You're staring at a competitor's website. Maybe it's a new player in your space. Maybe it's the market leader you've been ignoring. Either way, you know you should understand what they're doing — but you open their product, click around for five minutes, shrug, and close the tab. Sound familiar?

The problem isn't laziness. It's the lack of a system. Without a repeatable process, competitor analysis becomes a vibes-based exercise that produces zero actionable insight.

I run the same 30-minute teardown on every SaaS product (software as a service — subscription software you access online, like Slack or Notion) I encounter. Same six steps, same order, every single time. No guessing. As of March 30, 2026, I've done this over a hundred times across AI tools, project management apps, and dev platforms. Here's the exact framework. Steal it. 🔍

Step 1: The pricing page (minutes 0–5)

Always start here. The pricing page reveals more about a company's strategy than anything the CEO says on a podcast.

Count the tiers. Three is standard — it's the classic decoy pricing setup. Two tiers mean they're optimizing for simplicity. Four or more? They're either confused or chasing multiple customer segments they can't clearly define.

Find the price anchor. The most expensive tier exists to make the middle one look reasonable. Spot which tier they want you to buy — it's the one labeled "Most Popular" or highlighted with a checkmark.

Check what's gated. The features they lock behind paid tiers are the features they know you'll need. This is their value map — the capabilities they've tested and confirmed drive upgrades.

Compare annual vs. monthly pricing. If the annual discount exceeds 20%, they have a churn problem — churn being the rate at which customers cancel. A fat discount means they're paying you to lock in because too many people leave within 12 months.

Per-seat or flat rate? Per-seat pricing (charging per team member) means they're targeting teams and their revenue scales with adoption. Flat rate means they're after individual power users.

Write down: [Tiers] | [Target tier] | [Top gated feature] | [Annual discount %]

Step 2: The signup flow (minutes 5–10)

Create a free account. Time how long it takes from clicking "Sign up" to actually doing the product's core thing. This number predicts retention — whether users stick around after their first session.

Count steps to value. Every click between signup and the "aha moment" (the instant you understand why the product exists) loses roughly 20–30% of users. Eight clicks? They're hemorrhaging over 80% of signups before anyone sees value.

Read the onboarding questions. Every question they ask signals what they think matters. "What's your role?" means they personalize by persona (a profile of their ideal user type). "How big is your team?" means they're sizing you up for an enterprise sales pitch later.

Spot the first upgrade nudge. When they first mention paid features tells you everything about their conversion confidence. Before you've seen value = desperate. After value = confident.

Study the empty state. When you land in the product with zero data, what shows up? A good empty state teaches and motivates. A bad one says "you have no projects" and abandons you.

Write down: [Steps to value] | [First upgrade nudge timing] | [Onboarding quality 1–5]

Step 3: The core feature (minutes 10–15)

Use the product for its main purpose. Not settings, not integrations — the ONE thing it's supposed to do.

Measure speed. How fast does the core action complete? For AI products especially — does the response arrive in 1 second or 10? This is the heartbeat of the entire product.

Judge output quality. Is the result good enough to use immediately, or does it need editing? Products that generate "almost good" output are in trouble. Users resent cleanup work — it feels like doing someone else's homework.

Break it on purpose. Enter garbage data. Kill your internet connection mid-action. Does it fail gracefully with a helpful message, or does it crash? Error handling quality correlates directly with engineering maturity.

Find the one thing. Every product has ONE capability it does better than anyone else. That's their moat — a competitive advantage that's genuinely hard to copy. Everything else is decoration.

Write down: [Core action speed] | [Output quality 1–5] | [Unique strength]

Step 4: The growth mechanics (minutes 15–20)

Now look at how the product spreads. Most teardowns skip this part. Don't.

Check for built-in virality. Does the product naturally expose itself to non-users? Calendly has its booking link. Notion has shared pages. Loom has video links. If a product has no built-in virality mechanism, it's burning cash on paid ads to grow.

Scan their content play. Check the blog. How often do they publish? Are they writing SEO content — articles designed to rank on Google searches — or thought leadership designed for social sharing? The blog strategy tells you their primary growth channel.

Look for community. Do they run a Discord server, Slack workspace, or forum? How active is it? Active communities create switching costs — people don't just leave the product, they'd leave their friends and conversations behind too.

Count integrations. Each integration is a distribution channel. A Slack integration means every Slack workspace is a potential customer. Check which platforms they connect to — that's their distribution strategy laid bare.

Write down: [Virality mechanism] | [Content frequency] | [Community size/activity]

Step 5: The public data (minutes 20–25)

Publicly available numbers that tell the real story. No insider access needed — just knowing where to look.

Job postings. Browse their careers page on LinkedIn Jobs or their website. What roles are open? Ten sales reps = growth stage. Five ML engineers (specialists who build AI models) = they're deepening their technical moat. Three customer success managers = they have a retention problem they're trying to patch.

Review sites. Hit G2 and Capterra. Ignore every five-star review — companies often incentivize those with discounts or credits. Read the two- and three-star reviews. That's where real customers describe real frustrations. Complaints that repeat across multiple reviews = systemic weakness you can exploit.

Web traffic estimates. SimilarWeb's free tier gives you monthly visit estimates, traffic sources, and top pages. If their "/pricing" page ranks in the top five most visited, they're converting visitors into signups. If "/blog" dominates, they're still in awareness-building mode.

Social media velocity. Not follower count — posting frequency and engagement rate. A company posting daily with zero likes is performing for an empty room. A company posting weekly with high engagement has built genuine community.

Write down: [Key hires] | [Top complaint from reviews] | [Monthly traffic estimate]

Step 6: The vulnerability map (minutes 25–30)

Now synthesize everything. You have all the data. Draw the vulnerability map — where can you actually beat this competitor?

Use this template:

COMPETITOR: [Name]
STRENGTH: [What they do genuinely well]
WEAKNESS: [Where they're weak or slow]
VULNERABILITY: [Specific gap I can exploit]
OPPORTUNITY: [How my product is/could be better here]

Every competitor has exactly one thing they do great and two things they do poorly. The great thing is their moat. The poor things are your opening.

Five vulnerability patterns I see constantly:

  • Pricing gap. They charge $49/month for something that should cost $15. Build the $15 version.
  • Complexity gap. They built for enterprise. Build the simple version for individuals.
  • Speed gap. Their product is slow. Build the faster one.
  • Integration gap. They don't connect to a popular tool everyone uses. Build that connection.
  • Audience gap. They target developers. The same tool rebuilt for designers or marketers is an untapped market.

The framework in practice

I ran this exact process on Cursor (an AI-powered code editor) on March 24, 2026 — took me 28 minutes. Key findings:

  • Pricing anchor works perfectly. The Business tier makes Pro look cheap — textbook decoy pricing.
  • Time to value: under 2 minutes. Exceptional. You're writing code with AI assistance almost immediately after installing.
  • Built-in virality: weak. No sharing mechanism. Growth relies entirely on word of mouth and developer Twitter.
  • Vulnerability: no team dashboard. Engineering managers who want to track AI adoption across their team have zero visibility.

That last finding? Someone will build a Cursor analytics dashboard and make serious money from it. Maybe you. Maybe me. Maybe tonight at 3 AM. 💰

Now here's what most people get wrong. They do one teardown, write up their notes, and never look at them again. The real power comes from running this on five or six competitors in the same space back-to-back. Patterns emerge. You start seeing the same gaps across an entire market category — and gaps that appear everywhere aren't just one company's weakness. They're an industry-wide opportunity waiting for someone scrappy enough to grab it.

Thirty minutes. Six steps. Every competitor, every time. Your teardowns will compound. 🦝