A Vendor Tour Is Not a Decision
What a piece of software actually costs once you count the hours it adds, and what it costs to leave.
- LISTED · YEAR
- 1,188
- SETUP · 12 HRS
- 1,716
- WORKAROUNDS · 4 HRS/MO
- 6,864
- TRUE FIRST YEAR
- 9,768
Composite. Not a quoted price from any vendor. Not legal, tax, or billing advice.
Total cost of ownership
Procurement · digital operationsThe price is the part you can see. The cost is the price plus the hours plus the exit.
Everything you buy has a listed price and a real one. The real one is four numbers added together. What you pay. What it takes to set up. What it does to your hours every month. And what it costs to leave. The sum has a name: total cost of ownership.
Composite, one software decision I have watched more than once. Listed for the year: $1,188. That is the number on the website. Setup took 12 hours, which at a real hourly is $1,716. Then it added 4 hours a month of working around it — $6,864 a year.
True first year: $9,768. The sticker was twelve percent of it.
That third number decides most purchases and almost nobody prices it. A tool that saves four hours a month is not saving time in the abstract. It is saving four hours times what your hour is actually worth. A tool that adds two hours a month charges you that same rate. Quietly. Every month. Forever.
The fourth number is switching cost — the price of being wrong. It is built from data you cannot get out, work you would have to redo, and habits people have already formed. Vendors compete hard on the first number and design the fourth one. That is why the cheapest thing to buy is so often the most expensive thing to have bought.
So the question is never which is best. It is three questions. What does each cost in total? Does it move the one deadline that matters? What does it cost me if I am wrong? An hour with those three beats any demonstration.
Go deeper: three things a school would add
- Switching cost has an anatomy: data portability, retraining, re-integration and contract term. Ask about all four before you buy, because you cannot ask about them credibly afterwards.
- The export test is the cleanest question in procurement: can I get my data out, in a usable format, without asking permission? A vendor who cannot answer that plainly has told you their exit price.
- A tool that removes work and a tool that moves work look identical in a demonstration. The difference shows up in whose hours went up — usually the biller, usually silently.
- Buying expertise has an order: discovery first, then judgment. Discovery is often already published by people near you and costs an afternoon; judgment is what you should actually be paying for. Paying an expert to do the discovery is the most common way a small budget disappears.
Work the case
Two decisions. Choose before you read the reasoning. The wrong answer is the one worth understanding.
Tool A costs three times Tool B. Tool A removes five hours of work a month; Tool B adds one.
Choose one — the reasoning opens after you commit.
A vendor offers thirty percent off if you commit for two years.
Choose one — the reasoning opens after you commit.
Ask which system to use in any operator room and you get names. Eleven names, no reasons, and whoever answered first sets the consensus.
The names are the least transferable part of the answer, because the right choice depends on numbers that belong to you.
Here's the honest version from running one: for a solo practice, and even at two providers, the enterprise-grade systems aren't justified. I ran AdvancedMD and it couldn't earn its complexity at that size, the capability was real and the overhead of holding it was larger than what it returned.
What works at this scale is the combination of simple and cheap, and when you're starting, you count every dollar.
That's what makes SimplePractice the default.
But defaults have holes, and the second half of the answer is what fills them. What the practice system doesn't do, the documentation load, the clinical thinking captured while you work, is what we built Psynopsis to do beside it.
Disclosure: I own Psynopsis. Treat that as a data point with an interest attached, not a recommendation from a neutral party. The lesson is the pairing, not the product: a simple, cheap system of record plus one tool that removes the specific hours you actually lose.
Run any candidate through the four questions. What does it list at? What will setup cost, in hours? What does it add or remove every month, priced at your real hourly, the one from the utilization board, not the one on your offer letter?
And what would it cost to leave? Can you export the panel, the notes, the ledger, without asking permission?
The last of those is where people get hurt. A system you can't leave is a system whose price you no longer control, and the moment to establish the exit is before you have anything in it.
Before you buy expertise, harvest what's already public and already local. Every established practice in your city has patient-facing artifacts: consents, notices, policies, a fee page, a directory profile. Every one of them has been shaped by the same rules you practise under.
That's a free jurisdiction scan sitting in plain sight. Read four of them before you pay anyone to tell you what your state expects, then pay for review rather than for discovery.
The habit generalises: the local market publishes its answers, and almost nobody reads them.
The cheapest version of this lesson isn't software at all. Register with a group purchasing organisation and order the routine tests directly, for a few dollars each.
Don't, and the alternative arrives as a surprise bill: the patient is upset, you write another note, and you spend an afternoon arguing for coverage of a test they wanted anyway.
Price that afternoon at your real hourly and the decision makes itself. This is the hole that quietly makes a practice miserable, and it's a purchasing decision, not a billing one.
One more thing the rooms never say: a purchase that doesn't shorten the longest gate doesn't move your opening date.
Buy those anyway if they're cheap and they help, but buy them knowing they're comfort, not progress.
The question, as it actually gets asked
Composite questions — blended from the operator rooms, never one person's words.Read these for the method, not the answer. The next question you have will not be on this list — the point is that the concept above answers it anyway.
Which concept this really isTotal cost of ownership. The listed price is the smallest of four terms: price, setup, the hours it adds or removes each month, and what leaving costs.
How you work it outPrice all four terms before you look at a demonstration. Then the size answer: at solo, and even at two providers, the enterprise-grade systems don't earn their complexity. I ran one, and the overhead of holding it was larger than what it returned. What works at this scale is simple and cheap, because when you're starting you count every dollar. Pair a simple system of record with one tool that removes the specific hours you actually lose.
The mental model that makes it hardChoosing on demos and enthusiasm. Every tool demos well; the ones that hurt do so at the claim, six weeks later.
The Monday rule, what do you actually run this week?
The sticker is the smallest number in the decision. Price the hours, and price the way out.
Connected lessons
- show-rate The hourly this page multiplies by is the one that board computes — a stack decision needs your real rate, not your offer letter’s.
- four-gates A purchase that does not shorten the longest gate has not moved your opening date.
- the-split Software is a fixed cost: it is paid out of the pile of contributions, so it is priced in visits.
This is one of eleven lessons, free to read. One arrives by email each week, on composite numbers you can check.