Most people using Claude for market research are leaving serious money on the table. They type something like “research my competitors” and get back a generic wall of text that reads like a Wikipedia summary. I’ve watched solopreneurs spend three hours reading that output, then still not knowing what to actually do next.
According to McKinsey’s 2023 report, generative AI could add $2.6–$4.4 trillion annually to global productivity.
The problem isn’t Claude. Claude is genuinely one of the best AI models available in 2026 for deep analytical work — its 200K token context window alone makes it a research powerhouse. The problem is the prompts. Weak input, weak output. It’s that simple.
I’ve spent the last several months building and refining a library of Claude prompts specifically for market research. I’ve used them for my own solopreneur projects, for client work, and for testing new niches before committing budget. The 25 prompts in this guide are the ones that consistently deliver results you can actually use — not just summaries, but frameworks, gaps, positioning angles, and actionable intelligence.
Copy them, paste them, tweak the brackets, and run them. Let’s get into it.
Why Claude Outperforms Other AI Models for Market Research
Before we get to the prompts, a quick word on why I specifically use Claude — not ChatGPT, not Gemini — for this type of work.
Claude (Claude 3.5 Sonnet and Claude 3 Opus on Claude.ai Pro, currently $20/month) handles long, nuanced reasoning chains better than most models I’ve tested. When you’re doing competitor analysis or customer psychographic research, you need a model that can hold multiple variables in tension and synthesize them — not just list bullet points.
Claude also tends to push back when your assumptions are shaky, which is genuinely useful in research contexts. I’ve had it flag logical gaps in my competitive positioning arguments that I would have missed completely.
Here’s a quick comparison of how the main models stack up for market research tasks specifically:
| Feature | Claude 3.5 Sonnet | ChatGPT-4o | Gemini 1.5 Pro |
|---|---|---|---|
| Context window | 200K tokens | 128K tokens | 1M tokens |
| Analytical depth | ★★★★★ | ★★★★☆ | ★★★★☆ |
| Nuanced reasoning | Excellent | Good | Good |
| Handles pasted research docs | Yes (large files) | Yes | Yes |
| Price (Pro plan) | $20/month | $20/month | $19.99/month |
| Best for | Deep analysis, synthesis | Broad research, browsing | Very long documents |
One practical tip: paste your raw research (Reddit threads, review excerpts, competitor website copy) directly into Claude. Its large context window means you can dump a lot of source material in one go and ask it to synthesize across all of it.
Category 1: Competitor Analysis Prompts
These prompts work best when you already have some raw material — competitor website copy, pricing pages, or even just their product names. Paste that context in first, then run the prompt. The more specific your inputs, the sharper the analysis.
Prompt 1: Competitive Positioning Map
When to use it: Early in your research, when you’re trying to understand where the white space is in a market. Great for validating a new product idea or repositioning an existing one.
You are a senior market strategist. I'm going to describe [NUMBER] competitors in the [MARKET/NICHE] space. For each one, identify:
1. Their primary positioning statement (what they claim to be best at)
2. Their ideal customer profile based on their messaging
3. Their apparent pricing tier (budget / mid-market / premium)
4. One clear weakness or gap in their offering
After analyzing all competitors, give me a 2x2 positioning matrix using axes that make strategic sense for this market. Recommend where a new entrant could position themselves to avoid direct competition.
Here are the competitors: [PASTE COMPETITOR INFO OR NAMES]
Prompt 2: Reverse-Engineer a Competitor’s Strategy
When to use it: When you want to understand why a competitor is winning, not just what they’re doing. This is particularly useful when a competitor has grown fast and you can’t figure out their actual growth lever.
Act as a business intelligence analyst. Based on the following information about [COMPETITOR NAME] — including their website copy, pricing, social presence, and any reviews I'll paste below — reverse-engineer their likely go-to-market strategy.
Specifically identify:
- Their primary acquisition channel (organic, paid, partnerships, etc.)
- The core pain point they're solving that justifies customer switching costs
- Their retention mechanism (what keeps customers from leaving)
- Any pricing or packaging tactic that appears intentional and strategic
Be direct. If you're making an inference, say so. Don't pad the analysis.
[PASTE COMPETITOR INFO HERE]
Prompt 3: Competitive Gap Finder
When to use it: After you’ve gathered a set of customer reviews (from G2, Capterra, Amazon, Trustpilot, or Reddit) for two or more competitors. Paste them in and let Claude find the patterns you’d miss reading manually.
I'm going to paste customer reviews for [COMPETITOR A] and [COMPETITOR B]. Read all of them carefully.
Your job: identify the top 5 unmet needs that appear repeatedly across both sets of reviews — things customers want but aren't getting from either product.
For each unmet need:
- Quote 1-2 specific review excerpts that demonstrate it
- Estimate how frequently this complaint appears (rare / common / very common)
- Suggest how a new product or service could address this gap specifically
[PASTE REVIEWS HERE]
Category 2: Customer Psychographic and Persona Research
This is where Claude genuinely shines compared to other tools. Building a persona from survey data or forum posts usually takes a skilled researcher hours. These prompts compress that work dramatically — but only if you give Claude real source material to work with. Don’t ask it to invent a persona from scratch; feed it real customer language.
Prompt 4: Voice-of-Customer Synthesis
When to use it: When you have a pile of raw customer quotes — from interviews, surveys, Reddit threads, or reviews — and need to find the real patterns fast.
I'm going to paste raw customer quotes and comments about [PRODUCT/SERVICE/PROBLEM]. These come from [SOURCE: Reddit / customer surveys / review sites / interviews].
Read every quote carefully. Then produce:
1. The top 3 emotional drivers behind purchases in this category (what people are really buying — the feeling or outcome, not the feature)
2. The top 3 fears or anxieties that delay or prevent purchase
3. The exact words and phrases customers use to describe their problem (I will use these in my marketing copy)
4. Any surprising insight that wouldn't show up in a typical survey
Do not summarize generically. Use the actual language from the quotes.
[PASTE QUOTES HERE]
Prompt 5: Ideal Customer Profile Builder
When to use it: When you’re launching something new and need to prioritize which customer segment to go after first. This prompt forces Claude to make a recommendation, not just list options.
I'm building [PRODUCT/SERVICE DESCRIPTION]. Based on what I've described, I want you to define three distinct customer segments who might buy this.
For each segment, provide:
- Demographic snapshot (age range, job title or life situation, income level)
- Primary motivation for buying
- Biggest objection before buying
- Where they spend time online (specific platforms, subreddits, communities)
- A realistic willingness-to-pay range
After presenting all three, make a clear recommendation on which segment I should prioritize first and explain your reasoning in 3-4 sentences. Be direct — don't hedge.
Prompt 6: Psychographic Deep Dive from Reddit Data
When to use it: After spending 30 minutes reading a relevant subreddit and copy-pasting the most interesting posts and comments into a doc. This is one of my favorite research combos — Reddit raw material + Claude synthesis.
Below are posts and comments from [SUBREDDIT NAME] related to [TOPIC/PROBLEM]. These are real people talking unfiltered about their experiences.
Analyze this content and give me:
1. The core identity these people hold about themselves (how they see themselves, not how marketers see them)
2. What they brag about vs. what they complain about
3. The status markers in this community (what earns respect, what causes ridicule)
4. Three marketing angles that would resonate with this group — and one approach that would immediately alienate them
[PASTE REDDIT CONTENT HERE]
Prompt 7: Jobs-to-Be-Done Analysis
When to use it: When you want to understand not just who your customer is, but what “job” they’re hiring your product to do. Based on the JTBD framework popularized by Clayton Christensen — Claude applies it well with the right prompt.
Apply the Jobs-to-Be-Done framework to [PRODUCT/SERVICE/CATEGORY].
For this product, identify:
1. The functional job (the practical task it accomplishes)
2. The emotional job (how it makes the customer feel about themselves)
3. The social job (how it affects how others perceive the customer)
4. The "struggling moment" — the specific trigger that causes someone to seek out this solution
5. What they were doing before this product existed (the workaround they abandoned)
Then suggest one product feature or messaging angle that addresses the emotional or social job but is currently missing from most competitors in this space.
Category 3: Market Sizing and Opportunity Assessment
Claude can’t browse the web in real time (unless you’re using it with a tool like Perplexity or with Claude’s web access enabled in certain configurations). But it’s excellent at helping you structure a market sizing approach, sanity-check your assumptions, and build bottom-up estimates from numbers you feed it.
Prompt 8: Bottom-Up Market Size Calculator
When to use it: When you have some rough data points (number of businesses in a category, average spend, conversion assumptions) and want Claude to build a structured TAM/SAM/SOM analysis.
Help me build a bottom-up market sizing model for [BUSINESS IDEA/PRODUCT CATEGORY].
Here are the data points I have:
- [DATA POINT 1, e.g., "There are approximately 500,000 independent consultants in the US"]
- [DATA POINT 2, e.g., "Average spend on productivity tools is $150/year"]
- [DATA POINT 3, e.g., "Estimated 15% are actively looking for this type of solution"]
Using these inputs, calculate:
1. Total Addressable Market (TAM) — everyone who could theoretically buy
2. Serviceable Addressable Market (SAM) — realistic reach given my business model
3. Serviceable Obtainable Market (SOM) — realistic Year 1-3 capture based on a solo operator
Flag any assumption you think is overly optimistic and suggest a more conservative figure.
Prompt 9: Niche Viability Stress Test
When to use it: Before you commit to a niche. This prompt acts like a devil’s advocate — it’s specifically designed to find the holes in your plan before you do.
I'm considering entering the [NICHE] market with [PRODUCT/SERVICE DESCRIPTION] priced at approximately [PRICE POINT].
Play the role of a skeptical but fair-minded investor. Ask me the five hardest questions about this business idea — questions that would expose fatal flaws in the market opportunity, competitive dynamics, or customer acquisition assumptions.
After listing the questions, give your honest assessment of the one risk that would most likely kill this business in the first 12 months. Be direct and specific. Don't soften it.
Prompt 10: Trend Signal Scanner
When to use it: When you’ve collected a batch of recent news headlines, industry reports, or social media posts about a category and want Claude to extract the signal from the noise.
I'm going to paste a collection of news headlines, report excerpts, and social posts about [INDUSTRY/CATEGORY] from the past 6-12 months.
Analyze them for:
1. The 3 strongest trend signals (things that are clearly accelerating, not just noise)
2. Any countertrends or things slowing down that most people are ignoring
3. A "leading indicator" — something in this data that suggests where the market will be in 18-24 months
4. One emerging customer need that isn't yet being served by an established product
[PASTE CONTENT HERE]
Category 4: Pricing and Positioning Research
Pricing is where most solopreneurs lose money — either by undercharging because they’re scared, or by overpricing a product that hasn’t earned premium positioning yet. These prompts help you think through pricing strategy with real market data as input.
Prompt 11: Pricing Model Comparison for Your Market
When to use it: When you’re deciding between pricing models (subscription vs. one-time, per-seat vs. flat-rate, freemium vs. paid-only) and want a structured analysis based on what’s working in your specific market.
I'm launching [PRODUCT/SERVICE] in the [MARKET] space. Here are the pricing models used by my main competitors:
[LIST COMPETITOR PRICING, e.g., "Tool A: $29/month subscription, Tool B: $299 one-time, Tool C: freemium with $49/month Pro"]
Analyze these models and tell me:
1. What the dominant pricing model in this space signals about customer buying behavior
2. Whether there's a pricing model that's conspicuously absent and why that might be an opportunity
3. The psychological pricing sweet spot based on these benchmarks
4. A recommended pricing structure for my product with a rationale
My product differentiates by: [YOUR KEY DIFFERENTIATOR]
My target customer is: [CUSTOMER DESCRIPTION]
Prompt 12: Value Metric Identification
When to use it: When you’re not sure what your pricing should be tied to — time, usage, seats, outcomes. Getting the value metric right is the single biggest lever in SaaS and service pricing.
Help me identify the right "value metric" for [PRODUCT/SERVICE] — the unit of value that aligns what customers pay with the value they receive.
Here's how the product works: [DESCRIPTION]
Here's what outcome it delivers: [OUTCOME]
Here's who uses it: [USER DESCRIPTION]
Give me three potential value metrics, rank them, and explain:
- Which one most closely tracks the value customers actually receive
- Which one is easiest to measure and communicate
- Which one would create the most natural upsell path as customers grow
Recommend one and explain why in plain terms.
Prompt 13: Positioning Statement Generator
When to use it: When you have your research done and need to translate it into a sharp positioning statement — for your homepage hero, your pitch, or your ad copy.
Based on the following market research context, write 5 distinct positioning statements for [PRODUCT/SERVICE].
Context:
- Target customer: [DESCRIPTION]
- Primary pain point: [PAIN]
- Key differentiator vs. competitors: [DIFFERENTIATOR]
- Proof point or result: [SPECIFIC RESULT, e.g., "saves 5 hours/week"]
Each positioning statement should follow this format:
"For [customer], [product] is the [category] that [key benefit] unlike [alternative], because [reason to believe]."
After writing all five, rank them from most to least compelling and explain your top pick in 2 sentences.
Category 5: Survey Design and Interview Analysis
Primary research — talking to real customers — is still the gold standard. But Claude can make your survey design sharper and your interview analysis much faster. These prompts saved me probably 4 hours on a recent product validation project.
Prompt 14: Customer Survey Builder
When to use it: Before sending a survey to your list or running one on a platform like Typeform or Google Forms. Weak survey questions get you garbage data — this prompt helps you avoid common biases.
I need to design a customer survey to validate [PRODUCT IDEA / RESEARCH QUESTION].
Create a 10-question survey that:
- Starts with behavioral questions (what people actually do) before attitudinal ones (what they think)
- Avoids leading questions and social desirability bias
- Includes at least 2 open-ended questions that are likely to generate usable qualitative data
- Has a logical flow from broad context to specific product questions
For each question, briefly note what insight it's designed to capture and any common mistake to avoid when asking it.
My target respondent: [DESCRIPTION]
My core hypothesis to validate: [HYPOTHESIS]
Prompt 15: Customer Interview Transcript Analyzer
When to use it: After running customer discovery interviews. Paste the transcript in and let Claude pull out the insights — this is one of the highest-leverage uses of the 200K context window.
I'm going to paste a transcript from a customer discovery interview about [TOPIC].
After reading the full transcript, extract:
1. The interviewee's actual problem (which may differ from what they say their problem is)
2. Any "when X happens, I wish I could..." moments — these are product opportunities
3. The exact language they use to describe their situation (3-5 verbatim phrases)
4. Any signal about their current solution and why it's inadequate
5. One surprising thing they said that challenged my assumptions
Do not infer what I want to hear. Flag anything that contradicts my hypothesis.
[PASTE TRANSCRIPT]
Prompt 16: Pattern Finder Across Multiple Interviews
When to use it: When you’ve done 5-10 interviews and need to synthesize them into clear patterns. Paste all the summaries or key excerpts at once.
I've conducted [NUMBER] customer interviews about [TOPIC]. Below are summaries or excerpts from each one.
Read all of them and identify:
1. The 3 themes that appear in at least half the interviews
2. Any outlier perspective that only one or two people expressed but might be important
3. The one finding that most strongly suggests a clear product or messaging direction
4. Any assumption I came in with that the interviews failed to confirm
Present your findings as a research brief I could share with a collaborator — structured, specific, and referenced back to specific interview excerpts.
[PASTE INTERVIEW SUMMARIES]
Category 6: Content Gap and SEO-Aligned Market Research
Market research and content strategy overlap more than most people realize. Understanding what your market is searching for, what questions go unanswered, and what content is currently ranking tells you a lot about unmet demand. These prompts connect those dots.
Prompt 17: Content Gap Identifier from SERP Analysis
When to use it: After you’ve manually reviewed the top 5-10 search results for your target keyword and want Claude to identify what’s missing — what searchers want that none of the current results fully deliver.
I've reviewed the top search results for the keyword "[TARGET KEYWORD]". Below are summaries or excerpts from the top-ranking articles.
Analyze them and tell me:
1. What topic angles or subtopics none of them cover adequately
2. What type of content searchers probably want that isn't there (comparison table, step-by-step guide, specific use cases, real pricing data, etc.)
3. The implicit question behind this search that no article fully answers
4. A specific content angle I could take that would be genuinely more useful than what's currently ranking
[PASTE SERP SUMMARIES OR CONTENT EXCERPTS]
Prompt 18: “People Also Ask” Pattern Analyzer
When to use it: After you’ve exported a list of “People Also Ask” questions from Google or questions from AnswerThePublic/AlsoAsked for your target market. This prompt turns a raw list into a research map.
Here is a list of questions people are searching for related to [TOPIC/MARKET]:
[PASTE QUESTION LIST]
Analyze these questions and:
1. Group them into 4-6 thematic clusters
2. Identify which cluster represents the highest buyer intent
3. Identify which cluster reveals the deepest confusion or uncertainty in this market
4. Suggest one product, service, or content asset that could directly answer the most common unresolved question
Also flag: are there questions here that indicate people are frustrated with existing solutions? Quote the specific questions.
Prompt 19: Audience Language Mining from Review Data
When to use it: Specifically for ad copy and landing page writing. The words people use in reviews are the words that convert — this prompt extracts them systematically.
I'm going to paste Amazon/G2/Capterra reviews for products in the [CATEGORY] space.
Your job is to extract the exact language I should use in marketing copy. Specifically:
1. The top 10 "before" phrases — how people describe their situation before finding a solution (e.g., "I was spending hours every week...")
2. The top 10 "after" phrases — how people describe results (e.g., "Now I can...")
3. The most emotionally loaded words that appear repeatedly
4. Any specific numbers or metrics customers cite as proof of value
5. Three headline formulas based directly on this language
Present everything as a copy swipe file I can reference when writing ads or landing pages.
[PASTE REVIEWS]
My Real-World Experience
A few months back, I had a seller in Funchal asking me why her apartment wasn’t moving. She’d had it listed for 11 weeks, two viewings, zero offers. Before I could give her an honest answer, I needed data — not just a gut feeling based on driving around the area. I opened Claude, described the property (T2, 78m², older building, no sea view, Largo da Sé zone), and used one of the competitor analysis prompts from this list to structure my research questions. Within about 40 minutes, I had a working framework for a neighbourhood price trend report that would normally take me half a day to build from scratch. I used it to pull together publicly available data from Idealista and Imovirtual, organise it into a proper CMA narrative, and present it to the client the next morning. She dropped the price by €12,000. It sold in three weeks.
That’s the kind of use case where these prompts actually earn their keep. I’ve now been testing Claude prompts specifically for market research tasks for around 14 weeks, and I use them almost daily — competitor pricing summaries, buyer persona breakdowns for specific Madeiran neighbourhoods like Monte or Caniço, and demand-signal reports I put together for investors asking about the short-term rental market near the airport corridor. What used to take me 3 to 4 hours now takes closer to 45 minutes if I have the right prompt structure going in.
That said, there’s a real limitation I keep running into: Claude doesn’t have live access to property listing data or local MLS figures. It can help me structure the analysis and write the narrative around numbers I feed it, but I still have to do the actual data gathering manually from Portuguese portals. If you come in expecting it to spit out current price-per-square-metre stats for São Martinho, you’ll be disappointed. It’s a thinking and writing tool, not a data source.
If this article carried a rating, I’d give these prompts a solid 4.2 out of 5 — they genuinely compress the research-to-report pipeline for solo agents who are doing everything alone and can’t afford to spend a full workday building one CMA.
Bottom line: if you’re a solo agent managing your own research, listings, and client comms without an assistant, these prompts will save you real hours every week. Just go in with your own data ready — Claude will help you make sense of it faster than you can alone.
“`Bonus: 6 Advanced Claude Prompts for Deeper Market Research
These are the prompts I use when I want to go beyond standard analysis — stress-testing strategy, synthesizing cross-category insights, or pressure-testing a business model before pitching it.
Prompt 20: Second-Order Market Effect Analyzer
When to use it: When a major industry shift is happening (new regulation, new technology, a competitor getting acquired) and you want to think through second and third-order effects on your market.
A significant change is happening in the [INDUSTRY] market: [DESCRIBE THE CHANGE].
Think through the second and third-order effects of this change on:
1. Customer behavior (what will they do differently in 6, 12, 24 months?)
2.
Robson Penassi
Real estate consultant in Madeira, Portugal. Solopreneur since 2012. Testing AI tools since 2023 to automate his one-person business. Writes about what actually works — and what does not.
More articles by Robson →