"Octopus Deploy Went From $50 to $15,000. Nobody Is Exempt"

#The comment that said it in four words

There is a Hacker News comment from August 2026 that does more work than most essays. It is short. It is angry. And it ends with a phrase that should worry every team that rents the tools it depends on:

so we built a clone instead

The comment was about Octopus Deploy. A deployment tool that, by the account in the thread, went from about $50 a month to about $15,000 a year. Same software, same team using it, same job to do. Different number, and not a "costs went up" number. A number that says: we looked at what you are locked into, and we priced it accordingly.

The software world spent a week being angry at Harvest for its reported ~1500% increase. Octopus Deploy got a comment. But the two stories are the same story, and the Octopus one is worth telling on its own, because it shows the pattern is not one company's failure. It is the shape of the industry.

---

#What Octopus Deploy is, for anyone outside devops

Octopus Deploy is a deployment automation tool. Teams use it to ship software to servers in a repeatable way. It is the kind of tool that becomes invisible infrastructure: nobody posts about it on launch day, everybody notices when it breaks, and the team that runs it has used it for years.

That is exactly the profile of a tool that can be repriced. It is not a shiny new SaaS people churn through. It is a piece of the machine. Replacing it means re-plumbing deployments, retraining the team, and risking a week of broken releases. The switching cost is the leverage.

The reported change, from around $50 a month to $15,000 a year, is a 25x jump on the annual figure

. Whatever the exact plan and timing, the shape is the same as Harvest: a tool that was cheap enough to forget became expensive enough to think about, and the team using it discovered they were not the customer with the leverage. The vendor was.

#Why the comment hit so hard

The four-word comment resonated because it names the only real response a small team has to repricing: leave, and build what you need yourself.

Building a clone is not a good outcome for anyone. It is a team taking engineers off its actual product to rebuild a deployment tool, or a time tracker, or whatever got repriced, because the alternative, paying the new number, is worse. The clone is a symptom. The disease is the architecture that made the repricing possible.

And the comment is also a quiet confession. The people writing it are the ones who can build a clone. Most teams cannot. Most teams pay, because the alternative is rebuilding years of accumulated process around a different tool. The comment is the voice of the minority that had a way out, and everyone else felt the weight of it.

I am one of the minority, and I want to be honest about what that felt like. When my own Harvest bill changed, I did not write a comment about building a clone. I sat in front of my laptop and started one. The first version of owntime was not a business plan. It was spite with a timer attached. I was so angry about the reprice that I built the thing I wanted to use, and only later did I realize other people were angry the same way. Every time I read a comment like "so we built a clone instead," I recognize the exact feeling that made me open a code editor that week.

#The pattern, stated plainly

Put the Octopus story next to the Harvest story and the pattern is unmistakable.

1. A tool becomes essential. Teams build their workflow around it, their history accumulates in it, their people learn it.
2. It is priced cheaply enough that nobody audits it. $50 here, $130 there, a free tier, a teaser rate. It is under the threshold where anyone does the math on the risk.
3. The vendor needs more revenue. Maybe growth stalled. Maybe the company got acquired, as Harvest was by Bending Spoons. Maybe the pricing was always a loss leader pointed at this exact moment.
4. The price changes. A lot. The jump is not 10%. It is 15x, or 25x, or "a number you have to sit down for."
5. The customer does the math on leaving, discovers the switching cost is years of history and process, and either pays or spends months building the clone.

The reason this keeps happening is that steps 1 and 2 are features, not bugs. Vendors want you dependent and they want you not auditing. The repricing is the moment the investment pays off.

#Why "nobody is exempt"

The title of this post is not drama. It is the structural argument.

Every subscription tool you rely on has the Octopus lever built in, because every subscription tool controls both the software and the data, and the switching cost is what makes the control valuable. It does not matter if the current vendor is nice. It does not matter if the founders are beloved. Harvest was a well-liked company for twenty years before the acquisition. The niceness was not the protection. There was no protection.

The exemption you actually want is not a vendor who promises not to raise prices. It is an architecture where raising prices is not possible, because there is nothing to reprice.

That is the difference between a promise and a structure. A promise is a sentence in a blog post that a future owner can delete. A structure is the absence of the lever entirely.

#The two tools in the story, and the third one

Here is where I have to be transparent, because this post is not neutral journalism. I make owntime, a buy-once, local-first time tracker and invoicing app. It exists for exactly the reason the Octopus comment exists. I got the Harvest email, I did the math, and I built the thing I wanted instead.

The difference between my story and the Octopus comment is that I am trying to sell the clone, not keep it internal. So let me say the honest thing about that.

owntime is not exempt from every risk. If I disappear, the tool stops improving. That is a real cost, and the changelog that proves updates ship is also the record of how much you would lose. What owntime is exempt from is the specific lever that Harvest and Octopus pulled: there is no server, no subscription, no monthly invoice, and no database of your history on my side of the fence. Your timesheet is one SQLite file on your disk. I cannot reprice you, because there is nothing to reprice, and I cannot hold your history hostage, because I never held it.

The worst case of buying owntime is the tool stops improving and your file keeps working. The worst case of renting Octopus is a 25x number and a year of re-plumbing. Those are not the same category of risk.

#What to check in your own stack tonight

You do not need to wait for the email. Go through the tools you depend on and ask five questions.

1. Which tools are under the audit threshold? Anything under $100 a month is probably not being reviewed. Those are the ones that can jump 25x without you having a plan.
2. Where does the data live? If your history is in their database, the switching cost is the hostage. If your data is a file you hold, you can always leave.
3. What would it cost to leave? Do the math now, while you are calm. If the answer is "weeks of work and years of lost history," you are not a customer. You are a captive.
4. Can you export today? Try it. A tool with a real export is a tool you can leave. A tool where export is buried or partial has priced the exit into your future.
5. Is the price a promise or a structure? If the vendor says "we will never raise prices," ask what happens when they sell. If the tool has no server and your data is local, the question answers itself.

#The uncomfortable corollary

The Octopus story has a corollary that the angry comments do not like to admit.

If you are the kind of team that can build a clone, you have a duty to your own time to notice that you are spending it rebuilding rented infrastructure instead of building your product. The clone is not free. It is engineer-months that did not go to your roadmap.

The better move is to stop renting the category entirely. Buy the tool that cannot be repriced, hold the file, and spend your clone-building energy on the thing you actually sell. That is the whole argument for owntime, and it is the argument this post is making with someone else's cautionary tale.

#FAQ

Did Octopus Deploy really raise prices that much?

A Hacker News comment from August 2026 cites Octopus Deploy going from about $50 a month to about $15,000 a year, a jump the commenter answered by building their own tool. Exact figures vary by plan, and the number stands as reported in that thread.

Is this the same pattern as Harvest?

Yes. Both are subscription tools where teams accumulated years of history and process, making switching costly, followed by a sharp price increase after the vendor needed more revenue. Harvest's increase came after acquisition by Bending Spoons.

How is owntime different from Octopus or Harvest?

owntime is a buy-once, local-first desktop app. There is no server and no subscription, so there is no monthly price to raise and no database of your history to hold. Your timesheet is a local SQLite file you keep.

What should I do if a tool I depend on gets repriced?

Do the math on leaving before you decide. If your data is exported and your workflow is portable, leave. If you are locked in, the repricing has already happened and the cheapest exit is the one you start today, not the one you start after the next increase.

Is building a clone ever the right answer?

Sometimes, but it is expensive. A clone spends your engineering time recreating rented infrastructure. Buying a tool that structurally cannot be repriced and keeping your data in a file you hold is usually cheaper than building, and it frees your time for your actual product.

#The bottom line

The Octopus Deploy comment is four words long and it is the most honest review of the SaaS era ever posted. "So we built a clone instead." Because that is what repricing does. It turns customers into builders, whether they can afford to be or not.

The teams that cannot build a clone pay. The teams that can, spend their roadmap rebuilding what they already rented. Either way, the vendor wins twice: once on the way up and once on the way out.

There is a third path. Own the tools that matter, hold the files, and make the repricing conversation irrelevant. Nobody is exempt from the pattern. But you can opt out of the architecture that makes it possible.

Buy owntime once, $99 early bird · See how your data stays yours · Try it free for 14 days

---