Product Hunt launches are one of those things you can prepare for forever without ever feeling ready. We had been building SalesSheet for months — an AI-native CRM with built-in calling, email sync, conversational AI, and a mobile experience we were proud of. But there was always one more feature to add, one more edge case to handle, one more screen to polish. On Friday, February 14th, we made the call: we are launching on Product Hunt on Monday. That gave us 48 hours.
The decision was partly strategic, partly gut instinct. We had been watching Product Hunt's CRM and sales tool category and noticed a window with low competition. Monday morning Pacific Time is when Product Hunt resets its daily leaderboard, and the early hours are critical for building momentum. If we waited another week to feel more prepared, we might miss the window and launch into a more crowded day. So we committed. Friday night became the starting line.
Looking back, the 48-hour constraint was the best thing that could have happened. It eliminated scope creep, forced ruthless prioritization, and created an intensity that produced our best work. The sprint yielded 63 commits, 19 features shipped, 44 bugs fixed, and 315 files changed. Here is exactly how it happened.
Before writing a single line of code on Saturday morning, we spent Friday evening building the launch checklist. Product Hunt requires specific assets and preparation that you cannot rush on launch morning. We broke the checklist into three categories: Product Hunt assets, landing page, and product readiness.
For Product Hunt assets, we needed: a tagline under 60 characters ("AI-native CRM that replaces forms with conversation"), a description paragraph, a gallery of 5 screenshot images showing the AI chat, pipeline view, calling interface, mobile app, and analytics dashboard, an animated GIF demonstrating the AI assistant creating a contact from natural language, and a maker comment explaining why we built SalesSheet and what makes it different. We prepared all of these Friday night so launch morning would be pure execution.
The landing page needed work. Our existing homepage was functional but not optimized for the Product Hunt audience — technical founders and early adopters who evaluate products in under 30 seconds. We rewrote the hero section to lead with the AI conversation interface rather than a traditional feature list. We added a "Built with" section showing our tech stack (Remix, Supabase, Telnyx, OpenAI) because the PH audience respects transparency about technical choices. We added social proof placeholders that we could fill in post-launch with actual user testimonials and upvote counts.
The feature list was ambitious but focused. Every item was something users would notice in their first five minutes with the product:
Two decisions defined the sprint and could have derailed it entirely. The first was migrating telephony providers mid-sprint. We had a working Twilio integration — functional calling, basic recording, reasonable quality. But Telnyx offered better per-minute pricing (critical for a bootstrapped CRM competing on price), a cleaner WebRTC SDK that reduced our client-side calling code, and an API that made features like hold and speaker boost trivial to implement. The migration touched 38 files across the codebase. If anything had gone wrong — a misunderstood API parameter, a WebRTC compatibility issue, a billing configuration error — we would have lost hours we did not have. We tested the migration path on Friday night with a single test call, confirmed it worked, and committed to the full migration Saturday morning.
The second hard decision was adding Voice DNA at the last minute. Voice DNA was not on the original 17-feature sprint list. It was a "nice to have" that we promoted to "must have" on Saturday afternoon when we realized it would be our strongest Product Hunt differentiator. Every CRM claims to have AI. Voice DNA makes our AI personal — it writes emails that sound like you, not a generic assistant. Building it meant creating a Gmail analysis pipeline, a Slack message analyzer, a style profile generator, and integrating the resulting profile into every AI drafting context. We shipped it in under 12 hours. It became the most-discussed feature in our Product Hunt comments.
The constraint of 48 hours forced us to ask the right question: not "is this feature ready?" but "will someone on Product Hunt notice this in their first 5 minutes?" If yes, it shipped. If no, it waited.
Pricing for a Product Hunt launch is a strategic decision that affects your first impression with thousands of potential users. We considered three approaches: generous free tier with expensive paid plans, low-cost entry with per-seat pricing, or flat-rate pricing that rewards small teams. We chose the third option. SalesSheet's pricing is per-account, not per-seat — $29/month for Pro, $79/month for Ultra. A team of five pays the same as a solo founder. This pricing model resonates with the Product Hunt audience, which skews toward small teams and solo founders who are tired of per-seat CRM pricing that punishes growth.
We also offered a launch-day discount — an extended trial period for Product Hunt users who signed up on launch day. This created urgency without devaluing the product with a permanent discount code. The free tier includes core CRM features with AI chat, and the upgrade path is clear: more AI capacity, Voice DNA, calling, and advanced analytics on paid plans.
Claude Code was the force multiplier. As a solo founder, shipping 19 features in 48 hours without AI-assisted development would have been physically impossible. Claude Code handled the boilerplate — generating Supabase Edge Functions, writing Zod validation schemas, scaffolding React components, and creating test files. The human-AI workflow was tight iteration loops: describe what you want, review the generated code, adjust, commit, move on. The 63 commits tell the story of that rhythm — short, focused changes rather than massive pull requests.
The tech stack itself was a force multiplier too. Remix's server-side rendering meant fast initial page loads without a separate API layer. Supabase's realtime subscriptions meant live data updates without building WebSocket infrastructure. Telnyx's WebRTC SDK meant browser-based calling without a native app. Vercel's edge deployment meant instant deploys during the sprint — push to main, live in 30 seconds. Every tool in the stack was chosen for developer velocity, and that choice paid off when velocity was the only thing that mattered.
Product Hunt's daily leaderboard resets at midnight Pacific Time. We scheduled the post for 12:01 AM PT on Monday. The maker comment was pre-written and ready to paste. The first three hours after launch are the most critical — upvotes during this window carry more weight in the ranking algorithm than upvotes later in the day. We pinged our personal networks at 6 AM Pacific when the East Coast was waking up and again at 9 AM Pacific when the West Coast came online.
We monitored three metrics throughout the day: upvote count and rank on the daily leaderboard, traffic to app.salessheets.ai (tracked via PostHog), and sign-up conversions. The landing page conversion rate — visitors who clicked "Start Free" and completed registration — was the number we cared about most. Product Hunt upvotes feel good, but sign-ups are the actual outcome. We had PostHog dashboards open on a second monitor all day, watching the funnel in real time.
Engagement in the Product Hunt comments section mattered too. Every question got a detailed, technical answer from the maker (me). Product Hunt's audience respects founders who show up, explain their technical decisions, and engage honestly about tradeoffs. We answered questions about our AI architecture, our pricing rationale, our security posture, and our roadmap. Every comment was an opportunity to demonstrate that SalesSheet was built by someone who understands both the technical and business sides of CRM.
Five lessons from our Product Hunt launch sprint that apply to any product launch:
17 planned features. 19 shipped. 44 bugs squashed. Zero compromises. The CRM that launched on Product Hunt was complete, polished, and ready for real users. Every feature worked. Every screen was responsive. Every edge case was handled.
We did not cut a single feature from the launch list. If anything, we added two more because the momentum was unstoppable. That is what happens when a deadline is real and the product matters.
That is the story of our Product Hunt launch sprint. Not a story of compromise, but of relentless execution. 63 commits in 48 hours, and a product we are proud to show the world. If you are considering a Product Hunt launch, the best advice we can give is this: set a date, build the checklist, and sprint. You will ship more in 48 focused hours than in two comfortable months.