1,200 Looms Later: How Async Video Became My Development Superpower

I never planned to record over a thousand videos. In January 2021, I hit record on my first Loom — a quick demo for a client who couldn't make a call. Five years and 1,212 videos later, it's become one of the most impactful habits I've built as a founder and developer. Over 107 hours of recordings now document my demos, decisions, estimates, and code walkthroughs. Here's what I've learned.

The Numbers Tell a Story

I didn't set out to build an async video habit. It happened organically — and the growth curve surprised even me:

What started as an occasional demo tool became a daily habit. My peak month — February 2025 — saw 145 videos recorded. That's roughly 7 Looms per working day.

The average video is 5 minutes and 30 seconds long — but the median is just 1 minute 44 seconds. That gap tells an important story: most of my Looms are quick, focused messages under 3 minutes. The longer client demos and sprint walkthroughs pull the average up. No meeting preamble, no "can everyone hear me?", no waiting for latecomers.

Here's the full distribution:

Over a third of my recordings are under a minute — quick answers, visual pointers, "here's what I mean" moments. The majority (63%) are under 3 minutes. Only the deep demos and walkthroughs go longer.

What I Actually Use Loom For

Looking at my library, clear patterns emerge across five years of recordings:

Client Demos and PoCs (80+ videos)

This is where async video delivers the most impact for me. When I finish a sprint or build a proof of concept, I record a walkthrough instead of scheduling a demo call. The client watches it when they have time, pauses to take notes, and rewatches sections they want to discuss.

A few real examples from my account:

• "PoC SingBuddy" — walking a client through a new product concept before committing to a full sprint
• "Ambistream | Final demo for Sprint 5" — end-of-sprint delivery showing dynamic video layers in action
• "Amazon IVS Real-time Mobile Web Demo" — a technical PoC demonstrating live streaming capabilities for a music platform

The pattern is always the same: I open the app, walk through what I built, explain the decisions I made, and highlight what's next. Five minutes. Done.

Estimates and Financial Walkthroughs (119 videos)

This one surprised me when I looked at the data. I've recorded 119 videos walking clients through spreadsheets — project estimates, cost breakdowns, timeline projections.

Spreadsheets are notoriously hard to communicate over email. A 3-minute Loom where I point at cells, explain trade-offs, and walk through assumptions eliminates 90% of follow-up questions. The client can pause, rewind, and share the video with their team. It's far more effective than a PDF attachment with a two-line email.

Internal Tools and Dashboard Updates (153 videos)

My most recorded category — and the most underrated. Every internal tool I build, every dashboard update, every infrastructure change gets a quick Loom.

When a new team member joins MusicTech Lab and asks "how does the deployment pipeline work?" or "how do I use the Ableton export tool?", the answer is a 4-minute video, not a 30-minute onboarding session. These recordings aren't polished — they're raw, practical walkthroughs that serve as living documentation.

Code and Design Reviews (70 videos)

I started recording code walkthroughs when I realized my PR comments were getting too long. Instead of leaving 20 inline comments, I record a 5-minute Loom explaining the architecture decision, the trade-offs I considered, and what reviewers should watch for.

The same applies to design reviews. When I receive a Figma mockup, I record my feedback as a video — pointing at specific elements, suggesting alternatives, asking questions. It's more nuanced than written comments and much faster to produce.

Replacing Emails (24 videos)

Some things are just faster to show than type. I've recorded 24 videos that directly replaced what would have been long email threads — quick screen recordings answering questions that would have taken 500 words to write but only 90 seconds to show.

Why This Works for Me as a Founder

Running MusicTech Lab means I'm constantly switching between roles — writing code, reviewing designs, talking to clients, planning sprints, handling finances. Async video fits into all of these contexts because it's fast to produce and respects other people's time.

Time zone independence. Our clients span multiple time zones. A client in LA doesn't need to stay up for a demo with our team in Poland. They watch the Loom at 9 AM their time and leave comments. I respond the next morning. The work never stops, but nobody loses sleep.

Context preservation. When I record my thought process, I create an artifact. Six months later, when someone asks "why did we choose this architecture for the streaming module?", the answer exists as a searchable video — not as a fading memory from a call nobody recorded.

Protecting deep work. My most productive months — measured by commits and deployments — correlate with my highest Loom output. It sounds counterintuitive, but recording more videos means scheduling fewer meetings, which means more uninterrupted coding time.

What I'd Tell You If You're Starting

After five years and 1,200+ recordings, here's what I'd recommend:

Start with client demos

This is the highest-ROI use case. Record a 3-5 minute walkthrough of what you built, send it before the scheduled call. You'll immediately notice calls getting shorter and more productive. Clients love it — they can share the video with stakeholders who weren't on the original call.

Make it a reflex, not a process

I never made a rule about when to record. I just started hitting record whenever I caught myself typing a long explanation. Over time, it became second nature. The growth from 46 to 434 videos per year was entirely organic.

Keep it under 2 minutes

My median is 1 minute 44 seconds. Over 60% of my recordings are under 3 minutes. If a Loom is getting longer than a few minutes, I split it. Short videos get watched; long videos get bookmarked and forgotten.

Don't polish

The biggest barrier to async video is perfectionism. My Looms have typos on screen, "umm"s in the audio, and cursors that occasionally wander off. Nobody has ever complained. The content matters, not the production quality.

Build a searchable archive

At 1,212 videos, finding a specific recording became a challenge. I recently built a custom tool (an MCP server using Claude Code) that connects to my Loom account and lets me search, analyze, and generate statistics from my entire video library. That's actually how I pulled the data for this article — every number here comes from querying my own recordings programmatically.

The Compounding Effect

The real value of async video isn't in any single recording. It's in the compounding archive you build over time. My 1,212 videos are now:

• An onboarding library — new team members watch relevant Looms instead of attending knowledge-transfer sessions
• A decision log — architecture choices, business decisions, and trade-offs captured with full context
• A portfolio — demos and PoCs I can reference when scoping similar projects
• A content mine — this very article was built by analyzing my Loom data and reflecting on patterns I wouldn't have noticed without it

Every Loom I record today is a small investment in future efficiency. After 107 hours of recordings and over a thousand videos, I can say with confidence: async video isn't a productivity hack. For a developer running a remote-first agency, it's infrastructure.

I'm Mariusz, founder of MusicTech Lab — a software development agency focused on the music industry. If you're curious about how we work or want to discuss a project, let's talk.