[azhawkes]

I'm the solo founder of Loadster (https://loadster.app) which has been profitable for a while and is now making me pretty close to a comfortable living.

I really enjoy the freedom and variety that comes with running a SaaS, and while the grass is always greener on the other side, I think I'm pretty much ruined for a regular job at this point. So beware!

One piece of unsolicited advice: if I were to start all over again as a solo bootstrapper, I would probably do something less technical. As others here have pointed out, there is a LOT more to running a SaaS than just building software, and it's often hard to find the time or brain space to give all the facets of your business the attention they deserve.

[rlonn]

Great work w Loadster. Didn't know it was a solo effort! I founded another load testing service - loadimpact.com, which is doing very well but certainly not a solo effort and has taken on investors along the way. I'm now trying to build something solo again, and would echo the advice to go for something less technical. Plus do B2B if financial success is important. I have three projects atm, trying to figure out which has the most potential - 2 of them are consumer apps (one unreleased, the other is a crossword game for kids - https://puzzlepirate.net) and one is a technical data storage SaaS for IoT sensor data storage (https://pushdata.io). The latter product is a lot more demanding, from most aspects. A simple and maintenance-free product is worth gold when you're running it solo.

[azhawkes]

Hey thanks Ragnar it's good to hear from you! Those both look great. Pushdata seems like an excellent idea – having done some stuff with InfluxDB I can definitely see the appeal of fully managed timeseries data without the worry of scaling it yourself. I guess Amazon and other big competitors will want this space too, but they won't be able to touch the simplicity of something like this. Best of luck!

[plasma]

Hey azhawkes; I've used loader.io and a few other services to do load tests.

For all of them, I wish I had a "pay for credits" model, where I could just pay for each load test I ran, instead of a subscription.

Load tests for me were infrequent and feature/ad-hoc/test-case specific, so I wish I had a billing model that fit my use case.

Wish you the best!

[azhawkes]

Thanks!! The Loadster Fuel option is just that, where you buy a bunch of "fuel" (credits) and use it whenever you need to run tests. Each unit of fuel is enough to run 1 virtual user for 1 hour and it never expires.

Loadster's revenue split is something like 70/30 between MRR and Fuel. As the founder, of course, I love MRR, but many customers are like you and load testing is an occasional activity where it wouldn't make sense to have a recurring commitment.

I do always wonder how much friction it causes having two separate pricing models... pricing is hard.

[nathan_f77]

Thanks for sharing your experience! I've been thinking about adding a credit system to my PDF generation product [1], so that people can perform one-time batch jobs, or for companies that have very seasonal workloads. I agree that recurring revenue is the best, but there's also nothing stopping people from suspending their account if they don't need it anymore. I haven't had anyone do that so far, so maybe this would be the wrong direction.

[1] https://formapi.io

[rlonn]

At Load Impact we faced this issue also: us wanting a predictable revenue stream, and the customers often preferring a pay-as-you-go model. In my mind you should always be customer-centric and offer what the customers want. Otherwise someone else will.