Agents In Use
What’s new todayLast update
108 use cases added today

ChatGPT agent builds a five-competitor workbook

Faems reports that ChatGPT agent researched five competitors for his startup into a workbook, including customer-review evidence; he recognised all five from prior human research.

Source published linkedin.comAgent: ChatGPT agent
OUR ADAPTATION

Suggested prompt

Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.

Research competitors to [conference-networking product]. Build a workbook with company, website, location, offering and differences. Put positive and negative customer-review evidence with source links on a separate sheet; flag gaps.

Evidence at a glance

Sourced account
Source author
Dries Faems
Original account
Read the source ↗
Published
Source link checked
Reported result
Faems reports that ChatGPT agent researched five competitors for his startup into a workbook, including customer-review evidence; he recognised all five from prior human research.
Human involvement
Faems recognises the five competitors from previous research. The source does not independently verify all workbook fields or customer-review quotations.

A linked account supports attribution. It does not mean we independently reproduced its result.

Report an issue with this caseOpen correction email

Opens your email app with the case link included. Nothing is sent automatically.

How the agent did it

  1. Describe the conference-networking product and competitor criteria.
  2. Specify company, website, location, offering and differentiation columns.
  3. Request a separate sheet for positive and negative customer-review evidence.

Dries Faems tests ChatGPT agent against a startup idea for networking before conferences. The research task asks for a competitor workbook with columns for company name, website, location, offering and differences from his own product. A second worksheet captures positive and negative customer-review evidence about those platforms.

Faems describes roughly ten to fifteen minutes of agent work before receiving the result. The workbook identifies five competitors, all of which he recognises from his own previous research. That comparison provides a concrete human check on the competitor set, although it does not independently validate every review quotation or commercial detail in the file.

The source also explains that earlier, vaguer attempts produced weak outputs. He improved the instruction by specifying the sequence and spreadsheet structure. No connected private inbox is needed for this particular public-web research task; a calendar connection discussed elsewhere in the video should not be credited with producing the competitor workbook. The useful specification is to separate factual company fields, product comparisons and customer opinions into an inspectable artifact. The result supports examining a business idea, while market sizing and other experiments in the same video remain separate tasks rather than additional outcomes of this entry.

Prompt
Suggested prompt (written by Agents In Use, not the author's)

Source: LinkedIn · Dries Faems

Frequently asked questions

How was the research output structured?

One sheet contains competitor company fields and product differences; another holds positive and negative customer-review evidence.

What did Faems personally check?

He recognised all five competitors from his earlier research into the same startup idea.

Did it require inbox access?

The competitor task used public-web research. The source does not establish that a private email connection was needed.

More details

Remaining steps

  1. Compare the resulting competitor set with prior human research.

About this account

ChatGPT agent conducts public-web competitor research and organises company comparisons and review evidence into a multi-sheet workbook.

Firsthand account by Dries Faems; original publication date verified against Public LinkedIn JSON-LD datePublished 2025-10-06T05:49:54.356Z. Outcomes are the author’s reports, not independently reproduced. Suggested prompt written by Agents In Use; not the author’s original prompt.

Credit: Dries Faems

Source published

Added to Agents In Use

Source checked

Tools and topics

ChatGPT agent · Web research · Spreadsheet workbook

Competitor workbook · Customer reviews · Startup research

Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.