The Problem I've Had Since 2017

I'm a cyclist. I've been riding with a GoPro mounted on my handlebars for years — originally just in case of an accident, but I quickly realized the footage was more interesting than I expected. There was just one problem: the video only tells half the story.

On one side, you have the camera footage — landscapes, descents, the road ahead. On the other, your cycling computer is recording everything: speed, altitude, heart rate, cadence, power. Two data sources, completely disconnected.

Back in 2017, Garmin had a tool called VIRB Edit that could overlay telemetry data onto video. It was exactly what I needed. Then Garmin discontinued it. The workflow was also painful — processing 4K files frame-by-frame on a desktop app, and if something went wrong, you'd start over from scratch.

Today the only real alternative is Telemetry Overlay, a desktop application at $183 one-time. It works, but it's heavy, desktop-only, and aimed at power users. Not at the weekend cyclist who just wants a cool short for Instagram.

The Idea

What if you could do the same thing — overlay GPS telemetry onto your cycling video — but in the cloud? No desktop software, no heavy file processing on your machine. Upload your video, connect your Strava or Garmin data, sync them up, and get a vertical short with speed, altitude, heart rate overlaid — ready for Instagram, TikTok, YouTube Shorts.

Not just for cycling either. Anyone who carries a GPS device and an action camera: skiing, motorcycling, karting, surfing, skateboarding. The speed is the common denominator.

Before writing any code, I wanted to validate whether this idea made sense. So I sat down with Claude.

Using AI as a Co-Founder for Validation

I opened a fresh Claude chat and started from zero — no leading questions, no bias. I wanted Claude to act as a VC, a potential user, and a product manager simultaneously.

Step 1: Feed It Context

Claude can't browse the web, but it can read documents. I downloaded Telemetry Overlay's user manual and uploaded it. Within seconds, Claude understood exactly what the product does and how it works.

Step 2: SWOT Analysis of the Competition

I asked Claude for a SWOT analysis of Telemetry Overlay as a desktop application. The results were revealing:

Strengths — wide format compatibility, deep customization, one-time payment, works offline.

Weaknesses — high technical requirements (processing 4K video is brutal), steep learning curve, limited to one device, no mobile workflow, manual updates.

Opportunities — growing creator economy, more GPS-enabled devices, niche market without dominant players, potential for collaboration features.

Threats — cloud-based competitors (exactly what I'm proposing), evolving video editors, the cost of maintaining desktop software across three operating systems.

The SWOT confirmed what I felt: there's a gap for a cloud-native, social-first, mobile-friendly version of telemetry video overlays.

Step 3: Market Sizing

Claude estimated a Total Addressable Market of around $1B, with a Serviceable Addressable Market of 100-200M users — people who use both a GPS device and an action camera. The realistic target: capturing 0.1-1% of that market in the first years, meaning 150K-2M users.

For a side project, the relevant question isn't the TAM — it's whether I can find 50-200 people willing to pay.

Step 4: The PRD

I asked Claude to generate a Product Requirement Document. In 10 minutes, I had something that would have taken me weeks to write:

Executive summary and positioning
User personas and use cases
MVP feature scope (profile, activity sync, basic widgets, sharing)
Subscription tiers (Free with limits, Premium at ~$10/month)
Technical architecture recommendations (serverless, mobile-first)
Success metrics and first-year goals
Risk analysis with mitigations

The PRD wasn't perfect — Claude estimated 3-4 months of development when I thought I could ship an MVP in 2 weeks using AI tools. But the structure and thinking were solid.

Step 5: Landing Page and Video Brief

In the last few minutes of the session, I asked Claude for two things: a prompt for Lovable (an AI UI builder) to create a landing page, and a creative brief for a 30-second demo video with the concept: "Your video only tells half the story."

What I Learned

Claude as a validation partner is genuinely useful. In 90 minutes, I went from a scattered idea in my head to a structured business case with competitive analysis, market sizing, a PRD, and an MVP scope. I couldn't have produced this in two months on my own.

The idea validated. Three things confirmed it: (1) people are already paying $183 for Telemetry Overlay, proving demand exists, (2) the desktop-to-cloud transition is a clear market gap, and (3) the social/sharing angle opens the market beyond power users to anyone who rides with a GoPro.

But validation means nothing without execution. As I said at the end of the video: everything so far is worth absolutely nothing. You have to build it, and then you have to get it in front of people. That's the hard part.

What's Next

Next session: choosing a name, building the landing page with Lovable, and starting to think about marketing. The build-in-public journey continues.

This is part 1 of the "Building SportsSync" developer journal — a series documenting the journey of building a SaaS product from idea to launch, using AI tools at every step. Follow the full series.

How I Used AI to Validate a SaaS Idea in One Session