I've finally decided to shut Letterfuel down.
Letterfuel is a tool that makes it easy to build and send curated/round-up newsletters. You collect links throughout the week with a Chrome extension, then when you're ready to send, you drag your chosen links onto the builder, make some edits, then publish. It's rough around the edges, but the process is smooth.
I decided to build it after about a year of putting together a weekly design newsletter. I tried various newsletter services (MailChimp, MailerLite etc) and the process was ridiculous. Every week, I'd clone the previous email, delete the old content, then spend ages copying and pasting links, titles, images, etc. It was stressful, time-consuming, and error-prone.
I looked around for services that would make the process easier and found Revue. It seemed perfect, but I couldn't use it at the time because the images I was added to my newsletters were too large. There were a couple of other reasons it was a no-go, but as someone who was itching to build a solution myself, I didn't need much convincing.
At the time, I was quite early on in my career as a developer. I'd just started to become comfortable with Django and started dabbling with React. I knew that building a successful SaaS so early in my career was a long shot, but I justified it by telling myself that even if the product fails, I would learn a lot about Django, React, Stripe etc.
Six months later, Letterfuel was ready to go. I struggled through building with React, flip-flopped about what the product should and shouldn't do, and didn't manage to get more than a handful of beta testers. Launching on Product Hunt was the only logical thing to do.
The launch went reasonably well. About ~150 people created a newsletter, although the large majority were just dabbling. A few people stuck around and some ended up becoming paid customers. It looked promising.
What went wrong
The biggest issue was that by the time I launched, I hated the codebase and was sick of working on it. I wasn't anywhere near experienced enough with React to try build so complex with it. The result was a messy codebase that I hardly understood. This killed momentum after launch as I struggled to convince myself to make improvements and add features.
The next issue was that Letterfuel was attracting users who had never built newsletters before. This seemed great initially and may have been a good sign for a venture backed startup looking to grow a category, but as a solo founder, I didn't have the time or resources to help educate customers and increase their chances of success.
It turns out, no matter how streamlined the tool, starting and growing a newsletter requires a lot of work and patience. Most people gave up on their newsletters after a couple of weeks or so.
With enough motivation, I'm convinced I could have overcome the above issues, but some of my other products were seeing more success, so it was hard to justify spending more time and energy on Letterfuel.
I still use Letterfuel every week to curate the newsletters I send. I tried using one of the native builders with another provider and was instantly reminded why I built Letterfuel in the first place.
Instead of fully shutting Letterfuel down, I'm turning off the ability to send. That's the part that requires the most maintenance and cost. Instead, you can now build newsletters in Letterfuel, then export them to be sent with other providers.
That's how Letterfuel was supposed to work in the first place, but I somehow convinced myself that the builder without the ability to send wasn't enough.
That's essentially the story of Letterfuel from start to finish. I don't regret building it as it did indeed teach me a lot about code. It also saves me many hours of stress and frustration as I put together my newsletters every Friday. If I was to go back in time, I would have kept the scope tiny and would have launched sooner with way less features.