Podstack: Turning Newsletters Into Podcasts

The Challenge

In an age of information overload, newsletters are more valuable than ever — but also harder to keep up with. Busy professionals often subscribe to multiple newsletters across Substack, Medium, and independent publishers. Yet most of these remain unread, buried under daily email noise.

Our user research showed a simple truth: people want to read newsletters, but don’t have the time to sit down and go through them. Podcasts, on the other hand, fit into people’s routines — during commutes, workouts, or chores.

The Idea

Podstack bridges this gap. It fetches newsletters from your inbox, cleans and prioritizes the content, and converts them into short podcast-style audio episodes — personalized and delivered daily.

Our Approach

We focused on four core goals:

Seamless Integration: Connect to Gmail in one click.
Smart Filtering: Identify which emails are actually newsletters.
Content Optimization: Strip away noise and extract core message.
Natural Audio Output: Convert content into smooth, human-sounding narration.

Key Features

🔐 OAuth-based Gmail integration
🧠 AI-powered newsletter detection & parsing
🧹 HTML cleaner to strip out signatures, ads, and clutter
🎙️ Voiceover pipeline using [PlayHT / ElevenLabs] style TTS models
📱 Ready-to-consume podcast feed via RSS or embedded player

Tech Stack

Frontend: Next.js + Tailwind
Backend: FastAPI + Celery workers
Auth & Email API: Google OAuth2, Gmail API
Parsing & Summarization: Langchain + GPT-4-turbo
Text-to-Speech: Custom voice synthesis via [TTS vendor]
Storage & Pipeline: Firebase, Cloud Run, Redis

Results

Within two weeks of internal alpha, Podstack processed:

📬 7,200+ emails across 24 users
🔊 430+ podcast episodes generated
⏱️ Over 88% of users listened to at least one episode within 24 hours

Feedback highlighted:

A sense of “guilt-free” inbox clearing
Productivity boost from passively absorbing key insights
Preference for short, daily 5-7 min episodes

What We Learned

Users care more about relevance than voice quality.
Episode consistency (daily drops) builds habit retention.
HTML newsletters vary wildly — heuristic + LLM hybrid parsing worked best.

Next Steps

We're expanding Podstack with:

Smart queues for prioritizing urgent reads
Voice personalization options
Integration with Notion and Readwise for archival

Podstack is a perfect example of what we aim to build at Rola Labs — practical AI tools that solve real user pain points with thoughtful automation.

Want something like this for your idea?
Send us a brief →