Have an agent audit your whole catalogue for missing and weak content
Get a complete audit of missing descriptions, alt text, and metadata across your entire product catalogue in one report.
How it works today.
You spot-check products when customers complain or when you remember, but you know there are gaps everywhere. Some products have one-line descriptions, others missing alt text, and half your metafields are empty. You click through page after page of your admin, taking notes on what needs fixing, but you lose track of where you were.
By the time you finish checking 200 products, you have a messy spreadsheet and three weeks have passed. You know the other 800 products probably have the same issues, but starting over feels impossible. Meanwhile, thin content hurts your search rankings and customers bounce from products with no proper descriptions.
Before you start.
All of it has to be true, or step one fails in a way that is annoying to debug.
- Admin access to your store platform with ability to export product data
- A builder tool account (like Zapier, Make, or n8n) with your store connector enabled
- A generalist agent account (Claude, ChatGPT, or similar) that can process CSV files
- Write access to create and run automated workflows in your builder tool
The steps.
Export your complete product catalogue
Go to your store admin and export all products to CSV, including fields for title, description, images, alt text, and any custom metafields you use. Most platforms have this under Products > Export or similar. Download the file and note how many products you exported.
Set up the audit workflow in your builder tool
Create a new workflow that triggers manually. Add your store connector as the first step and configure it to fetch all products with the same fields you exported. Test the connection to make sure it pulls current data, then save the workflow.
Configure the content analysis step
Add your generalist agent as the next step in the workflow. Set it to receive the product data and analyze each item for content gaps. The agent will need to process the data in batches if you have more than 100 products.
Paste thisAnalyze this product data and flag items with content issues. For each product, check: 1) Description under 50 words or generic phrases, 2) Missing alt text on images, 3) Empty metafields that should have values, 4) Titles that are just model numbers. Return a CSV with product ID, issue type, severity (high/medium/low), and specific recommendation.
Test the workflow with a small batch
Run the workflow on just 10-20 products first to check the output format. Look at the CSV results to make sure the agent correctly identifies real issues and gives actionable recommendations. Adjust the prompt if it flags too many false positives or misses obvious problems.
Run the full catalogue audit
Execute the workflow on your complete product set. This may take 10-30 minutes depending on catalogue size. The workflow will process products in batches and compile a master audit report. Monitor for any errors or timeouts during processing.
Generate priority recommendations
Feed the complete audit results back to your agent to create a prioritized action plan. Focus on high-severity issues that affect customer experience and search visibility first.
Paste thisReview this complete audit data and create a priority action plan. Group issues by: 1) Quick fixes (under 2 hours total), 2) Medium effort (1-2 days), 3) Major projects (1+ weeks). For each group, list the specific products and exact changes needed. Estimate the SEO and conversion impact of fixing each group.
Set up automated monitoring
Modify your workflow to run weekly and check only products modified in the last 7 days. This catches new products added with missing content before they go live. Configure it to email you a summary of any new issues found.
Create content templates for common gaps
Use your agent to analyze products with good descriptions and generate templates for common product types. This gives your team a starting point when fixing flagged items.
Paste thisAnalyze these well-performing product descriptions and create templates for [your main product categories]. Each template should include: required elements, tone guidelines, optimal length, and 3 example descriptions. Make templates specific enough that someone can fill them out in 5 minutes per product.
What you keep.
Automating the typing does not move the accountability. These stay with a person.
- Reviewing and approving all content changes before they go live
- Deciding which audit recommendations to prioritize based on business impact
- Writing or editing the actual product descriptions, alt text, and metadata
- Setting content standards and brand voice guidelines for your team to follow
Once it works.
The first run is the demo. These are where the time actually comes back.
Run the audit workflow every Sunday night to catch content gaps in products added during the week.
Set the workflow to automatically audit any product when its status changes to 'published' or 'active'.
Expand the audit to check competitor pricing, review sentiment, and seasonal relevance alongside content gaps.
Send audit results directly to your content team's project management tool with pre-filled tasks for each flagged product.
Run this next.
One workflow is a tip. Chained, they are how a week actually changes shape.
The work this replaces.
These are the O*NET work activities this workflow covers, and the categories of tool that address them.
