Over the past year, I've had 12 SaaS ideas that felt worth pursuing.
You know the feeling. You're in the shower, or walking the dog, or half-asleep, and suddenly: "Wait. What if there was a tool that..."
Historically, I'd either: a) Get excited, start building, lose steam after 2 weeks b) Add it to a notes app where ideas go to die c) Talk myself out of it before doing anything
This year, I tried something different. I scored each idea systematically, using the same framework, before writing any code.
Here's every idea, how it scored, what I learned, and the patterns that emerged.
The 12 Ideas
Idea #1: AI Meeting Notes for Sales Teams
The pitch: Record sales calls, auto-generate summaries, track commitments, update CRM.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 8 | Sales teams spend $50B+ on tools, clear budget | | Competition | 3 | Gong, Chorus, Fireflies, Otter—massively funded | | Technical | 6 | Doable, but reliable transcription is hard | | Monetization | 8 | Clear B2B value, existing budget line items | | Personal Fit | 4 | Never worked in sales, don't know the workflows | | Timing | 7 | AI transcription just got good enough |
Total: 5.85
What killed it: Competition score. The space has multiple $100M+ funded companies. Even if I built something great, I'd be fighting Gong's marketing budget.
Lesson: Big markets attract big players. "There's room for everyone" is usually a lie.
Idea #2: Budget Tracker for Freelancers
The pitch: Simple income/expense tracking designed for solo freelancers, not accountants.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 6 | Millions of freelancers, but fragmented | | Competition | 5 | Wave, FreshBooks, but they're complex | | Technical | 9 | Standard CRUD app, bank API integrations | | Monetization | 4 | Freelancers notoriously cheap, race to bottom | | Personal Fit | 7 | Been freelance, understand the pain | | Timing | 5 | Not new, not trending |
Total: 5.85
What killed it: Monetization score. Freelancers will use free tools forever. The successful players (Wave, FreshBooks) monetize through payments processing and upsells, not subscriptions.
Lesson: Your target customer's willingness to pay matters more than your product quality.
Idea #3: Developer Portfolio Generator
The pitch: Auto-generate beautiful portfolio sites from GitHub activity + LinkedIn data.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 5 | Mostly juniors and job-seekers, small market | | Competition | 6 | GitHub profile, personal sites, LinkedIn suffice | | Technical | 8 | GitHub API, templating, straightforward | | Monetization | 3 | One-time purchase at best, hard to justify subscription | | Personal Fit | 8 | Developer, understand the need | | Timing | 5 | Remote hiring peaked, market normalizing |
Total: 5.55
What killed it: Monetization and market size. The people who need portfolios most (juniors) have the least money. Seniors don't need them—their work speaks.
Lesson: If your target customer can't afford your product, no amount of product excellence helps.
Idea #4: Cold Email Sequence Builder
The pitch: AI-powered cold email generation with A/B testing and deliverability optimization.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 7 | Sales teams, agencies, founders—proven demand | | Competition | 4 | Lemlist, Apollo, Instantly—well-funded, feature-rich | | Technical | 7 | Email APIs, LLM integration, deliverability monitoring | | Monetization | 8 | Clear ROI story, existing budget categories | | Personal Fit | 5 | Have done cold outreach, not passionate about it | | Timing | 6 | AI adds new angle, but crowded |
Total: 6.10
What killed it: Competition + personal fit combo. The space is red ocean, and I'd need to love the problem to persist through the grind.
Lesson: Competitive markets require obsession. If you're "meh" about the problem, you'll quit.
Idea #5: SaaS Idea Validation Tool
The pitch: Systematic scoring, competitor research, market analysis—all in one place for founders.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 6 | Solo founders, indie hackers—passionate but niche | | Competition | 7 | Mostly spreadsheets, no dominant tool | | Technical | 8 | Web app + AI integrations, in my wheelhouse | | Monetization | 7 | Founders will pay for tools that save time | | Personal Fit | 10 | This IS my problem, I'd use this daily | | Timing | 8 | AI makes research 10x faster now |
Total: 7.45
What made it win: Personal fit was a 10. I was living the problem. Every time I had an idea, I wished this existed.
Lesson: Personal fit isn't just nice-to-have. It's what sustains you through the inevitable hard parts.
Idea #6: No-Code Internal Tool Builder
The pitch: Retool/Appsmith but simpler, for non-technical ops teams.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 8 | Enterprise internal tools, massive market | | Competition | 2 | Retool ($200M+ funding), Appsmith, Airplane | | Technical | 5 | Complex—need databases, permissions, integrations | | Monetization | 8 | B2B, seats-based, proven model | | Personal Fit | 6 | Built internal tools, understand the space | | Timing | 6 | Category is mature |
Total: 5.60
What killed it: Competition score destroyed it. Retool alone has 200+ engineers. I'd be competing with an army.
Lesson: "Simpler version of X" only works if X is truly overbuilt. Retool isn't.
Idea #7: AI Code Review for Solo Devs
The pitch: Get senior engineer-quality code reviews on your PRs, without a team.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 7 | Solo devs, small teams, bootcamp grads | | Competition | 5 | Some players, but none dominant | | Technical | 7 | LLM integration, GitHub API, manageable | | Monetization | 6 | Devs will pay for learning/improvement | | Personal Fit | 9 | Miss code reviews, would use constantly | | Timing | 9 | LLMs just got good enough for this |
Total: 7.10
Why it's strong: Timing + personal fit. LLMs crossed the quality threshold in 2024. The problem exists. No one's nailed the solution.
Lesson: Sometimes you're waiting for technology to catch up to the idea. Timing matters.
Idea #8: Newsletter Analytics Dashboard
The pitch: ConvertKit/Substack analytics are basic. Deep dive into what content performs.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 5 | Newsletter creators, growing but still niche | | Competition | 6 | Basic analytics built into platforms | | Technical | 6 | Need API integrations with every platform | | Monetization | 5 | Creators have limited budgets | | Personal Fit | 6 | Run newsletters, care about analytics | | Timing | 6 | Newsletter trend still growing |
Total: 5.65
What killed it: Monetization + market size. Creators with small lists can't afford tools. Creators with big lists have sponsors or are on enterprise plans with better analytics.
Lesson: The middle market (too big for free, too small for enterprise) is often a desert.
Idea #9: Micro-SaaS Directory
The pitch: Curated directory of profitable micro-SaaS with revenue data, tech stacks, founder stories.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 4 | Very niche, mostly aspiring founders | | Competition | 5 | Indie Hackers, Twitter threads, scattered info | | Technical | 8 | Content site, manageable | | Monetization | 4 | Directories monetize poorly, advertising is dying | | Personal Fit | 7 | Love the space, would enjoy curating | | Timing | 5 | Not new, not trending |
Total: 5.30
What killed it: Monetization. Directories make money through ads (declining), sponsorships (unreliable), or premium tiers (hard to justify). The business model is broken.
Lesson: Business model viability matters as much as product viability.
Idea #10: Customer Feedback Aggregator
The pitch: Pull reviews from G2, Capterra, Twitter, Reddit—unified sentiment dashboard.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 6 | Product teams, customer success teams | | Competition | 3 | Medallia, Sprinklr, Qualtrics—enterprise giants | | Technical | 6 | Many APIs, sentiment analysis, data normalization | | Monetization | 7 | B2B, clear value prop | | Personal Fit | 5 | Understand the need, not passionate | | Timing | 5 | Stable market, no new catalyst |
Total: 5.35
What killed it: Competition. The enterprise players have this locked. Going SMB means competing on price with razor margins.
Lesson: "Enterprise product but simpler/cheaper" rarely works. Enterprise buyers want enterprise features.
Idea #11: Changelog as a Service
The pitch: Auto-generate changelogs from Git commits + PRs, publish beautiful update pages.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 5 | SaaS companies, dev teams—specific use case | | Competition | 6 | Beamer, LaunchNotes, Canny—but none perfect | | Technical | 7 | GitHub integration, templating, webhooks | | Monetization | 6 | B2B, but low priority purchase | | Personal Fit | 6 | Would use, not obsessed | | Timing | 5 | Not new |
Total: 5.80
What killed it: Low priority purchase. Changelogs matter, but they're rarely someone's #1 problem. Hard to sell "nice to have."
Lesson: Products solving #5 priority problems compete with "do nothing" more than competitors.
Idea #12: AI-Powered PRD Generator
The pitch: Turn rough ideas into structured product requirement documents using LLMs.
| Dimension | Score | Reasoning | |-----------|-------|-----------| | Market | 6 | PMs, founders, product teams | | Competition | 5 | Notion AI, generic LLMs, but no specialized tool | | Technical | 7 | LLM prompting, structured output | | Monetization | 6 | PMs have budgets, but varies | | Personal Fit | 8 | Write PRDs constantly, would love this | | Timing | 8 | LLMs just became capable enough |
Total: 6.55
Why it's interesting: Could be a feature of the validation tool (#5), not a standalone product. Integration potential.
Lesson: Some ideas are better as features than products.
The Patterns That Emerged
After scoring 12 ideas, clear patterns appeared:
Pattern 1: Competition Is Usually Underestimated
8 of 12 ideas died primarily because of competition. I kept thinking "I can build something better" until I researched who I'd be competing against.
The fix: Research competitors FIRST, before falling in love with the idea.
Pattern 2: Personal Fit Predicts Persistence
The ideas I was most excited about (#5, #7) weren't necessarily the biggest markets. But they were problems I genuinely cared about solving.
The fix: Weight personal fit heavily. You'll need that enthusiasm in month 8 when nothing's working.
Pattern 3: Monetization Requires Existing Budgets
The ideas with clear monetization (#1, #4, #6) all targeted customers who ALREADY spend money on similar tools. The ones with weak monetization targeted customers who historically don't pay.
The fix: Don't try to create a new budget category. Capture an existing one.
Pattern 4: Timing Creates Windows
#5 and #7 both benefited from AI timing. The ideas weren't new, but the enabling technology just matured. That's a window.
The fix: Ask "why now?" for every idea. If there's no good answer, the timing might be wrong.
Pattern 5: Ideas Cluster Into Themes
Looking at my 12 ideas, they cluster:
- Developer tools (3)
- Sales/marketing tools (3)
- Creator tools (2)
- Internal tools (2)
- Founder tools (2)
That clustering revealed something about my interests and expertise. The founder tools cluster is where I have real insight.
The fix: Notice your idea patterns. They reveal where you should focus.
What I'm Building Now
I went with Idea #5 (SaaS Idea Validation Tool), incorporating elements of #12 (PRD generation) as a feature.
Why?
- Personal fit is a 10 (I use it daily)
- Timing is favorable (AI enables faster research)
- Competition is manageable (no well-funded dominant player)
- Monetization path is clear (founders pay for tools that save time)
- I can be customer zero
Three months in: 200 beta users, positive feedback, early revenue. Validation validated.
The Meta-Lesson
Scoring ideas isn't about finding the "best" idea objectively. The best idea for me isn't the best idea for you.
It's about:
- Forcing honest evaluation (no more "I'll just build and see")
- Comparing ideas systematically (not just gut feel)
- Identifying fatal flaws early (before you've invested months)
- Understanding why (not just which idea, but why it wins)
The scoring framework killed 9 ideas I might have otherwise wasted time on. It surfaced 3 worth pursuing. It gave me confidence to commit.
That's not a guarantee of success. But it's better than random selection.
Your Turn
You probably have ideas sitting in your notes right now. Maybe 5. Maybe 20.
Here's what I'd suggest:
- List them all (no filtering yet)
- Score each across the 6 dimensions
- Kill everything below 5.5 (ruthlessly)
- Deep-dive the survivors
- Pick the one with highest total AND highest personal fit
- Commit to validating it properly
The scoring takes a weekend. It could save you a year.
Ready to score your SaaS ideas systematically? Launchcrew is the tool I built to solve this exact problem—AI-powered validation, competitor analysis, and scoring so you can find your winning idea faster.