I've been a full-stack engineer since 2007. Sales, marketing, and networking has never been my strong suit. In June I decided to do something about that.
Molecule.dev began as a passion project of mine in 2021, something I've been indirectly working towards since the start of my professional career 15 years ago. I've always wanted to minimize the time and effort required to build high quality applications, because I've always believed that innovation is the key to improving everyone's quality of life. The faster people can throw their ideas against the wall, the sooner they can find what sticks. My goal has always been to accelerate innovation.
It became very clear that for Molecule.dev to catch on, as with any serious tech company, I would need to find product-market fit for short term traction, while positioning it for the long term vision of instant integration of almost any tool, feature, and/or platform into existing systems.
Very early on, at the beginning of the year before the initial prototype was ready, I made one very unexpected sale, which I used as the guiding light for the initial (assumed) product-market fit and targeting for early users. I stayed up all night and made sure customer #1 had everything he was expecting (and actually gave him more than that), and in less than 2 weeks he was using it in production within an existing system. To me, this was enough validation that I should continue working on Molecule.dev, and it really lit a fire under me to get it to a point where this kind of sale could be replicated and make the lives of developers a little bit easier. (Thank you, Randy!)
The question then became, "How do I find more people like Randy?" I made it my mission for the month of June to learn as much about sales and marketing as I reasonably could. I asked around, queried some search engines, and read some blog posts, and the consensus seemed to be to simply ask potential customers if they'd be interested, so that's what I did.
The most common advice seemed to be that cold emails are just what B2B companies do and that it is widely accepted, and maybe this is true for some, but it just doesn't feel right to me. Personally, I associate cold emails with spam, and I loathe spam so much that I made a custom email app to help combat it, built with a very early version with Molecule.dev. (A new and much improved open source, self-hosted version is coming!)
In an attempt to replicate that first sale, I used Apollo.io, a company based in San Francisco backed by YCombinator (leading me to think this is just how people do things these days), to send out roughly 800 emails to 400 targeted businesses (a cold email plus a followup email a few days later) over the span of a few weeks to potentially relevant startups and app development companies, with each message tailored to each recipient. Out of those 400, 3 responded kindly (thank you for being kind), and a couple responded very negatively (I don't blame them). A good portion of them probably quietly reported the emails as spam (and rightfully so).
The intention was to simply let companies know about Molecule.dev, as it's something developers can use to build high quality applications faster, but developers are particularly averse to any kind of direct marketing approach, so it did not go as intended.
While the short cold email campaign was a dead end, I learned some valuable lessons. Despite little technical progress being made throughout June, it was an incredibly productive month in terms of understanding how Molecule.dev can fit into existing markets (and maybe even eventually create a new one! more on that later) while determining the most realistic trajectory for growth.
So what's next for Molecule.dev?
Returning to the original plan of building out more features and integrations and putting out quality content that is actually of interest to people.
The public repositories on GitHub will be getting another makeover. The custom Git merge approach works well for assembling each Molecule, but it's very hard to follow along and contribute. We'll soon remove the hashed branches and refactor everything as a monorepo with installable packages like you're used to, with the option to "eject" when you want full control over the internals. There will still be a need for
mlcl, as some glue (extra code) will almost always be necessary to produce fully working applications from combinations of said packages.
At this time of this writing, there's only one preassembled Molecule, the Cross-Platform Typescript React Postgres SaaS Starter on Digital Ocean. Many more are on the way. In the mean time, you can use the advanced assembler to create a custom Molecule of your own and/or submit requests for nearly anything you want, and thanks to the cold email campaign, you may need check your spam folder.
Thanks for reading!