I spent 11 years doing competitive analysis the hard way — pulling numbers from property listings by hand, copying competitor pricing into spreadsheets, and writing market summaries at midnight before client meetings. Last quarter I ran a full competitive analysis on 23 properties across four Madeira neighborhoods using Claude. It took me 94 minutes. The same job used to eat an entire afternoon. That gap is why I’m writing this tutorial.
This is a step-by-step walkthrough of exactly how I use Claude AI for competitive analysis in my real estate business — the specific prompts, the settings, the workflow order, and the one part where Claude still lets me down. If you run a solo operation and need actionable competitor intelligence without hiring an analyst, this is the process.
What You’ll Build by the End of This Tutorial
By following these steps, you’ll have a repeatable Claude AI workflow that produces:
- A structured competitor profile for each rival listing or business
- A side-by-side feature and pricing comparison table
- A gap analysis identifying what competitors are missing
- A positioning summary you can drop straight into a client presentation or your own marketing
The whole sequence runs inside Claude.ai with no paid plugins required. I use Claude Pro ($20/month) because the longer context window matters when you’re pasting in multiple competitor descriptions at once — but I’ll flag which steps work fine on the free tier.
Prerequisites Before You Start
- A Claude.ai account (free tier works for Steps 1–4; Pro recommended for Steps 5–7)
- Raw competitor data: listing descriptions, pricing, amenities, or marketing copy you’ve collected manually or via a site like Idealista, Rightmove, or your local MLS equivalent
- A clear picture of your own offer — what you sell, at what price point, to whom
- A Google Doc or Notion page open to paste outputs as you go
You do not need any Claude tools or extensions. Everything here runs in a standard browser chat window.
Step 1: Set the Context With a System-Level Brief
Claude performs better when it understands the business context before you throw data at it. Open a new conversation and paste this as your first message — edit the bracketed fields for your situation.
You are a competitive analysis assistant for a solo real estate consultant
based in [YOUR LOCATION]. My business focuses on [YOUR NICHE — e.g., luxury
holiday rentals, urban apartments under €300k, commercial units].
My typical client is [BRIEF CLIENT PROFILE].
Throughout this conversation I will give you competitor data. Your job is to
help me identify patterns, gaps, and positioning opportunities. Be direct.
Use bullet points and tables. Flag anything that looks like a pricing anomaly
or a marketing weakness I could exploit. Do not summarize what I've already
told you — just analyze.This brief does two things: it stops Claude from giving generic advice, and it tells it to skip the filler and go straight to useful output. I keep a saved version of this in Notion so I’m not retyping it every session.
Step 2: Feed In Competitor Data One Block at a Time
Do not dump everything in one message. I learned this the painful way — Claude handles structured analysis better when you introduce competitors sequentially and ask it to hold the data before comparing.
For each competitor, use this format:
Competitor [NUMBER]: [NAME OR REFERENCE CODE]
Listing price / fee / rate: [PRICE]
Property type / service type: [TYPE]
Key selling points mentioned in their listing or marketing:
[PASTE THEIR COPY HERE — listing description, About page, social bio, etc.]
Notes I've observed: [YOUR OWN NOTES — e.g., "always drops price after 30 days",
"no mention of management fees", "photos look outdated"]
---
Do not analyze yet. Confirm you've stored this competitor profile.That last line is important. Telling Claude to wait before analyzing means it won’t start drawing conclusions from incomplete data. Once I’ve loaded all competitors, I give the command to compare — not before.
Step 3: Build the Competitor Comparison Table
After you’ve loaded all your competitors, send this prompt:
Now create a competitor comparison table using all profiles I've provided.
Columns should include: Competitor Name, Price Point, Primary Selling Angle,
Target Client Type, Weaknesses I Should Know About, and Missing Elements
(things they don't mention that buyers/clients often care about).
Format it as a markdown table.Claude will generate a clean table you can paste directly into a Google Doc or Notion. If you need HTML for a client deliverable, just follow up with: “Now convert that table to HTML.” Takes about four seconds.
Here’s an example of the kind of output structure you’ll get — simplified for illustration:
| Competitor | Price Point | Main Selling Angle | Target Client | Key Weakness | What They Miss |
|---|---|---|---|---|---|
| Agency A | €450k avg | Ocean views, luxury finishes | Northern European retirees | No local area context | Rental yield data, running costs |
| Agency B | €280k avg | Value for money, fast process | First-time buyers | Generic photography | Post-sale support, residency info |
| Independent C | €320k avg | Personal service, local knowledge | Returning expats | No online presence | Digital documents, remote signing |
Step 4: Run a Gap Analysis Prompt
The comparison table shows you what’s there. The gap analysis tells you what’s missing — and that’s where the real opportunity lives. Send this next:
Based on all the competitor profiles and the table you just built, identify:
1. The three most common weaknesses across all competitors
2. Topics or client concerns that NO competitor addresses in their marketing
3. Price positioning gaps (segments that appear underserved)
4. One specific angle I could own that none of my competitors are currently using
Be specific. Don't give me generic marketing advice — base everything on
the actual competitor data I've provided.That fourth question is the one I find most valuable. When I ran this for my Madeira portfolio analysis, Claude identified that none of my local competitors were explicitly addressing the Golden Visa transition — what happened to investor clients after the scheme changed. I built a FAQ around that and it became my most-clicked content that quarter.
Step 5: Generate Your Positioning Statement
Now you use the gap analysis to write your own positioning. This is where Claude earns its keep for solo operators — it takes everything you’ve fed it and helps you articulate where you fit.
Using the gap analysis and competitor data, write three versions of a
positioning statement for my business. Each should be 2-3 sentences.
Each version should emphasize a different competitive angle from the gaps
you identified.
Write for a client who is considering multiple agents and hasn't decided yet.
Tone: confident but not aggressive. No clichés like "your dream home" or
"passion for real estate."You’ll get three solid drafts in under 30 seconds. I usually take one, edit it for my voice, and it ends up on my website or in my email signature within the same day.
Step 6: Build a Client-Ready Report in One Pass
If you need to present this to a client — a seller who wants to know how their property compares, or an investor evaluating the market — you can generate a formatted report with one more prompt. This is where Claude Pro’s longer output capacity matters.
Write a client-facing competitive market analysis report using all the data
from this conversation. Structure it as follows:
1. Market Overview (3-4 sentences summarizing the competitive landscape)
2. Competitor Profiles (brief paragraph on each competitor — no more than
4 sentences each)
3. Comparative Analysis Table (reuse the table from earlier)
4. Market Gaps and Opportunities (bullet list, plain language)
5. Recommended Positioning for [MY BUSINESS NAME]
6. Next Steps (3 concrete actions the client should consider)
Tone: professional but readable. Avoid jargon. This will be read by
a client who is not a real estate expert. Length: 600-800 words.Export that output, paste it into a Google Doc, add your logo, and you have a deliverable. I’ve sent versions of this report to clients within the same hour I started the analysis.
Step 7: Save the Workflow as a Reusable Template
This is the step most people skip, and it’s the reason they end up reinventing the wheel every month. After you finish your first analysis, do this:
- Copy all six prompts into a single Notion page or Google Doc titled “Claude Competitive Analysis — Master Prompts”
- Note which competitor data format worked cleanest (the structured block from Step 2)
- Save one example of Claude’s gap analysis output so you can show clients what they’re getting
- Set a recurring calendar reminder to run the analysis quarterly — competitor landscapes shift faster than most solo operators realize
I run this workflow every 90 days for Madeira and every 60 days for the Lisbon market I started tracking in late 2025. Each run takes me under two hours including data collection.
My Real-World Experience Running This in Madeira
In February 2026 I needed to prepare a competitive analysis for a client — a German couple relocating from Frankfurt who wanted to understand whether the property I was recommending was fairly priced relative to similar listings in the Câmara de Lobos area. They were thorough people. They’d already done their own research and came to the meeting with printed Idealista screenshots. I had 48 hours to produce something more useful than a printout.
I collected data on 14 comparable listings across four agencies and two independent sellers. Using the exact workflow above — starting with the context brief, loading competitors one by one, then running the comparison and gap analysis — I had a structured first draft of the report within 90 minutes. That included the time I spent collecting the raw data manually, which Claude obviously can’t do for me.
The gap analysis flagged something I hadn’t consciously noticed: three of the five competing listings mentioned “sea views” in the headline but buried the actual view angle (partial, from upper floor only) in paragraph three or four. My recommended property had a direct frontal view from the living room. Claude suggested I lead with that distinction and call it out explicitly in comparison rather than assuming clients would notice it themselves. I rewrote two paragraphs of the property description based on that one insight.
The client meeting lasted 40 minutes. They made an offer the next day. I’m not attributing the sale entirely to Claude — the property was genuinely good — but the analysis gave the clients confidence they weren’t overpaying, and it gave me a structured, defensible answer to every comparison question they asked. Before I started using this workflow, that kind of preparation would have taken me four to five hours and still felt somewhat improvised. This time it took 90 minutes and felt solid.
I’ve now run this process 11 times across different property segments in Madeira — short-term rentals, urban resale apartments, and rural quinta properties. The average time per analysis has dropped from about 4.5 hours (my old manual method) to under 2 hours including data gathering. That’s roughly 27 hours recovered across those 11 sessions. For a solo operator, that’s more than half a working week handed back to me.
Where Claude AI Falls Short in Competitive Analysis
I want to be straight about this because the limitation is significant: Claude cannot go get data for you.
It has no ability to browse Idealista, pull listing prices from Rightmove, or scrape competitor websites in real time. Every piece of competitor data in this workflow comes from manual collection before you open Claude. If a competitor updated their pricing yesterday, Claude has no idea — because you’re the one feeding it the data.
This matters more than it sounds. I’ve seen people get excited about “AI competitive analysis” expecting some kind of automated intelligence sweep of the market. That’s not what this is. What Claude does extremely well is process and structure data you’ve already collected — it finds patterns, writes coherent comparisons, and surfaces gaps you’d likely miss reading the same listings for the third time at 10pm. But the legwork is still yours.
I’ve experimented with pairing Claude with Perplexity AI (which does have web access) to handle the data-gathering layer — but that’s a separate workflow for a separate article. For now, just know that this tutorial assumes you arrive at Step 1 with real competitor data already in hand.
Troubleshooting: When Claude’s Analysis Misses the Mark
Problem: The gap analysis is too generic.
Fix: Your competitor data probably wasn’t specific enough. Add price-per-square-meter figures, exact amenity lists, or quoted phrases from their marketing copy. The more granular the input, the sharper the output.
Problem: Claude keeps repeating the same weakness for every competitor.
Fix: Add a line to your gap analysis prompt: “Do not repeat the same weakness across multiple competitors unless it is genuinely universal — in that case, flag it as a market-wide pattern.” That forces it to differentiate.
Problem: The positioning statement sounds like every other real estate agency.
Fix: Add a line to the prompt: “Avoid any phrase that could appear in a generic real estate agency description. The positioning must be specific to the gaps you identified in this analysis.” Then also tell it your one most distinctive credential — mine is 14 years of local market data in Madeira, which most competitors can’t match.
Problem: The client report is too long or uses too much jargon.
Fix: After generating the report, send: “Rewrite this at a reading level suitable for someone who reads a quality newspaper but has no real estate background. Cut any sentence that uses industry terms without immediately explaining them.” Works every time.
Quick Reference: The Full Prompt Sequence
| Step | What You Send Claude | What You Get Back | Works on Free Tier? |
|---|---|---|---|
| 1 | Context brief | Confirmation + framing | Yes |
| 2 | Competitor data blocks (one at a time) | Stored profiles | Yes (up to ~5 competitors) |
| 3 | Comparison table request | Markdown or HTML table | Yes |
| 4 | Gap analysis prompt | Weakness list + opportunity angle | Yes |
| 5 | Positioning statement request | 3 draft positioning statements | Yes |
| 6 | Client report request | 600-800 word formatted report | Pro recommended |
| 7 | Save prompts to Notion/Doc | Reusable template
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 → |