Opinion
Why owning your software is cheaper than renting it after year two
Renting software is cheaper in year one, and in my view owning it is cheaper from about year two on, mostly because of the hours, not the subscriptions.
Here is the math I use, and when it does not hold. Most owners who act on it start with Connect at $5,000 or Consolidate at $20,000.
What this usually looks like
A typical stack: a CRM, a scheduler, a form builder, a project tool, a field app and an accounting tool. Each bills per seat, every month, forever, and the price goes up when the vendor decides.
Then the hidden part: someone spends hours every week re-typing between those tools. Those hours do not show up on any invoice, but you pay for them.
Say ten tools at $40 a seat for 12 people. That is $4,800 a month, or $57,600 a year, before anyone counts the re-typing.
Why it got this way
SaaS is priced to be easy to start and hard to leave. Every year the bill renews, and every year the data is harder to move.
You pay for 100 percent of each tool and use 10 to 15 percent. Owning means paying once for the part you use, plus support if you want it.
The fix, three ways
Connect, $5,000
keep the tools, fix the plumbing
For $5,000, keep your tools and stop the re-typing. This is the cheapest way to test the math: if the hours saved do not pay for it within a year, owning more is not for you.
Consolidate, $20,000
one app for the daily work, keep what still earns its place
For $20,000 and optional support at $1,500 a month, the daily work moves into one app you own, and the tools it replaces get cancelled. Compare that with what those tools and the re-typing cost you each year.
Replace, $75,000 to $100,000
rebuild the part of the big platform you actually use
For $75,000 to $100,000, the big platform goes, and the part you use is yours. This is where the year-two math is strongest, because large platforms are the most expensive rent.
What it looks like when it's done
Spark is the real-world version of this math. Spark replaced four tools and a payroll done by hand in Excel. It saves BigDog Solar 20+ hours a week of manual work and roughly $60,000 a year, and billing issues now live in 1 system instead of 3.
Read more: Spark for BigDog Solar.
When you should not do this
If your stack costs under roughly $500 a month and nothing is re-typed by hand, renting is cheaper. Keep renting.
If you want to rent the result, owning is the wrong model. Everything I build is yours, and if you want a subscription product, that is a different business.
Questions about this
Is it cheaper to build or buy software?
Buying is cheaper to start. Building costs more up front and less every year after, because you stop paying for seats and features you do not use, and you stop paying people to re-type between tools. In my view, for most small businesses the lines cross around year two.
What does owning software cost after it is built?
Hosting in your own cloud accounts, billed at cost, plus an optional support plan: $500 a month for Connect up to $7,500 for Exclusive. You can cancel support with 30 days' notice and keep everything.
What if my business changes after the build?
You change the software. Small changes are covered by the monthly block of hours in a support plan, and bigger ones are a fixed-price change order. Because you own the code, any developer can make changes, not just me.