We recently went through a version of this with Asana when we discovered their "per seat" pricing is sold in groups of 5 seats. Going from 5->6 users adjusts your bill to 5->10 seats, doubling your price.
It isn't about the money, the principle is gross. Thankfully Asana's core business is now a prompt. We are nearly done replacing it with a native, self hosted client. Will post about it once it is hardened through use. Cancelling our Asana subscription this week.
To SaaS providers (myself included) it has never been easier to leave your service due to mistreatment. Take note.
You need to read the “Starting from” text on their pricing page right before those prices you see.
You can have those lower monthly fees if you pay a higher percentage of your billings.
The article acknowledged that.
> There’s also a “flex” plan which costs $12.65/mo… but that plan now skims an additional 2% on top of the usual ~3% card transaction fees.
The problems with the Harvest acquisition driving up prices has been discussed a lot among consultants and contractors. This isn’t one person reading their billing page wrong. This is happening to all of their customers.
It’s how the acquiring company operates: Acquire companies, increase billing, accept that most customers will leave but some will stay and make up for the losses.
Indeed, if aphyr billed $5000 a month, just to use a conservative figure if it's a small but significant part of business, that would be $12.65/mo + $100/mo. Oh, that would actually be cheaper. That hike is a doozy. And I think I went way too conservative on the figure.
I think it's including pay for usage. From the pricing page, beneath both Teams and Enterprise:
> Your base rate includes core features. As your team grows, additional invoices, projects, clients, and tasks are billed based on what you use, so you're never overpaying.
That's a bit bewildering, lol. Especially since they have annual plans.
This clarifies it:
> On Flex billing your first (sign-up) invoice usually covers seats only, because no projects, tasks, clients, or invoices have been tracked yet. Usage-based charges are calculated from the previous billing period, so they first appear at your renewal — not at sign-up. This is why a bill can jump (e.g. from a seats-only $44 to $404 once usage is added).
That's roughly a 10x multiplier, just like aphyr's jump from $13.32 to $133.98!
You're right something doesn't ad up, and seems fishy (to me, bizarre even). But that isn't the author, it's Harvest after the Bending Spoons aquisition.
Fable built me a version that covered all of my needs in an hour. (Clients pay me via ACH/wire - maybe it would have taken a day if I needed a stripe integration?). Runs on localhost with SQLite - incredibly basic, but I enjoy it twice as much as Harvest.
I believe your question answers itself, if you could do this then people would[2]. The next question might be 'What prevents the current LLM based coding tools from being able to do this?"
Now there has always been software that people have used that they "could" write for themselves but they didn't because they didn't want to maintain it. Seems like things that are like that would be good candidates for this right?
But this question shares at least a spiritual geneology with the response to some new startup code that does some thing, and the comment here is "Gee, I can't believe that got this funded, I could write that in a weekend."
The problem is, there are many things that are easy to describe but difficult to pin down, and reality is complicated. So many times I've gone into a project thinking it would be "easy" only to start uncovering layer after layer that are much more involved than I expected.
So no, you cannot '1-shot prompt' pretty much any significant software, and that goes doubly true if it affects your life in any potentially devastating way.
One of the things I'm constantly amused by[1] is the Solitaire program on iOS that is clearly being vibe coded these days because it comes up with daily challenges like 'win a hand in -2739551 moves'. Like no human would ever write code that screwed up basic math but LLMs do it all the time.
[1] Obviously I am easily amused :-). But it is sort of the 'pelican riding a bike' test where you can easily tell what a human would do vs what an LLM does.
[2] EDIT: another commentator says they vibe coded it in a weekend. That would support the hypothesis that the application could be at lease a small number of shots vibe coded. That said, I'd love to see that code and the prompts that led to it.
I’ve found that while vibe coding can be effective, anything with complexity takes more than a 1-shot prompt to get there. The amount of work that comprises a single scrum story is about the right amount to do with a single prompt and even that requires some review and revision. So 1-shot your payment system, hell no, but writing a payment system without writing code is possible although I would think it still would take enough time that I wouldn’t choose that over a reasonably priced SaaS.
I've had a Harvest account for over a decade, and it was always admittedly a little bit too cheap. But two months ago, they changed the pricing based on what you gross, so essentially putting a flat tax on my earnings.
I knew it was going to be more expensive, but when I got the bill this month and saw the price was over $200 and the Bending Spoons LLC tagline at the bottom, that was the end.
Claude is backing up all of my data via the API and will be vibe coding my next invoice.
We recently went through a version of this with Asana when we discovered their "per seat" pricing is sold in groups of 5 seats. Going from 5->6 users adjusts your bill to 5->10 seats, doubling your price.
It isn't about the money, the principle is gross. Thankfully Asana's core business is now a prompt. We are nearly done replacing it with a native, self hosted client. Will post about it once it is hardened through use. Cancelling our Asana subscription this week.
To SaaS providers (myself included) it has never been easier to leave your service due to mistreatment. Take note.
The Harvest website lists the following (annually billed) plans:
- Free (obviously free)
- Teams $9 a month
- Enterprise unlimited $14 a month
The writer of this article claims they got upgraded to $133.98 a month plan, but looking at the site, such plan does not even exist.
Something is a bit fishy here or they accidentally added an extra 3 to the $13.98 or mistook the yearly price for a monthly price.
You need to read the “Starting from” text on their pricing page right before those prices you see.
You can have those lower monthly fees if you pay a higher percentage of your billings.
The article acknowledged that.
> There’s also a “flex” plan which costs $12.65/mo… but that plan now skims an additional 2% on top of the usual ~3% card transaction fees.
The problems with the Harvest acquisition driving up prices has been discussed a lot among consultants and contractors. This isn’t one person reading their billing page wrong. This is happening to all of their customers.
It’s how the acquiring company operates: Acquire companies, increase billing, accept that most customers will leave but some will stay and make up for the losses.
Indeed, if aphyr billed $5000 a month, just to use a conservative figure if it's a small but significant part of business, that would be $12.65/mo + $100/mo. Oh, that would actually be cheaper. That hike is a doozy. And I think I went way too conservative on the figure.
I think it's including pay for usage. From the pricing page, beneath both Teams and Enterprise:
> Your base rate includes core features. As your team grows, additional invoices, projects, clients, and tasks are billed based on what you use, so you're never overpaying.
That's a bit bewildering, lol. Especially since they have annual plans.
This clarifies it:
> On Flex billing your first (sign-up) invoice usually covers seats only, because no projects, tasks, clients, or invoices have been tracked yet. Usage-based charges are calculated from the previous billing period, so they first appear at your renewal — not at sign-up. This is why a bill can jump (e.g. from a seats-only $44 to $404 once usage is added).
https://support.getharvest.com/hc/en-us/articles/31447072608...
That's roughly a 10x multiplier, just like aphyr's jump from $13.32 to $133.98!
You're right something doesn't ad up, and seems fishy (to me, bizarre even). But that isn't the author, it's Harvest after the Bending Spoons aquisition.
I vibe coded my own harvest or toggl app in one weekend.
Fable built me a version that covered all of my needs in an hour. (Clients pay me via ACH/wire - maybe it would have taken a day if I needed a stripe integration?). Runs on localhost with SQLite - incredibly basic, but I enjoy it twice as much as Harvest.
Do you have a repo you could point people too? That would be interesting to see.
Can you not 1-shot prompt an agent to build an equivalent piece of software for yourself?
I believe your question answers itself, if you could do this then people would[2]. The next question might be 'What prevents the current LLM based coding tools from being able to do this?"
Now there has always been software that people have used that they "could" write for themselves but they didn't because they didn't want to maintain it. Seems like things that are like that would be good candidates for this right?
But this question shares at least a spiritual geneology with the response to some new startup code that does some thing, and the comment here is "Gee, I can't believe that got this funded, I could write that in a weekend."
The problem is, there are many things that are easy to describe but difficult to pin down, and reality is complicated. So many times I've gone into a project thinking it would be "easy" only to start uncovering layer after layer that are much more involved than I expected.
So no, you cannot '1-shot prompt' pretty much any significant software, and that goes doubly true if it affects your life in any potentially devastating way.
One of the things I'm constantly amused by[1] is the Solitaire program on iOS that is clearly being vibe coded these days because it comes up with daily challenges like 'win a hand in -2739551 moves'. Like no human would ever write code that screwed up basic math but LLMs do it all the time.
[1] Obviously I am easily amused :-). But it is sort of the 'pelican riding a bike' test where you can easily tell what a human would do vs what an LLM does.
[2] EDIT: another commentator says they vibe coded it in a weekend. That would support the hypothesis that the application could be at lease a small number of shots vibe coded. That said, I'd love to see that code and the prompts that led to it.
Go ahead and vibe code your payment system. What could go wrong?
Harvest is a stripe wrapper with time keeping. They are not suggesting to vibe code stripe.
I’ve found that while vibe coding can be effective, anything with complexity takes more than a 1-shot prompt to get there. The amount of work that comprises a single scrum story is about the right amount to do with a single prompt and even that requires some review and revision. So 1-shot your payment system, hell no, but writing a payment system without writing code is possible although I would think it still would take enough time that I wouldn’t choose that over a reasonably priced SaaS.
So many one shot opportunities and yet so few actual links
I've had a Harvest account for over a decade, and it was always admittedly a little bit too cheap. But two months ago, they changed the pricing based on what you gross, so essentially putting a flat tax on my earnings.
I knew it was going to be more expensive, but when I got the bill this month and saw the price was over $200 and the Bending Spoons LLC tagline at the bottom, that was the end.
Claude is backing up all of my data via the API and will be vibe coding my next invoice.
Meanwhile in October...
Berkshire Hathaway invests further $330M in Bending Spoons
https://investors.bendingspoons.com/newsroom/berkshire-hatha...
Wow, harvest still exists?
My first tech job ever pretended we were supposed to track our hours in harvest. Which of course no one did. So long ago.
You could try plinx