Onboarding and risk management tools for financial firms. Monthly spend is about $1500, which is about 10% of revenue. I can probably spend some time and bring that spend down to $1000 with some work (but currently time is better spent getting revenue up)
Edit: I would like to add that with a combination of fly.io and digital ocean the actual server costs are way under $100. Most of the costs are paying for other Saas needed for the product (document analysis, emails etc) and tools (github, copilot, sentry, etc)
B2B SaaS here, if you are asking about the gross margin (revenue minus cost of goods sold), I've always kept it above 90%. I'm careful not to use expensive hosting (like AWS) and I keep a close eye on COGS at all times.
I run a serverless stack on AWS and my only cost is a minimal RDS instance that runs me about $15/month.
I'm genuinely curious as to where this reputation for expensive hosting comes from.
I'm surprised there are no comments here, I think this is a very interesting question.
We are b2b and need to have high isolation between customers which means dedicated resources. Because of this, it's likely out our margins are lower than many other SaaS categories.
We target 80-90% margin on most of our subscription plans.
> We are b2b and need to have high isolation between customers which means dedicated resources.
I'm curious - is that to avoid "noisy-neighbour" and provide performance guarantees through dedicated hardware, is it for security/privacy/etc or is it a combination of factors?
70% gross margins is acceptable, 80% gross margins are normal, 90% are exceptional. Remember, in most cases you are losing 4% to payment processing right off the top.
3% on payment processing, subscription management and collection.
7% on AWS.
10% on product support, account management and retention.
If you’re fast growing this is about right. If you’re starting to slow down then you can afford to spend more time optimising these, eg 2, 4 and 9, giving you a GM of 85% which is elite for a B2B SaaS.
I like the % on AWS (or infra let’s say) because it is possible to be too cheap and get engineers to solve problems the clouds can offer you a solution off the shelf. Since the Engineers are making the decisions the may balk at spending $10k/m for something in a cloud but if it is framed as 0.1% ARR it can be put into perspective: this is 1/1000th of your infra spend, is it worth it? Or would you rather have X? The idea being to untie personal spending decisions and fear i to a business decision.
Crypto accounting b2b tool here. We spend 90% of our time doing onboarding of accountants. 2 ppl company, before salary our operating costs are around 5%. We just have a simple Frontend deployed on heroku and then a bunch of nodejs workers pulling the data from various API. Our tech cost are probably less than 1%, the rest are tools and api cost.
B2B is cheap if you don’t do stupid things like running kubernetes :)
50$ on Digital Ocean and it brings in 2-3k$ per month.
I'm selling API access to data (the servers process plenty of requests, I need to scale them up every once and then).
I've read about someone on linkedin with a 7$ VPS making 500k per year (half of that is selling data, the other half is talking to people via mail) but barely using the server, it's just to keep up a fancy dashboard.
I worked at a SaaS startup, where after 5 years, the monthly earnings would barely pay for the AWS bill and 1 mid level engineer. Gross profit was single digit percentage, based on the official COGS. COGS was likely low, since it expected to make the bad numbers look better. Most of the company got laid off. Even then, like 150% to 200% of revenue was going to payroll.
Edit: I would like to add that with a combination of fly.io and digital ocean the actual server costs are way under $100. Most of the costs are paying for other Saas needed for the product (document analysis, emails etc) and tools (github, copilot, sentry, etc)