Every subscription in your stack has a 24-month price, and it is the monthly figure multiplied by twenty-four. Nobody line-items that number, which is why it never gets challenged.
We are not going to tell you what your stack costs. We do not know, and any number we invented would be worth exactly what you paid for it. The arithmetic below is yours to run, and it takes about twenty minutes.
Where most people stall
The stall is not ignorance of the monthly price. Every finance team knows the monthly price. The stall is that software is budgeted monthly and owned never, so the comparison that would matter is never drawn.
A tool at $400 a month is a $9,600 decision over two years. Nobody approves a $9,600 line item without a conversation. Everybody approves $400 a month.
That is the tax: not the price, the framing. The monthly cadence makes each renewal too small to argue about, and the argument that never happens is the one that compounds.
The system: rent the commodity, own the differentiator
The useful question is not "is this expensive." It is "would owning this change anything." For most of the stack the honest answer is no, and renting is correct.
Rent anything where the vendor's version will always beat yours and the switching cost is low. Email delivery, payments, video calls, storage. You will never out-build these and you gain nothing by trying.
Own the layer where your process is the product. If a tool encodes how you qualify, score, route or follow up, then renting it means renting your own operating logic back from someone who can reprice it, deprecate it, or sell to a competitor.
The comparison, drawn once
| Rented | Owned | |
|---|---|---|
| Cost shape | Monthly, forever, rising at renewal | Build once, then maintenance |
| At 24 months | Monthly multiplied by 24, and still renting | Paid, and still yours |
| Your process | Fitted to the tool's model | The tool fitted to your model |
| Data | Exportable in their format | In your database |
| Repricing risk | Theirs to decide | None |
| Deprecation risk | Theirs to decide | None |
| Right when | Commodity, low switching cost | The process is the differentiator |
The diagnostic: twenty minutes, your numbers
Open the card statement, not the tool list. The tool list is aspirational; the card statement is true.
- List every recurring charge. Include the ones a department expenses without telling anyone.
- Multiply each by twenty-four. That is the two-year commitment you have already made by default.
- Mark each one R or O. R if the vendor will always be better and you could leave next quarter. O if it encodes how you specifically operate.
- Total the O column. That is what you are paying, over two years, to rent your own process.
The O total is usually smaller than people fear and larger than they expected, and it is the only number in this exercise that should change a decision.
What we admit
Owning is not free and the build is the cheap part. Maintenance, hosting, the person who understands it, and the cost of being your own support desk at 2am are all real, and they do not appear in a build quote.
Four cases where renting is simply right, and we will say so on a call rather than sell around it.
The tool is a genuine commodity. You will not out-build a payments processor, and owning one buys you nothing but liability.
Switching costs are low and the market is competitive. If you can leave next quarter, the vendor cannot hold you up, and renting is priced correctly.
Your process is not settled. Building a system around a workflow you are still arguing about produces expensive concrete. Rent until the shape stops moving.
The O total is small. If renting your own process costs less over two years than one engineer-month, own nothing and go back to work.
Our receipts
We have shipped 10 revenue MicroSaaS and 10 operating systems on this architecture, every one demoable on a fifteen-minute call. Two are public, named cases: Entrepreneurs Oasis and Lab Experience.
The Signal Sales Engine is 8 plugins and 118 skills with a published architecture, and three gates enforced in CI rather than by convention: a context floor, a plan hold, and version-hashed approval on every send. It is the same discipline we are describing, applied to ourselves.
Behind that: 500 people trained on Claude with operating systems built for more than 50 of them, and 20 years carrying B2B quota across the Gulf and the Levant at Cisco, Avaya and Uniphore. We are not describing a theory about software procurement. We have been on both sides of the invoice.
Where to start today
Run the four steps above on your own statement. It is twenty minutes and a spreadsheet.
If the O column totals something that makes you uncomfortable, that discomfort is the finding. Bring it to a call and we will tell you which line is worth owning and which three are not, including when the answer is that none of them are.
Frequently asked questions
What is the SaaS tax?
Should we build or buy our sales and marketing tools?
When is renting still the right answer?
Map your pipeline in 30 minutes.
A strategy call, not a pitch. We map your funnel, name the leaks, and tell you whether an install even makes sense.
Book the call