Video summary

Give Me 36 Minutes and I'll Teach You How to Find $1M SaaS Ideas

Main summary

Key takeaways

Business

Summary (business-focused)

Core claim & dataset

  • Choosing the right SaaS idea can lead to outcomes like $10k, $100k, or $1M/month—instead of “nothing at all.”
  • The speaker uses a research-backed approach:
    • Surveyed 200+ SaaS founders
    • All reported making at least $1/month (many were well above $100k/month)
    • Goal: identify how founders found profitable SaaS opportunities
  • Prevalence of idea-finding sources:
    • Method 1 (day job problem): 48%
    • Method 2 (copy/enter existing category): ~10%
    • Method 3 (freelancing): ~10%
    • Method 4 (solving spouse/friend problem): ~8%
    • Method 5 (online problem search): ~3%
    • Method 6 (scratch your own itch): 16%
    • Method 7 (emerging tech): ~3%
    • Method 8 (build on existing product): 4%
  • Final takeaway:
    • 72% found their idea through work experience (day job, incumbent pain, agency clients, or while building something else).

The 8 founder “idea discovery” methods (with tactics)

1) Find a problem at your day job (48%)

How it works (framework-like checklist)

  • Look for:
    • Tools you use daily that frustrate you (or coworkers)
    • Manual processes that should be automated
    • Whether incumbents are actually old/clunky—or just disliked

Why it’s powerful (speaker’s reasons)

  • Deep context from lived experience
  • Access to customers/beta testers via coworkers and industry contacts
  • Clear understanding of the buying process (who signs off, budget cycles, objections)
  • Ability to validate while employed (lower risk)

Risks / watch-outs (actionable constraints)

  • Confirm the problem exists beyond your employer (avoid a “uniquely dysfunctional” bubble)
  • Check IP clauses in employment agreements
  • Don’t use proprietary/customer data from your employer
  • Use personal equipment and keep boundaries clean

Concrete examples

  • Jitbit (Alex Yumishev): help desk tooling missing “boring” enterprise features (SSO, AD integration, provisioning, on-prem Windows deployment)
  • Upstream Edge (Zach): forecasting/valuation tools driven by 10-year oil & gas frustration; 100x speed + higher-level features
  • Didomite (Lee Dido): desktop/production workflow problems repeatedly seen in a hardware-centric market; UX gaps drove a decade-later pivot
  • Blue Gamma (Ahmed Babikir): consultancy repeatedly producing interest rate data spreadsheets; productized the repetitive workflow
  • Loody (Steve): canvas retro tool idea from watching a PM write notes by hand

2) Copy/enter an existing category with proof of demand (~10%)

What the method really means

  • Not blind copying—enter an existing category where customers already understand the “type of tool”

Why it works (speaker’s mechanism)

  • Proof of demand: customers already pay/validate
  • Learn from incumbent failures (negative reviews/support forums)
  • Differentiate via:
    • Price
    • UX
    • Features
    • Niche
  • If the market is big, a small slice can be viable for bootstrapped founders

Risks / watch-outs

  • Small-market copying can mean “fighting for scraps” (needs enough market size)
  • Real differentiation is required (not just “we’re newer”)
  • Incumbents often win on brand/SEO/integrations/trust
  • Execution alone isn’t a strategy—you must know why you’ll win

Decision questions (practical prompts)

  • Why are customers unhappy (price, features, support, usability, other)?
  • Does your differentiation matter to customers?
  • Can you reach customers via channels incumbents underuse?
  • Is there an underserved segment that wants a purpose-built product?

Concrete examples

  • Drip (Rob Walling example): entered marketing automation/email automation after noticing competitors were expensive/contract-heavy with poor software UX; positioning leveraged category understanding
  • t.ly (Tim Leland): cheaper URL shortener + QR tooling after Bitly custom-domain costs became prohibitive
  • BigMailer (Lilia Tobin): email marketing platform for agencies/franchises; built in-house after costs were too high for their newsletter size
  • Checkout Joy (Merijn de Wet): custom checkout pages for e-learning using local payment processors + LMS connections; validated by quick conversion improvements
  • Hururu.ai (Simon Thompson): AI answering service tailored to Australian business context (accents/sensibilities), assuming US majors wouldn’t move quickly into AU

3) Discover problems through freelancing (~10%)

Mechanics

  • Repeated client pain reveals demand beyond one company
  • Clients pay you for research/customer development
  • Domain expertise builds faster

Risks / watch-outs

  • Freelancing is time-demanding; hard to build a product alongside
  • Need clear IP ownership before productizing
  • Transition to a “product owner” mindset is hard (learning to say no to features)
  • Avoid building overly bespoke solutions for a single client (no broader market signal)

Examples

  • Drum (Ben Walker): repeated internal-like ERP functionality for consulting clients → productized
  • Civic Review (John Reynolds): repeated requests for permitting software led to selling a tailored SaaS instead of rewriting custom solutions
  • StatusGator (Colin Bartlett): annoyance with not knowing whether an API status page existed; centralized ad API health + notifications
  • Dump Truck Dispatcher (Joe Walling): kept IP in exchange for charging clients less for the software

Action questions

  • What functionality have you built 3+ times?
  • What internal tools have you built for efficiency that others might want?
  • Can you shift from service revenue to product revenue—or will you be stuck forever as both?
  • Are your clients’ industries ones you want to serve long-term?

4) Solve a problem for spouse/friend/colleague (~8%)

Why it works

  • Direct access to a real user (fast interviews/feedback)
  • Less initial cold-calling
  • The relationship drives follow-through
  • Intros to others in the field enable early validation/sales

Risks / watch-outs

  • The “unique” problem may not generalize to the market
  • They might not provide honest feedback
  • Mixing personal/business relationships needs early expectation-setting
  • You still need to understand the market beyond their expertise

Examples

  • Quill Therapy Solutions (John Sister): wife (therapist) suggested using AI for progress notes/documentation without recording sessions; privacy/ethics constraint
  • CrankWheel (Yoav Sigurðsson): co-founder sales background + Yoav’s exposure to real-time communications; asked what salespeople need to show prospects—found “no tools” due to complexity

Action prompts

  • What does your partner complain about in daily work?
  • Can they introduce you to 5+ similar problem owners?
  • Is this a real business opportunity, not just a personal fix?
  • Do you have patience to learn the domain you don’t yet know?

5) Find problems online (only ~3%)

Core tactics

  • Lurk in:
    • Support forums
    • Reddit
    • Slack groups
    • Facebook groups
  • Look for repeated complaints, unmet needs, and gaps in existing solutions

Why it works (when it does)

  • Find problems in industries you haven’t worked in
  • Validate via:
    • Search volume
    • Complaint frequency
    • Community size
  • Negative reviews can reveal differentiation opportunities

Risks

  • Low signal-to-noise (complaints may not mean willingness to pay)
  • Lack of insider context can lead to misunderstanding
  • Requires time/patience
  • Selling to strangers in unfamiliar domains is hard

Examples

  • String Thing (Tolu Akinola): Shopify app for selecting delivery date; mined Shopify support forums; prioritized issues where apps had many reviews and many negative ones
  • Hovercode (Rami Koufach): dynamic QR generator; combined hypothesis with keyword research (Ahrefs); demand was present

Where to search (tools/places)

  • Shopify/App store reviews
  • G2/Capterra
  • Support forums
  • Keyword tools: Ahrefs, Semrush, Google Keyword Planner

6) Scratch your own itch (16%)

Mechanics

  • Build for your own problem, then verify others share it
  • Don’t treat “I have this issue” as validation (it proves one $0 customer)

Validation approach described

  • Example: Drip originated from scratching an itch, then the founders polled 17 founders and moved forward after getting 10–11 “yeses.”

Risks / watch-outs

  • “Market of one”
  • Building what you want instead of what the market wants
  • Your case may be too niche/edge-case
  • Founders may assume their needs match typical customers

Examples

  • Dialog Shift (Olga Khusar): hotel guest frustration → WhatsApp-like instant answers for hotel logistics/services
  • Simple Booklet (Scott Brownlee): accessible brochures via QR/tablet after years sailing without up-to-date information

Action prompts

  • How many people have the exact problem?
  • Are you representative—or an edge case?
  • Would others pay (not just be mildly annoyed)?
  • Can you separate personal preferences from market needs?

7) Take advantage of emerging/fast-growing technology (~3%)

Idea pattern

  • Use new tech openings (new APIs, platforms, AI, browser capabilities) before the space crowds

Why it works (speaker’s logic)

  • Early movers can establish presence before crowding
  • New tech creates new problems requiring new solutions
  • Ride the attention wave
  • Customers may tolerate rough edges early

Risks / watch-outs

  • Timing: too early (immature tech/customers uneducated) vs. too late (crowded)
  • Hype cycle ≠ willingness-to-pay
  • Tech may not mature as expected
  • Big competitors can move faster and crush smaller teams

Examples

  • WA Notifier (Rom Shingale): WhatsApp marketing momentum + WhatsApp API partner opening → WhatsApp marketing tool (especially for WhatsApp-heavy regions)
  • Gen Text AI (Alex Charles): connecting AI (post-ChatGPT) to Microsoft Word to reduce hallucinations
  • PodSqueeze (Tiago): AI-assisted podcast repurposing (social clips, summaries, transcripts, etc.) after frustration with post-production

Action prompts

  • Is the tech ready for real use?
  • Are businesses already spending in adjacent areas, or is it speculative?
  • What unique angle do you have (beyond hype)?
  • Can you build something useful quickly before the window closes?

8) Build on an existing product (4%)

Mechanics

  • Expand from what you already know / already sell
  • You’re already talking to customers, have momentum, and sometimes revenue to fund exploration

Why it works

  • You’re already in the market (close customer contact)
  • You see problems in real time
  • You can turn “tool for yourself” into broader demand
  • You leverage accumulated learning

Risks / watch-outs

  • “Shiny object” chasing
  • Splitting focus can kill both products
  • Escaping hard problems rather than pursuing real opportunity
  • Sunk cost can distort judgment

Action prompts

  • Is it genuinely bigger/better, or just boredom with the current product?
  • Can you validate quickly without fully abandoning the current product?
  • Do you have resources to run both—or must you choose?
  • What must be true to justify switching?

Examples

  • Authored Up (Ivana): LinkedIn analysis/content creation Chrome extension after recruiter/hiring tooling didn’t find customers (customers didn’t know what they wanted/needed)
  • Super Data (Refael): YouTube transcript scraping became a broader API tool; monetized via RapidAPI
  • SignWell (Ruben Gomez): dissatisfaction with e-signature options while running another business; improved offerings with security/compliance at an affordable price

Frameworks / playbooks explicitly or implicitly used

  • “Don’t tell me your idea—tell me what problem it solves.” (problem-first positioning)
  • “Enter an existing category” strategy (category understanding already exists; differentiation happens via UX/features/price/niche)
  • Validate that others will pay (especially for itch-scratching)
  • Practical validation-by-talk:
    • Interview / feedback loops
    • Polling other founders/users (example: 17 founders → 10–11 affirmations)
  • Read incumbent mistakes using negative reviews/support forums
  • “Market of one” test: are you edge-case or representative?

Metrics / KPIs / targets mentioned (as research signals)

  • Survey sizing and thresholds:
    • 200+ SaaS founders
    • Founders surveyed: at least ~$1/month, many >$100k/month
  • Method shares (used like “founder distribution” KPIs):
    • 48%, ~10%, ~10%, ~8%, ~3%, 16%, ~3%, 4%
  • Conversion-style proof example:
    • MVP that produced an “immediate bump in conversions” (Checkout Joy)
  • Early validation sample size:
    • Polling 17 founders → 10–11 yeses (Drip example)
  • Scale metrics in examples:
    • BigMailer founder grew email list to 300,000+ subscribers
  • No explicit CAC/LTV/churn targets provided in the subtitles

Additional resources / actionable recommendations

  • Free resource mentioned:
    • robwalling.com/ideas (free 9-to-5 audit; prompts for finding ideas “hiding at your work day”)
  • Marketplace add-on reference:
    • microconf.com/marketplaces (list of 75+ marketplaces)

Presenters / sources

  • Presenter: Rob Walling
  • Survey/example founders cited:
    • Alex Yumishev (Jitbit)
    • Zach (Upstream Edge)
    • Lee Dido (Didomite)
    • Ahmed Babikir (Blue Gamma)
    • Steve (Loody; formerly Metro Retro)
    • Tim Leland (t.ly)
    • Lilia Tobin (BigMailer)
    • Merijn de Wet (Checkout Joy)
    • Simon Thompson (Hururu.ai)
    • Ben Walker (Drum)
    • John Reynolds (Civic Review)
    • Colin Bartlett (StatusGator)
    • Joe Walling (Dump Truck Dispatcher)
    • John Sister (Quill Therapy Solutions)
    • Yoav Sigurðsson (CrankWheel)
    • Tolu Akinola (String Thing)
    • Rami Koufach (Hovercode)
    • Steven Gladney (OrderNerd)
    • Olga Khusar (Dialog Shift)
    • Scott Brownlee (Simple Booklet)
    • Rom Shingale (WA Notifier)
    • Alex Charles (Gen Text AI)
    • Tiago (PodSqueeze)
    • Ivana (Authored Up)
    • Refael (Super Data)
    • Ruben Gomez (SignWell)

Original video