中文 Book a Workshop
AI workflow · Marketing · Research

Competitive Research with AI

Competitive research with AI compares what competitors publicly say — positioning, audience, pricing, proof — in one table where every cell cites the page it came from, so the comparison can be checked and updated.

TeamMarketing
LevelBeginner
ToolsChatGPT · Claude · Copilot
How it usually goes
  • Someone makes a comparison slide once a year.
  • It is out of date in a month.
  • Nobody knows where the numbers came from.
The workflow · 5 steps
  1. 1

    Fix the list and the pages

    Three to five competitors. For each: homepage, pricing page, one product page. Same page types for everyone, or the comparison is unfair.

  2. 2

    Fix the columns before you start

    Who it is for, the main promise, pricing model, proof they show, what they do not mention.

  3. 3

    Extract, do not evaluate

    First pass is quotation: what does each page actually say. Evaluation comes after, and it is yours.

  4. 4

    Ask for the gaps

    What does every competitor claim (table stakes), and what does only one claim. That second list is where positioning lives.

  5. 5

    Date it

    Put the retrieval date on the table. Rerun quarterly with the same prompt.

The prompt · copy and paste
Compare the companies below using ONLY the page text I paste for each.

Build a table. Rows = companies. Columns:
- Who it is for (their words)
- Main promise (quote the headline)
- Pricing model (or "not published")
- Proof shown (logos, numbers, case studies — list what is there)
- Notably absent

After the table:
- Claims every company makes
- Claims only one company makes
Cite the page for each cell. Write "not stated" rather than inferring.

PAGES:
[Company A — homepage text, pricing text, product text]
[Company B — …]
Example output · fictional company
| Company | Who it is for | Main promise | Pricing | Proof |
| A | "operations teams at mid-size distributors" (home) | "Plan every route in minutes" (home) | Per vehicle / month (pricing) | 3 logos, 1 case study |
| B | not stated | "AI-powered logistics" (home) | not published | none shown |

Only one company claims: A — named customer results.
Before it goes out · a person checks

Review checklist

  • Every cell can be traced to a page you pasted.
  • "Not stated" appears where the page is silent.
  • The retrieval date is on the document.
  • Evaluation and opinion are in a separate section, written by you.
Don’t use it for

Where this workflow stops

  • Anything behind a login, or material obtained under NDA.
  • Market-share or revenue estimates — the model will guess.
  • A substitute for talking to customers who chose a competitor.

Part of AI for Marketing. Not sure your team is ready for workflows like this? Take the AI Readiness Assessment — 10 questions, no email. Want your team to build this on its own work? See how training works.

Related

Workflows that pair with this one

All workflows →
Marketing

Customer Interview Analysis with AI

Synthesise customer interviews into themes with quotes and counts, without the model inventing a pattern.

IntermediateClaudeChatGPT
Open workflow →
Sales

Prospect Research with AI

Turn a company website, a LinkedIn profile and your CRM notes into a one-page prospect brief before outreach.

BeginnerChatGPTClaude
Open workflow →
Leadership

Executive Brief Creation with AI

Build a one-page weekly brief for leadership from team updates: what changed, what is at risk, what needs a decision.

IntermediateCopilotChatGPT
Open workflow →

Bring one task.
Leave with it running.

Book a 30-minute call. You’ll have a recommended format and a written quote within 24 hours.

Book a Workshop

Or email [email protected] · text TRAINING to (469) 425-4142