Notice: This page requires JavaScript to function properly.
Please enable JavaScript in your browser settings or update your browser.
Lære Comparing Products in Google Sheets | Prompting Agents to Take Action
AI Agents in Action

Comparing Products in Google Sheets

Sveip for å vise menyen

For the second workflow, you add the Chrome extension into the mix. The agent will go out on the open web, research five products, and drop a structured comparison into a Google Sheet.

This is not just shopping. If you are launching a Shopify store, comparison research is how you scout what already exists in your category, what is selling, and where the pricing sits.

Even if you are not launching anything, this workflow is useful for everyday decisions: buying gear, evaluating tools, choosing software, or anything that would normally take hours of review reading.

The Scenario

Pick a product category you are considering selling, or choose something you would genuinely shop for.

You want the top 5 contenders in a realistic price range, compared using the criteria that matter:

  • Price
  • Key specs
  • Top 2 features
  • One-line standout for each product

You want the agent to:

  • Research five products that fit your criteria using the Chrome extension.
  • Extract comparison data from product pages and reviews.
  • Structure everything into a Google Sheet you can scan in 30 seconds.

Connectors used: Claude for Chrome and Google Sheets.

Note
Prompt

Research the top 5 [product category] under $[price] currently available.

For each product, capture:

  • Product name
  • Brand
  • Price
  • [Key spec 1]
  • [Key spec 2]
  • Top 2 features
  • One-line standout feature

Use the Chrome extension to gather data from product pages and review sites. Stick to products available on Amazon, the brand's direct site, or major retailers.

Output the results as a Google Sheet called [Category] Comparison.

Use one row per product and one column per data point. Add a header row.

If you cannot find one of the data points for a product, leave that cell blank and note it in a Comments column.

Stop when you have 5 complete product rows.

Why the Chrome Extension Matters

Cowork can browse the web on its own, but the Chrome extension makes it much better at live web research.

With the extension, the agent can:

  • Scroll real webpages
  • Open product pages
  • Click into reviews
  • Expand product specifications
  • Move from search results to detailed pages

The extension acts like the agent's eyes on the live internet. This workflow shows why that matters.

Why Use a Spreadsheet Instead of a Doc?

Comparison data works best in a table.

A paragraph like "Product A has a 75-minute battery, while Product B has 180 minutes" is harder to scan than two values in a spreadsheet column. Choose the output format that matches the data:

Iteration Patterns

You will probably want to improve the sheet over time. Practice extending it with short follow-up prompts.

Each tweak can be a one-line reply. The agent updates the existing sheet instead of starting over.

question mark

Why is Google Sheets the best output format for the product comparison workflow?

Velg det helt riktige svaret

Alt var klart?

Hvordan kan vi forbedre det?

Takk for tilbakemeldingene dine!

Seksjon 4. Kapittel 3

Spør AI

expand

Spør AI

ChatGPT

Spør om hva du vil, eller prøv ett av de foreslåtte spørsmålene for å starte chatten vår

Comparing Products in Google Sheets

For the second workflow, you add the Chrome extension into the mix. The agent will go out on the open web, research five products, and drop a structured comparison into a Google Sheet.

This is not just shopping. If you are launching a Shopify store, comparison research is how you scout what already exists in your category, what is selling, and where the pricing sits.

Even if you are not launching anything, this workflow is useful for everyday decisions: buying gear, evaluating tools, choosing software, or anything that would normally take hours of review reading.

The Scenario

Pick a product category you are considering selling, or choose something you would genuinely shop for.

You want the top 5 contenders in a realistic price range, compared using the criteria that matter:

  • Price
  • Key specs
  • Top 2 features
  • One-line standout for each product

You want the agent to:

  • Research five products that fit your criteria using the Chrome extension.
  • Extract comparison data from product pages and reviews.
  • Structure everything into a Google Sheet you can scan in 30 seconds.

Connectors used: Claude for Chrome and Google Sheets.

Note
Prompt

Research the top 5 [product category] under $[price] currently available.

For each product, capture:

  • Product name
  • Brand
  • Price
  • [Key spec 1]
  • [Key spec 2]
  • Top 2 features
  • One-line standout feature

Use the Chrome extension to gather data from product pages and review sites. Stick to products available on Amazon, the brand's direct site, or major retailers.

Output the results as a Google Sheet called [Category] Comparison.

Use one row per product and one column per data point. Add a header row.

If you cannot find one of the data points for a product, leave that cell blank and note it in a Comments column.

Stop when you have 5 complete product rows.

Why the Chrome Extension Matters

Cowork can browse the web on its own, but the Chrome extension makes it much better at live web research.

With the extension, the agent can:

  • Scroll real webpages
  • Open product pages
  • Click into reviews
  • Expand product specifications
  • Move from search results to detailed pages

The extension acts like the agent's eyes on the live internet. This workflow shows why that matters.

Why Use a Spreadsheet Instead of a Doc?

Comparison data works best in a table.

A paragraph like "Product A has a 75-minute battery, while Product B has 180 minutes" is harder to scan than two values in a spreadsheet column. Choose the output format that matches the data:

Iteration Patterns

You will probably want to improve the sheet over time. Practice extending it with short follow-up prompts.

Each tweak can be a one-line reply. The agent updates the existing sheet instead of starting over.

Alt var klart?

Hvordan kan vi forbedre det?

Takk for tilbakemeldingene dine!

Seksjon 4. Kapittel 3
some-alt