← The blogOwnership

"Price Certainty Is Architecture, Not a Promise"

Contents

#The promise every SaaS makes

Somewhere in the onboarding flow of almost every subscription tool, there is a moment that feels like safety. A pricing page that says "from $9/month." A founder who tweets "we will keep the free tier free." A blog post that says "we will never sell your data." A sales call that ends with "and your price is locked as long as you stay."

They are all promises. And promises are just words with a timestamp.

I am not saying the people who make them are lying. Most of them believe it when they say it. The problem is not intent. The problem is that a promise is only as good as the incentives of the person who made it, and incentives change. A founder who promises "no price hikes" is gone the day the company sells. A board that promised "free forever" reverses it the quarter growth stalls. The promise did not break. The people who made it just stopped being the people in charge.

This is not cynicism. It is the last three years of software pricing, playing out on repeat.

#The thing a promise cannot survive

Here is the test that separates a promise from a structure

Renting
Data on their servers. Price can change overnight. History is a hostage. Works only online.
Owning
Data on your disk. Price is done at purchase. History is a file you hold. Works offline, forever.
: what happens when the person who made the promise no longer works there?

If the answer is "the new owner can reprice you," you had a promise. If the answer is "the new owner cannot reprice you, because there is nothing to reprice," you had a structure.

Harvest was a beloved product for twenty years. The people who ran it for most of that time were, by every account, genuine about serving small agencies. Then Bending Spoons bought it, and the reported price increase was around 1500%. The people who made the old promises did not break them. They just stopped being the ones making decisions. The new decision-makers faced the architecture they inherited, a subscription with a captive audience, and they used it.

A promise would not have survived that transition. A structure would have.

#What price certainty looks like in owntime

✓License does not expire
✓Vendor cannot reprice you
✓Data is a file you hold
✓Works offline, forever

I want to show you the difference with the specific case of owntime, because I have been on both sides of this line.

owntime is a one-time purchase. $99 during the early bird, $149 after. One payment, per license, and one license covers up to three people, one Mac each. The transaction ends when you pay. There is no monthly invoice, no subscription to renew, no account that can be repriced.

Now, the promise version of this would be: "I, Amit, promise never to raise your price." And I mean it. But the honest version is: it does not matter whether I mean it, because I structurally cannot do it. There is no server holding your license. There is no subscription record to update. The license key verifies against a signature baked into the app, offline, once. Even if I wanted to charge you again, there is no mechanism to do it. The lever does not exist.

That is the difference between a promise and a structure. A promise depends on my character. A structure depends on the absence of a lever. I will be honest and say you should not have to trust my character. You should be able to verify the architecture and see for yourself that the lever is not there.

#The offline test, again

I keep coming back to this because it is the cleanest demonstration. The night I decided to build owntime, I turned off my wifi on purpose. If the app I was imagining needed a server, it would fail the test, and I would know the whole idea was a subscription wearing a buy-once label.

With the wifi off, the timer ticked. The entries saved. The invoice generated. The license key verified. Nothing failed, because nothing was trying to phone home. The app works the same online and offline, because the only thing a network connection adds is downloading updates, and updates are signed and optional.

If owntime had a server, the offline test would have failed, and I would have known I was building the same thing I was trying to leave. The fact that it passes is not a feature I added. It is the absence of a feature I refused to add.

#Why this matters beyond pricing

The worst case of buying software is better than the best case of renting it.

The architecture point is bigger than price, and it is the part people miss.

A tool with no server cannot reprice you. That is the pricing claim. But the same architecture also means it cannot hold your data hostage, because your data is a file on your disk. It cannot force an update you do not want, because updates are signed and optional. It cannot disappear when the company does, because the app does not depend on the company existing. It cannot read your timesheet, because there is no server to send it to.

Every one of those is the same structural point applied to a different fear. The absence of a server is not just a pricing feature. It is the thing that makes every other promise unnecessary, because the lever for breaking them does not exist.

#The honest limits of structure

I want to be fair about what a structure does not protect you from, because the list is real.

A structure cannot protect you from me disappearing and the app stopping improving. That is the trade. A structure cannot protect you from a bug that does not get fixed because there is no team to fix it. A structure cannot protect you from the app falling behind on macOS compatibility three years from now. These are real risks, and they are the reason a buy-once tool from a solo developer is not the right choice for everyone.

What a structure does protect you from is the specific, predictable, well-documented risk of being repriced, locked in, or held hostage by a vendor whose incentives changed after an acquisition. That is the risk that hit Harvest. That is the risk that hit Octopus Deploy. That is the risk that hits every subscription tool the moment the economics turn.

You are trading the risk of "the developer stops" for the risk of "the vendor reprices." Those are not the same size of risk, and in the current market, one of them is happening every month and the other has not happened yet.

#The bottom line

A promise not to reprice you is just words with a timestamp. Price certainty is what happens when the vendor structurally cannot reprice you, because the lever does not exist.

You should not have to trust a founder's character to keep your price stable. You should be able to turn off your wifi, watch the app keep working, and know that the thing that would let them charge you again was never built in the first place.

That is what buy-once means. Not a promise. An absence.

Try owntime free for 14 days · See the honest pricing · Read what we will not build


Sources & further reading

  • owntime pricing (https://owntime.click/pricing), accessed 2026-09-02
  • Hacker News discussion of Harvest repricing, August 2026

Your time tracker can't raise your price if you own it.

Try owntime free for 14 days. No email wall. Or buy once for $99 early bird.

Download free trialSee pricing