AI SWOT Analysis: 4 Prompts to Run Yours (With Examples)

Table of Contents

• Traditional SWOT vs. AI SWOT
• Before you start: gather your data
• Prompt 1: strengths and weaknesses (internal analysis)
• Prompt 2: opportunities and threats (external environment)
• Prompt 3: competitor simulation
• Prompt 4: from matrix to plan (CAME + priority)
• Example 1: e-commerce in Latin America
• Example 2: B2B SaaS or digital service
• 7-day plan to do your SWOT with AI
• Mistakes that ruin an AI SWOT
• Next step

A SWOT analysis done with artificial intelligence works when you supply the data and the AI supplies the structure: it organizes the quadrants, crosses the factors and forces you to prioritize. When you just ask "make me a SWOT of my business" with no context, it returns a list of generalities that fits a bakery as well as a startup.

This guide is the AI version of our step-by-step SWOT analysis, which explains what goes in each quadrant and includes the manual template. Here you'll find four copy-ready prompts, two examples (a Latin American e-commerce store and a B2B SaaS) and a seven-day plan to put it into practice.

Traditional SWOT vs. AI SWOT

	Traditional SWOT	AI SWOT
Time	Days or weeks of meetings	One or two work sessions
Source	The team's perception	Your real data, challenged by the model
Bias	Confirmation: everyone defends their own turf	Less internal bias, but the model may invent context
Iteration	Done once and filed away	Rerun whenever the numbers change
Main risk	Stopping at the matrix	Accepting the answer without checking it

The real advantage isn't speed. It's that the AI has nothing to defend: if you give it an 8% margin and a break-even point you don't reach, it points that out bluntly.

Before you start: gather your data

An AI SWOT is only as good as the data block you paste in. Collect this before opening the chat:

• Business model: what you sell, to whom and through which channel.
• Finances: margin per product or service, break-even point and available cash. If you don't know your margin, start with the profit margin calculator.
• Customers: repurchase, cancellations (churn) or satisfaction, even if only estimates.
• Stack: the tools you use and how locked in you are to them.
• Competition: the three or four that really take sales from you, not the list of "everyone in the sector".

An important privacy note: don't paste customers' personal data or confidential information you wouldn't put into an external service. Aggregated figures are enough. If it worries you, first read the guide on AI security and privacy in your business.

Prompt 1: strengths and weaknesses (internal analysis)

This prompt audits what depends on you. Replace the bracketed parts with your data.

Act as a senior operations and financial auditor specialized in [e-commerce / SaaS / digital services].
Audit the internal strengths and weaknesses of my business using only the data I give you.

Rules:
1. Start with a summary of at most 30 words on the overall health of the business.
2. Deliver a table with the columns: Area | Strength or weakness | Factor | Impact (high/medium/low) | Supporting data.
3. If a factor has no supporting data, mark it as "assumption" instead of inventing a figure.
4. Evaluate these areas: processes and manual work, tech stack and vendor dependence, financial health (break-even point, separation of personal and business accounts) and customer retention.

My data:
- Business and model: [describe]
- Current stack: [tools]
- Finances (margin, break-even point, cash): [real or estimated figures]
- Customers (repurchase, churn, NPS): [figures]

Prompt 2: opportunities and threats (external environment)

Here you ask it to look outward with two well-known frameworks: PESTLE (political, economic, social, technological, legal, environmental) and Porter's Five Forces.

Act as a strategy consultant specialized in environment analysis.
Identify the external opportunities and threats of my business by applying PESTLE and Porter's Five Forces.

Rules:
1. Deliver at most 3 opportunities and 3 threats, ordered by impact on my business.
2. For each one: what it is, why it affects me (not the sector in general) and what data or signal I should verify.
3. Do not cite market figures or studies if you can't give me the source; in that case write "verify".
4. Clearly separate what comes from my data from what is your assumption.

Context:
- Sector: [sector]
- Target market: [country or region]
- Model: [B2B / B2C / D2C]
- Trends or obstacles I already perceive: [list]

Prompt 3: competitor simulation

This is the prompt most people skip and the one that helps most. You ask the AI to play against you.

Act as an aggressive competitor of my business. You know my value proposition and my weaknesses.
Propose 3 moves you would make in the next 6 to 12 months to take customers or margin from me.

For each move provide:
- What the competitor would do and which weakness of mine it exploits.
- Risk level (critical, high or medium).
- What I can do in under 30 days to defend myself.
- What counterattack would make sense at 6 months.

My value proposition: [describe]
Competitors (direct and indirect, including AI tools that could replace me): [list]
My current barriers to entry: [list]

Prompt 4: from matrix to plan (CAME + priority)

The CAME matrix turns each SWOT cross-pair into a decision: Correct weaknesses, Face threats, Maintain strengths, Exploit opportunities.

Cross-pair	Strategy	What you do
Strength + Opportunity	Offensive	Exploit the opportunity with what you do well
Strength + Threat	Defensive	Use your strengths to face the risk
Weakness + Opportunity	Reorientation	Correct what keeps you from seizing the market
Weakness + Threat	Survival	Reduce the weakness before the threat exploits it

With that framework, paste the results of prompts 1 to 3 and use this one:

Act as a chief operating officer. With the SWOT I paste below, generate the CAME cross-pairs (offensive, defensive, reorientation and survival) and sort the initiatives into an impact vs. effort matrix:
- Quick wins: high impact, low effort (under 14 days).
- Major projects: high impact, high effort (30 to 90 days).
- Fill-ins: low impact, low effort.
- Discard: low impact, high effort.

For quick wins and major projects deliver a table: Initiative | Owner | Resources (time, money, tools) | Tracking metric.
Do not propose more than 7 initiatives in total.

My SWOT:
[paste the results here]

The cap of seven initiatives isn't arbitrary: without a limit, the AI returns twenty actions and you execute none.

Example 1: e-commerce in Latin America

A summary of what the result might look like for an online store selling to several countries in the region (the points are an illustrative case, not data from a real store):

• Strengths: sales through social media and WhatsApp, a store built on a no-code platform with low fixed costs, catalog and promotions adapted to local dates.
• Weaknesses: doesn't calculate break-even per product, mixes personal and business cash, loses margin on cross-border payments and returns.
• Opportunities: personalized recommendations to raise repurchase, secondary cities where e-commerce is growing, automatic sequences to recover abandoned carts.
• Threats: global marketplaces lowering price expectations, local currency devaluation, tariff changes or logistics delays.

The cross-pair that weighs most here is weakness + threat: without knowing its break-even point, a devaluation or a tariff change can turn orders into losses before anyone notices. The survival action would be to calculate the real margin per product this week, with a margin calculator, and adjust prices before launching any new campaign.

Example 2: B2B SaaS or digital service

For a software startup, the SWOT has three common blind spots that AI helps you see:

• Costs and stack dependence. If you built on a no-code platform, check how it charges (by usage, credits or users), what happens to your bill if traffic multiplies, and whether you can export your code if you ever need to migrate. That dependence is a weakness even if everything works today.
• Retention by account, not by contact. In B2B you lose customers when a company stops using the product, not when one user unsubscribes. Usage alerts per company (for example, "this account hasn't created a project in 14 days") are worth more than a mass email.
• The generative AI threat. Vibe coding tools like Cursor lower the cost of building software, and anyone can clone your interface. What's left as an advantage is your own data, your integrations with the customer's workflow and how fast you solve their problem end to end. If you can't name which of the three you have, that is your main weakness.

For this case, the "quick win" is usually measuring and automating the alert for inactive accounts; the "major project," reducing vendor dependence or going deeper on a key integration. If you need to sort out your value proposition before the SWOT, the interactive Business Model Canvas gives you the whole map.

7-day plan to do your SWOT with AI

1. Day 1: gather the data block (finances, customers, stack, competition).
2. Day 2: run prompt 1 and fix the "assumptions" it flagged with real data.
3. Day 3: run prompt 2 and verify the three external claims that weigh most.
4. Day 4: run prompt 3 and note the two defenses you can start this month.
5. Day 5: build the CAME cross-pairs with prompt 4.
6. Day 6: filter with the impact vs. effort matrix and discard without guilt whatever doesn't fit.
7. Day 7: assign an owner, date and metric to each initiative and put it on your calendar.

Mistakes that ruin an AI SWOT

• Asking without data. Without your numbers, the result is generic.
• Accepting market figures without a source. Models can give percentages with great confidence and no basis. Verify whatever weighs on a decision.
• Stopping at the matrix. If it doesn't reach actions with an owner and a date, it's an exercise, not a strategy.
• Not repeating it. When your margin or main channel changes, or a competitor appears, rerun the prompts with the new data.

Next step

If you don't have your foundation clear yet, start with the step-by-step SWOT guide and come back here to speed it up with AI. Then define who you sell to with the buyer persona tool, and if you need to turn the plan into measurable goals, read OKR vs. KPI.