, or is there a smarter way to make the Message Assistant choose the best keyword and only pass along that single result?
Thanks in advance!
Please share your workflow
