7 ms·
What I learned running a SaaS for a year
- rozenmd 5y agoHey, I'm the creator of OnlineOrNot. I mainly started building it because I wanted to learn more about running a business, marketing, and SEO (all the technical parts of the business were relatively easy in comparison).
- ushakov 5y agoi have same feeling for technical vs marketing side i don’t regret launching a failing SaaS for me or for client because it gave me the opportunity to learn ins and outs for almost no cost by the way, how does you product compare to instatus.com?
- rozenmd 5y agoAt the moment, I just handle the monitoring (static websites, APIs, and now JavaScript apps) and alerts. You can then use that data to write accurate status updates on your system's status page, but we do not host public status pages.
- eatonphil 5y agoI'm having the same experience, 8 months in. The software is by far the easy part.
- lbriner 5y agoPricing is hard but we also make it harder than it needs to be, especially if we have doubts about our product and the imposter syndrome (our product isn't perfect so can I charge premium?) Once (as the OP pointed out) we see that we are solving a problem rather than selling a service/product, it becomes easier to price, at least in the right ballpark. If you are saving perhaps 10 hours of time per week for a customer, you can price roughly how much you save them (probably a lot) and then you can charge accordingly. If you are saving whole employees from needing to be employed, then the sky is the limit. In the same way, if you were renting a Bouncer to a nightclub, you wouldn't be bothered that they aren't quite tall enough or don't speak French. Those things would be nice to haves but they still perform a job!
- bachmeier 5y ago> If you are saving perhaps 10 hours of time per week for a customer, you can price roughly how much you save them (probably a lot) and then you can charge accordingly. You're omitting one critical piece of pricing. Saving someone 10 hours per week doesn't mean much unless you can convince them they will save 10 hours per week before they've used the product and you actually deliver so that they stick with you in the future. Oh, and if there's a competitor that saves 9 hours per week at half the cost, you're not going to have many customers.
- mjwhansen 5y agoThanks for sharing! This is such a great example of having your pre-existing belief questioned, noticing user behavior, and adapting accordingly: > Contrary to popular belief (for B2B SaaS), people actually work from their phones. > Something like 50% of traffic to OnlineOrNot.com comes from folks on mobile. They tend to quickly create an account, add a few pages to monitor, then eventually get on their laptop/desktop to review their checks from time to time. > For about 6 months I didn't support mobile well, and folks that signed up on their phone churned rapidly. I eventually took the time to build responsive views for mobile, and now new mobile users are sticking around. Just curious -- in addition observing their behavior, have you talked to those customers (either via support/email or interviews) to understand this workflow?
- rozenmd 5y agoI've mainly been emailing customers to try understand their processes better. I read your book though, and I've been working up the courage to do customer interviews over Zoom/calls!
- mjwhansen 5y agooooh, you should share the email copy you're using on Twitter! would happily RT, so helpful to have real-life examples let me know how it goes when you work up to interviews! :)
- ushakov 5y agonot the OP, but i have actually observed a similar thing in the past - people land on marketing page on the phone, sign up, go to dashboard, etc. then then go to desktop and pick up where they left off
- TeeWEE 5y agoHow is your Monetisation going? I guess its hard to make a living from building your own product (I'm trying todo the same). But not impossible... I often sees these "twitter i'm successful founders" who keep mentioning "I'm almost doing 10k MRR" but not talking about costs... Its easy todo 10k per month, if you spend 15k on google ;-) I'm doing close to 2k per month on MRR but costs are also quite high.
- TeeWEE 5y agoAh found it in your blog! https://onlineornot.com/eight-months-of-onlineornot https://onlineornot.com/eight-months-of-onlineornot
- rozenmd 5y ago$3.5k USD total over the year, roughly.
- TeeWEE 5y agoNot bad! What's invaluable is what you are currently learning about running a business. Something I learned was that I totally underestimated marketing :p
- sjapps 5y ago
- dpcan 5y agoAm I reading this correctly that you worked 2 hours a day and only made $3.5k all year? Or about $300 a month?
- bachmeier 5y agoHe gets a valuable education, so basically he's getting free tuition plus enough money to buy a good lunch every day.
- deleted 5y ago
- tommiegannert 5y ago> Use analytics, set-up funnel tracking Any thoughts on which analytics services, regarding privacy and "acceptable to users" kind of metrics? (Former Google employee, who don't quite get the hatred Google receives online, but feeling that Google Analytics has a really complex UI for my needs.)
- XCSme 5y agoIf you like self-hosting: https://www.uxwizz.com/ https://www.uxwizz.com/
- rozenmd 5y agoI use Plausible analytics for site analytics - you can do quite a bit just by firing events when users land on a page (don't need to know who they are, just need a count of people that see page A, B, and C).
- withinboredom 5y agoI’ve found Azure’s Application Insights to be quite privacy focused. You can host the instance in whatever region you prefer. The UI isn’t all that great, but it seems pretty decent.
- cuu508 5y agoAre any Azure regions currently operated by independent companies that license Azure code from Microsoft? (if not, what difference does the region make privacy-wise, if the parent US company can ultimately still access the data?)
- withinboredom 5y agoDoes it matter if you’ve turned off all PII collection? If that’s your threat model to your own app though (ie, doing stuff that’s illegal in the US but not where you serve traffic), self hosting is probably the way to go.
- 5y ago
- vincvinc 5y ago> half of my time went to actually solving the problem I wanted to solve (knowing if a site is down, and alerting folks when that happens). The other half went to building a SaaS platform around that problem. You’d think in today’s software landscape there would be tools to simplify or automate the part or building a SaaS business that everyone needs to do. He mentions Stripe for payments of course, but isn’t there anything broader to get set up more quickly? Like a shopify for SaaS businesses?
- Veuxdo 5y agoStripe pretty much is that. There's only so much abstraction you can do on payment processing.
- bytebln 5y agoPaddle.com is pretty simple. I implemented Paddle in my App in less than 2 hours. Stripe API became too complex for my needs.
- PhilipA 5y agoDepending on the technology stack there are frameworks like SaaS Starter Kit for Phoenix: https://fullstackphoenix.com/saas-starter-kit https://fullstackphoenix.com/saas-starter-kit
- rozenmd 5y agoThere probably is, but I like to avoid sharing around customer data unnecessarily.
- dbbk 5y ago> Like a shopify for SaaS businesses? This is Paddle. As a one-person startup, there's no way I could go with Stripe. The tax implications alone are too much.
- zild3d 5y agothat's still just payments provider vs "Shopify for SaaS" which handles a lot more
- tnolet 5y agoWelcome to the club! Reminds me a lot of the first ~1,5 years hacking on my then bootstrapped monitoring product.