PasteWaves lets people upload and share audio files instantly. Most users were anonymous—they’d upload, share, and disappear. I needed sign-ups for retention and eventual subscriptions.

So I added a banner: “Please sign in before uploading.”

It didn’t work. Users would land with one clear intent—share this audio file right now—and I was asking them to create an account first.

I was breaking their flow at the worst possible moment.

The Problem with “Login First”

Friction kills conversions.

When someone lands on PasteWaves, they have a job to do. They don’t want to create an account—they want to upload a file.

By forcing authentication upfront, I was solving my problem at the expense of theirs. Users responded predictably: upload anonymously and leave.

The Hypothesis: Upload First, Claim Later

What if users could upload anonymously, get their link immediately, and then be encouraged to sign in?

The flow:

1. User uploads files (no account required)
2. Files upload successfully (they get their links)
3. Banner appears with value proposition
4. User signs in (they’re invested now)
5. Files automatically claimed (seamless)

Deliver value FIRST, ask for commitment LATER.

The Technical Implementation

Anonymous Upload:

1. Backend generates cryptographic claim token (ct_8f3a2e1b...)
2. Storage record created with owner="anonymous", claim token, 7-day TTL
3. Browser caches token in localStorage
4. User gets shareable link immediately

Automatic Claiming: When the user signs in via Google OAuth:

1. Frontend reads claim tokens from localStorage
2. Sends claim requests to backend
3. Backend validates token, updates storage atomically:
   • owner → user’s ID
   • Retention upgrades (7d → 14d for Free, or permanent for Pro)
   • Claim token removed
4. Files appear in user’s account

From the user’s perspective: seamless. Sign in, files appear.

The Psychology

The hypothesis relies on behavioral psychology:

Sunk Cost: Once someone uploads files, they’ve invested effort. The banner reminds them: “Sign in to keep your file for 14 days instead of 7.” They don’t want to lose what they’ve created.

Delayed Commitment: Old approach asked for commitment before value. New approach delivers value before commitment.

Tiered Incentives: Natural upgrade path emerges:

• Anonymous: 7-day retention
• Free (authenticated): 14-day retention, file manager
• Pro: Permanent storage (25GB), advanced tools

If the hypothesis holds, users will see the value difference and convert.

Key Principles

1. Never break the user’s flow at the critical moment. Let them do the job first, ask for commitment later.

2. Deliver value before asking for commitment. Let users experience the product, then invite them to stay.

3. Serverless enables sophisticated UX. Public endpoints, authenticated APIs, TTL-based cleanup, and atomic updates make this pattern possible without infrastructure complexity.

4. Measure everything. Track the funnel, watch conversions, iterate on data.

5. Psychology matters as much as code. Understand why people hesitate and how to reduce friction.

The Experiment

I can’t claim victory yet—this is an experiment in progress. The hypothesis: removing friction will increase anonymous uploads, better post-upload incentives will drive sign-ins, and claimed files will improve retention.

I’m collecting metrics and observing as the experiment unfolds.

If you’re struggling with user onboarding, consider this: what would happen if you let people use your product before asking them to sign up?

I’m about to find out.

The Car Ride Epiphany

One weekend, driving from Oslo to Gjøvik, I started thinking about it.

How much audio actually fits in 1TB?

Turns out: a lot. An absurd amount, really.

Here’s what made me stop and think:

• 1TB of compressed audio holds roughly 300,000 songs
• At today’s S3 prices, that’s about $23/month in storage costs
• Most people would never come close to filling it
• Yet every platform makes you pay per upload, per minute, per feature

Storage has quietly become almost trivial. This isn’t a hard problem anymore—it’s just that nobody’s bothered to give it away.

That realization changed everything.

Building It

So I started building a simple web app where you can drag in an audio file and get a link back.

The core idea was simple:

• No account required
• No setup friction
• No artificial limits on the free tier
• Just upload and share

I used Claude Code and put together the first version over a weekend. The entire stack runs serverless on AWS—React frontend, Go lambdas, DynamoDB for metadata, S3 for storage. Zero servers to maintain.

What It Became

After that first version worked, I started adding features as I needed them:

• Google OAuth login for permanent storage (25GB for Pro users)
• Subscription tiers with Stripe integration
• Album artwork extraction and display
• Browser-based audio editor with trim, effects, and fades
• Direct recording so you can capture and share from the same place
• Comments system for collaboration

But the free tier stayed generous. Anonymous users get 7 days of storage from a shared 1TB pool. Authenticated free users get 14 days. No credit card, no bullshit.

It started as a small frustration and turned into something I use all the time. If you’ve ever just wanted to send someone a piece of audio without turning it into a whole project—you’ll probably get it.

What’s Next

Right now I’m focused on making the upload experience smoother, improving the player UI, and adding password-protected files for sensitive content.

The goal isn’t to build the next SoundCloud. It’s to stay simple, fast, and generous with storage. Because in 2026, that shouldn’t be a hard ask.

If you’re a musician, producer, podcaster, or just someone who shares audio—give it a try. I built this for people like us.