I Built 2 SaaS Apps to $50K MRR: Here's My 90-Day Playbook

SUMMARY

Roman Czerny, a former mechanical engineer turned SaaS founder, shares his playbook for building two apps to $50K MRR, emphasizing customer validation, selling before coding, cold outreach, and multi-channel distribution in a 47-minute interview.

STATEMENTS

Roman Czerny transitioned from mechanical engineering to SaaS after freelancing, affiliate marketing, and selling a mobile game during COVID, seeking to compound knowledge in a stable business.
His first SaaS attempt failed because the team copied a successful U.S. text-to-buy app and adapted it to WhatsApp for Europe without validating customer demand, wasting six months.
European customers were not ready to pay via text or WhatsApp prompts, leading to quick pilot dropouts and highlighting the need for direct customer conversations.
Pivoting, Roman cloned a profitable WhatsApp Shopify app already succeeding in Europe, launching an MVP in two weeks using existing code from the failed project, Firebase, and basic features.
The cloned app gained immediate traction with 5-10 clients in France due to low competition, proving the value of copying validated, market-proven ideas.
Roman's role focused on marketing, while his team included two developers initially for Coco AI, now three members: one for marketing, one for product, and one for coding.
Growth to $500K ARR for Coco AI relied almost exclusively on cold outreach in the French market, targeting Shopify stores over $10K monthly revenue, especially those using AI tools.
Daily outreach involved 100 calls, 1,000-2,000 cold emails, and LinkedIn messages, aiming to secure demos with 60-70% close rates.
Leads were sourced from LinkedIn interactions, Sales Navigator, Store Leads, and Fox Leads, focusing on high-intent users engaging with marketing content.
A key hack was emailing customer service teams with demo videos to forward to CEOs, yielding near-100% replies and avoiding spam filters.
Coco AI was sold for seven figures after two years because the French market saturated, international expansion proved difficult with lower close rates, team members wanted out, and no clear vision existed.
The sale bypassed brokers like Acquire.com by directly outreach to Shopify app portfolio builders and spotting a LinkedIn post from a buyer, closing in two months for 90% upfront payment.
For GojiBerry, Roman sold a high-intent leads service first, generating $7K MRR manually via Fiverr hires before automating, validating demand without coding.
Previous GojiBerry versions failed: V1 (AI UGC) reached $500 MRR but lacked evolution path; V2 (AI note-taker like Fireflies) was hard to sell after two-week MVP.
Distribution for GojiBerry combined outreach with inbound: Reddit posts garnered 11 million impressions and tens of thousands of visits by sharing value-driven stories with proof.
Roman rejects single-channel focus early on, instead posting heavily on one, optimizing, then layering new channels like X, YouTube, SEO, and Threads to dilute time per channel.
Reddit strategy involved three weekly posts in 10 SaaS subreddits, telling stories (e.g., YC rejection, influencer campaigns) backed by proof and including links, boosted by a friend group for initial upvotes.
Outreach remains primary because it provides immediate feedback; tools like Goji AI (internal) for LinkedIn and Instantly AI for emails send 5,500 messages daily across five accounts.
Instead of direct demo requests, outreach offers free blueprints (10-page value PDFs) ending with optional booking links, leading many to self-signup via free trials.
Newsletter sponsorships in French yielded profitable but modest results over 2-3 months, harder to track long-term; LinkedIn influencers proved more ROI due to lower prices and viral potential.
Affiliates contribute 10-15% revenue with 30% lifetime commissions (planning to raise to 60%), providing ready-made promotion kits to overcome promoter inertia.
TrustMRR ads, including a top spot, drove 50-200 daily visitors profitably, amplified by X visibility and retweets from founder Marlu, leveraging past affiliate revenue display for credibility.
Roman's 90-day playbook starts with self-assessment of skills, market research on competitors, creating a 7-slide PowerPoint pitch, validating via 100-500 calls, selling as service first.
If validated with 5-10 paying customers, build MVP via co-founder or hire, then scale outreach with $300-500 tools; layer inbound on Reddit, LinkedIn, X for feedback and trials.
Best ideas stem from personal or observed problems turned into SaaS, always starting with customer willingness to pay, copying only after verifying competitors' real advantages.
Automate outreach by scaling proven campaigns weekly with new leads, minimizing changes until response rates drop, freeing time for responses.
GojiBerry's 2026 vision is a lean GTM copilot AI that identifies optimal contacts, timing, and messaging for sales, differentiating from basic outreach tools, targeting 500+ customers.
Major mistakes included building unvalidated products and wasting on fake influencers; needle-movers are relentlessly scaling what works amid global competition requiring high energy.

IDEAS

Cloning market-proven apps in underserved regions like Europe can shortcut validation, but only if you adapt minimally and leverage existing codebases for quick launches.
Targeting customer service emails instead of founders boosts reply rates to nearly 100% because they always respond, even if just to forward, turning potential gatekeepers into allies.
Selling a service manually before coding confirms demand with real revenue, allowing bootstrapped founders to hire help like Fiverr workers for fulfillment until automation scales.
Failed MVPs like AI UGC or note-takers reveal that easy-to-build ideas often lack sales traction, emphasizing the need to test market fit before heavy development investment.
Multi-channel distribution starts with one platform for mastery, then stacks others to avoid over-reliance, reducing burnout while compounding traffic sources over time.
Reddit virality hacks like friend-group upvotes secure initial 10 upvotes, propelling story-based posts with proof and links to millions of impressions without ad spend.
Blueprints as outreach bait lower commitment barriers compared to demos, delivering 10 pages of value that educate leads and subtly drive self-signups via embedded trials.
LinkedIn influencers outperform newsletters in ROI because lower competition allows price negotiation for more posts, plus viral mechanics can multiply exposure exponentially.
High affiliate commissions (up to 60% lifetime) incentivize sustained promotion, but success hinges on providing plug-and-play kits to eliminate promoters' execution hurdles.
Displaying aggregated past revenue on profiles like TrustMRR builds instant credibility, even from defunct ventures, redirecting curiosity to current products.
Post-exit reflection shows recreating success demands immense energy; low-churn, happy-customer products are rare and not easily replicated without deep domain knowledge.
Outreach's edge over inbound lies in quantifiable feedback loops—if 5,000 emails yield no responses, pivot fast—enabling rapid iteration absent in slower channels like YouTube.
PowerPoint pitches for validation keep ideas lean, adapting based on 100-500 customer calls to align solutions with actual pains, bridging non-technical founders to development.
Being the core user of your product, like Roman with outreach tools, ensures precise feature evolution, contrasting blind builds that lead to visionless saturation.
Global competition in SaaS intensifies, pitting bootstrappers against funded teams worldwide, underscoring that relentless volume in proven tactics separates winners.
YC acceptance signals hype but not success—98% fail—teaching that copying buzzworthy ideas without customer talks wastes months on unviable adaptations.
French market saturation forces international pivots, but cultural/language barriers slash close rates from 60% to 15-20%, highlighting bootstrapped scaling limits.
Exits without brokers via targeted outreach to acquirers yield better multiples than platforms demanding one-year revenue history, often undervaluing young SaaS.
GTM copilots evolve from automation to AI advisors on who, when, and how to engage, prioritizing high-response prospects for outsized sales efficiency.

INSIGHTS

True SaaS success begins with customer obsession, measuring problem urgency and payment willingness before any build, turning guesswork into validated revenue streams.
Copying works when paired with regional tweaks and data tools like Store Leads, but blind replication ignores cultural mismatches, as seen in Europe's SMS aversion.
Cold outreach's power stems from its feedback velocity and scalability, allowing solo founders to mimic team output with minimal tools, outpacing inbound's slow burn.
Service-first validation de-risks ideas by generating cash flow manually, revealing product-market fit through fulfillment challenges before tech commitments drain resources.
Layered distribution prevents single-point failures, evolving from outreach dominance to inbound amplification, sustaining growth as channels mature and overlap.
Storytelling on free platforms like Reddit democratizes traffic, where value-packed narratives with proof convert curiosity into trials, bypassing paid acquisition barriers.
Team dynamics matter: Marketing-led founders thrive by outsourcing dev, but vision gaps or member exits signal when to sell rather than force expansion.
High-intent targeting via engagement signals refines leads, boosting responsiveness and close rates, as engaged prospects already signal problem awareness.
Automation follows proof: Scale unchanging campaigns with fresh leads weekly, automating drudgery to focus on high-leverage responses and iteration.
Affiliates flourish with generous, lifetime incentives and done-for-you assets, transforming passive signups into active promoters who compound long-term traffic.
Post-sale hindsight underscores rarity of defensible moats; low-churn traction demands ongoing energy, making repeat successes harder without personal domain mastery.
Global bootstrapping demands adaptability—local dominance saturates fast, but international hurdles like lower conversions require strategic exits over forced scaling.

QUOTES

"Everything starts with the customer, not with the product. How much do they want to solve this problem and how much are they willing to pay?"
"We spent 6 months for nothing just because we did not talk to customers and if we had talked to customers in one week we would have known that we should not launch that."
"The game was just getting demos and then it was almost always a win."
"Having money before you code is a great great sign. It means that really you you are into something."
"I'm not a big fan of that especially in the beginning. So what I do is um my process is posting a lot on one channel, understanding how it works and then adding more and more channels every day."
"Outreach is definitely the number one. Um you need one LinkedIn account, you need a few email inboxes and that's it. You can you can know very fast if uh your project has potential or not."
"Saying yes to a blueprint is way easier than saying yes for a demo."
"The more you give to affiliates the more they promote um and I've seen it with some friends doing uh affiliate marketing Okay, when you increase the affiliate commission from 30 to 40, well, you make more than that."
"It's getting harder and harder. You have people that are highly motivated from all around the world. So you are competing against a lot of people."
"We want to create one of the best tools out there in terms of GTM copilot."

HABITS

Conduct daily cold outreach quotas, such as 100 calls and 1,000-2,000 emails, to maintain momentum and test messaging efficacy.
Post value-driven stories three times weekly on Reddit in targeted subreddits, backing them with proof and links to drive organic traffic.
Review and adapt outreach campaigns weekly by adding new leads without altering copy, ensuring sustained performance until rates decline.
Target high-intent leads by monitoring LinkedIn interactions on relevant posts, prioritizing engaged users for higher response potential.
Provide affiliates with ready-made promotion kits including videos and text templates upon signup to accelerate their efforts.
Stack distribution channels progressively, mastering one before layering another, to balance effort across inbound and outbound sources.
Solicit feedback via 100-500 customer calls early, refining pitch decks based on real pains to align offerings pre-build.
Display aggregated revenue histories on profiles to leverage past credibility for current product visibility without active maintenance.
Focus relentlessly on scaling proven tactics, like doubling message volume upon positive demo rates, to compound results amid competition.

FACTS

98% of YC companies fail, despite the hype around acceptance signaling success.
Coco AI reached $500K ARR primarily through French market outreach, with international close rates dropping to 15-20% from 60%.
GojiBerry generated $7K MRR from manual high-intent lead services before any coding, using Fiverr for fulfillment.
Reddit posts for GojiBerry achieved 11 million impressions and tens of thousands of website visits in four months without ad spend.
TrustMRR ads for GojiBerry brought 50-200 daily visitors, with one retweet yielding ROI exceeding costs.
Roman and his co-founder each earned about $2.5 million from past affiliate and course ventures, totaling $4.8 million displayed for credibility.
GojiBerry V3 launched six months after Coco AI's February sale, reaching $33K MRR in four months via distribution.
Cold outreach close rates hit 60-70% on demos for Shopify WhatsApp tools in Europe due to low competition.
Affiliates currently drive 10-15% of GojiBerry's revenue, with plans to increase commissions from 30% to 60% lifetime.

REFERENCES

One Text: U.S. SMS text-to-buy app in YC that inspired failed European clone.
Store Leads: Chrome extension for analyzing Shopify app downloads and store data.
Firebase: Backend tool used in building WhatsApp Shopify apps.
Sales Navigator: LinkedIn tool for targeted lead sourcing.
Fox Leads: Tool for extracting founder contacts from Shopify stores.
Instantly AI: Cold email automation platform for high-volume outreach.
Goji AI: Internal tool for LinkedIn message automation across five accounts.
Acquire.com: Broker platform rejected due to lack of one-year revenue.
Flipper: Another acquisition site that undervalued the business.
Fireflies AI: Note-taking app similar to failed GojiBerry V2.
Y Combinator: Accelerator program; rejection story shared on Reddit.
TrustMRR: Ad platform and revenue transparency site for SaaS visibility.
Justin W Newsletter: Example of targeted sponsorship for B2B tools.
X (formerly Twitter): Platform for posts, retweets, and affiliate promotion.
Threads: Emerging social channel recently added for traffic.

HOW TO APPLY

Assess your skills and unfair advantages, then brainstorm ideas solving personal or observed problems, researching competitors' funding and marketing paths to choose bootstrapping or raising.
Create a basic 7-slide PowerPoint pitch including logo, pain points, solution overview, product workflow, and pricing to communicate your offer clearly without full development.
Reach out to 100-500 potential customers via messages or calls, probing their problems and previewing your solution to gather feedback and refine the pitch iteratively.
Once pains align, secure pre-payments from 5-10 prospects by delivering as a manual service, refunding if needed, to validate demand with real revenue before coding.
If validated, recruit a developer or co-founder to build an MVP, focusing on core features that address confirmed pains, aiming for a two-week launch timeline.
Initiate cold outreach using affordable tools like LinkedIn automations ($300-500 setup), sending high volumes to book demos, tracking 60%+ close rates as success metrics.
Layer inbound efforts post-outreach stability by posting feedback-seeking content on Reddit, LinkedIn, and X, asking for reviews to build trials and organic traction.

ONE-SENTENCE TAKEAWAY

Validate SaaS ideas by selling services first through relentless outreach, cloning proven models, and stacking distribution for rapid, customer-driven growth.

RECOMMENDATIONS

Prioritize customer interviews over assumptions to avoid six-month failures, ensuring ideas match real pains and payment intent.
Clone successful apps in niche markets using data tools like Store Leads, launching MVPs in two weeks to test traction quickly.
Target customer service for outreach to guarantee replies and internal forwarding, boosting demo bookings without founder gatekeeping.
Sell manual services before building to generate early revenue, hiring freelancers for fulfillment until automation justifies development.
Use blueprints in emails to lower barriers, providing value that educates leads toward self-signups and free trials.
Stack distribution channels starting with outreach, adding Reddit stories and LinkedIn influencers for compounded, low-cost traffic.
Offer escalating affiliate commissions with promotion kits to turn signups into proactive revenue drivers over lifetimes.
Exit strategically without brokers by networking acquirers directly, securing better multiples than undervaluing platforms.
Automate proven outreach campaigns weekly with fresh leads, focusing founder time on responses and iteration.
Build as the core user to maintain vision, evolving features from deep domain knowledge rather than external guesses.
Scale relentlessly what works, like message volumes, to outpace global competitors demanding high daily energy.
Display transparent revenue histories on sites like TrustMRR to borrow past credibility for current launches.

MEMO

Roman Czerny, a mechanical engineer disillusioned with corporate boredom, stumbled into SaaS after freelancing and affiliate hustles during COVID. His first venture, a WhatsApp adaptation of a buzzy U.S. text-to-buy app, crashed after six months of development without a single customer conversation. European shoppers weren't primed for impulse SMS purchases, a lesson in validation's primacy. Undeterred, Czerny pivoted by scouring Shopify's app store for proven WhatsApp tools raking in millions. Cloning one with minimal tweaks—leveraging Firebase and prior code—he launched Coco AI in two weeks, securing 5-10 clients overnight in France's underserved market.

Growth exploded through brute-force cold outreach: 100 daily calls, thousands of emails, and LinkedIn pings targeting Shopify stores above $10K monthly revenue. Czerny's hack? Bypassing founders to email customer service with demo videos, netting near-perfect replies as teams forwarded to CEOs. Close rates soared to 70% on demos, pushing Coco AI to $500K ARR. Yet saturation hit; France's small pond limited expansion, with German and Spanish closes plummeting to 15%. Team fractures and a hazy vision prompted a seven-figure sale—brokered informally via LinkedIn outreach to app portfolio builders, closing in months for 90% upfront.

Post-exit, Czerny applied hindsight to GojiBerry, a high-intent leads tool. Discarding quick-build failures like AI UGC ($500 MRR dead-end) and note-takers (unsellable MVP), he sold the service first. Twenty thousand emails promising targeted leads yielded $7K MRR manually, fulfilled via Fiverr blueprints before automating. This de-risked the build, confirming demand from dental clinics to e-commerce brands seeking nurse hires or marketing responders. Now at $50K MRR in six months, GojiBerry thrives on layered distribution: Reddit stories of YC rejections and influencer playbooks amassed 11 million impressions, while TrustMRR ads and LinkedIn shoutouts added viral boosts.

Outreach remains the engine, with tools like Instantly AI firing 5,500 messages daily across channels. Blueprints—value-packed PDFs—outmaneuver demo asks, guiding leads to trials. Affiliates snag 10-15% revenue via 30% lifetime cuts and ready kits, with plans to hike to 60%. Newsletters underperform influencers, where low prices and repost virality multiply ROI. Czerny's ethos: Stack channels post-mastery, from Reddit upvotes via friend networks to X retweets, dodging single-thread risks in a hyper-competitive SaaS arena.

Looking ahead, GojiBerry eyes 2026 as a lean GTM copilot, AI advising on optimal contacts, timing, and copy for sales pros—evolving beyond automation. Czerny's mistakes, from fake influencers to unvalidated builds, underscore energy's toll; recreating Coco's low-churn magic demands founder-as-user clarity. In a world of global bootstrappers, he preaches relentless scaling of winners: If 10 messages book a demo, send 1,000. For aspiring B2B founders, his 90-day blueprint—skill audit, pitch refinement via calls, service sales, then outreach-inbound hybrid—offers a roadmap from zero to viable traction.