Hey folks, quick update about my startup, Glitter AI.
First, a word of thank you - I launched on this subreddit in May of 2024. I wrote openly about being both excited to launch - and scared.
That's because I went AGAINST common advice, and built for over 6 months before going live, with almost no market research.
Mind you, this was in the age of pre-Claude Code and stuff, and I coded A LOT of it myself.
I figured there was a market for a solution for creating SOPs (standard operating procedures) because I saw a bunch of Scribe alternatives out there, but I thought I could build a better version.
So I did :)
That said, my growth has been A LOT slower than I planned, and I'm disappointed:
Today, 2.5 years in, I've reached $18.2K MRR, and I want to share some of what I've learned.
So why am I disappointed?
When I launched, I said I would reach $1M ARR (about $83K MRR) within 1 year.
It's been 2.5 years, and I'm only 20% of the way there.
Meanwhile, I've seen other startups experience meteoric success, including some personal friends of mine (one buddy of mine went from $17K => $330K MRR in the last few months - $4M ARR) and it's hard not to compare myself.
That said, I wanted to take this opportunity and tell my story and some learnings.
Some interesting lessons from the last 2.5 years
1. Don't just copy - when I launched Glitter, I entered the market to compete as the best alternative to Scribe. However, beyond thinking "there's a market for an SOP solution" I had very little strategy.
I kept on building features and functionality (e.g. SOPs with AI Voiceover and AI Avatar) instead of doubling down on a specific persona and building for them.
Turns out that if you want to succeed as a bootstrapped startup against a multi-billion dollar company, you should probably pick a specific sub-niche (common advice, I know - but still...).
2. Invest in marketing BUT lean into your strengths - I hated marketing, and every time I tried to do it, it felt disingenuous.
So I always went back to coding features.
What took me a long time to understand is that I should do marketing my own way, by leaning into my coding skill as a strength - I love building, so now I build systems for marketing (more on this later).
3. Do the customer acquisition math - Scribe pricing starts low, but because they sell to enterprise, they end up making a lot of money per customer on average. If your LTV (amount of money you make per customer in their lifetime) is $200 and the market is saturated, you can't really invest in paid ads, which will limit channels for growth.
That was the case for Glitter, and I just couldn't run a business on a $20/month plan for something a whole company used.
So I changed it.
What am I doing differently today:
Applying learnings from all 3 points above.
1. Differentiate Glitter AI as an AI SOP generator that your whole company can use for a single price - I made viewer seats free, and added AI Voiceover and AI Avatar which Scribe doesn't do.
Also, do research. When you Google or ask ChatGPT something like "Does Scribe charge for viewers?" you realize a lot of are disappointed that with Scribe, every team member is a paid seat, including people who only read, and viewer licenses are Enterprise only. So on Glitter AI viewers are free. Here's hoping that was a smart move :)
2. Marketing as code - I love building so I'm leaning into that. Instead of "yelling into the void" on socials and trying different things to see what sticks, I create mini-tools that sell the product for me.
For example, I recently built a Scribe pricing calculator. I'm not going to link here but google "Scribe vs. Glitter AI" (in quotes and you'll find it if you want to have a look - I recently dropped the ball on SEO so I don't rank first without the quotes anymore).
3. I changed the offer (and raised the minimum price) - this one is tricky. I was sure that to compete with Scribe, I would have to have lower entry prices. But Scribe's model works differently - they get you in "cheap" and then upsell on their "enterprise" plan with a sales-led motion.
With Glitter AI, I decided that I want pricing to be transparent, so you pay a fixed price for unlimited viewers. My self-service plans have lots of features that Scribe locks behind their enterprise offering only (like translation and SSO). Glitter AI translates to 99 languages on a self-service plan, and I offer SSO as an add-on on any plan.
This did in practice mean I had to raise the entry price for new customers, but it lets me invest more in paid acquisition than I could before (though I'm still figuring it out).
Anyway, just thought some of you may be interested in hearing about my journey building Glitter AI. While it's not (yet) the crazy success I want it to be, the company is growing, and I'm having the time of my life.
Ask me anything :) I'm here to answer.