I can totally see a world where a small team of PM + engineers can recreate internal business tools, you need to run the numbers to see if paying these people is significant cost savings relative to paying for the tool
The reality for most enterprise apps is that customers are using a small fraction of all available functionality and are paying for much more than the featureset they are using
So if a business can identify this featureset and give a team empowerment to build and maintain this featureset, it could work
Running a business is an optimization problem though - if you are spending resources on internal tooling, and people are more expensive than money, are you taking away from those resources being spent on things that drive revenue?
Plus, this falls apart when you start dealing with anything involving compliance. Part of paying for an app is you offload the risk of maintaining compliance to that app
Once you need to get legal/compliance/accounting involved, it's unlikely to be worth the cost anymore, so it depends on the data being processed by the SaaS
Even if cloning a saas is one click away (it isn’t now, but let’s imagine), you would still need to run it in production. If you care a bit about it, you probably want to keep it secure and up to date, and keep backups, and add monitoring/alerting. While AI can help with all of this, is still some work to be done by a (knowledgeable) human.
I doubt a large percentage of any user base is willing to do all of that
You could clone easily before AI. When learning a language, "Build a twitter clone" is a tutorial project.
That doesn't give you the business. It give you no users, no market, no branding. You don't have the operations set up to run at scale, nor the customer base to even need it. Not to mention the leadership to build out a vision. "Clone someone else" is actually a bit of a red flag in terms of vision.
If anything, the ease of launching an MVP should prove what many folks already know - coding is not the bottleneck that prevents business success. It is the tool that allows you to step up to the starting line.
An app that you now have to maintain. The point of spending $5 a month is not to have to deal with that hassle. It makes zero sense to do it to save money. SaaS customers really aren’t in the business of SaaS cloning.
The only people that might look into cloning an existing business are competitors, and by virtue of being competitors they already were a danger to the business.
Companies are happy to pay for SaaS products that are being run by a team of people who are experts in that domain and are also responsible for anything that goes wrong. Cloning something in house, no matter how it's done, doesn't bring you the subject matter experts and it shifts the responsibility/blame internally.
Those aren't serious users anyway, business users who are focusing on their business don't have the time to clone and maintain their own software. People who claim to successfully clone and replace SaaS are never good customers anyway, they will either churn after free trial has ended, or post on reddit searching for free alternative
That's true that businesses don't want to waste their time by maintaining that software but it's also true that a bigger number of users will start cloning in the future especially fable, it's very good at this thing.
Start? We're well past that. If your SaaS is just a fancy wrapper on some CRUD it's already been cloned by AI a dozen times. Even if it has some level of sophistication, unless there is a trove of data that only you have access to, it's already been cloned.
This doesn’t work now unfortunately as described. Cloning is multi-months effort and it’s hard even when you know what you’re doing. If you want to have a sellable clone, it’s even harder challenge as users wouldn’t tolerate sloppy work when better alternatives exist.
Is there an actual SaaSpocalypse, or just people saying it is going to happen? Because I haven’t seen it with my eyes, and it’s not like business owners are flooding to HN to complain how AI has killed their business.
What's going away is really shitty business models out of spite. I'd rather not build it myself, I'd rather just pay someone else to do it, but if the vendor is being a bunch of shitheads, I'll roll up my sleeves and replace them.
I make online software so advanced nobody knows it was even possible to begin with. So I think I'm good, Ai can try all day long.
And I'm not even joking.
I can totally see a world where a small team of PM + engineers can recreate internal business tools, you need to run the numbers to see if paying these people is significant cost savings relative to paying for the tool
The reality for most enterprise apps is that customers are using a small fraction of all available functionality and are paying for much more than the featureset they are using
So if a business can identify this featureset and give a team empowerment to build and maintain this featureset, it could work
Running a business is an optimization problem though - if you are spending resources on internal tooling, and people are more expensive than money, are you taking away from those resources being spent on things that drive revenue?
Plus, this falls apart when you start dealing with anything involving compliance. Part of paying for an app is you offload the risk of maintaining compliance to that app
Once you need to get legal/compliance/accounting involved, it's unlikely to be worth the cost anymore, so it depends on the data being processed by the SaaS
Even if cloning a saas is one click away (it isn’t now, but let’s imagine), you would still need to run it in production. If you care a bit about it, you probably want to keep it secure and up to date, and keep backups, and add monitoring/alerting. While AI can help with all of this, is still some work to be done by a (knowledgeable) human.
I doubt a large percentage of any user base is willing to do all of that
I have replaced software that costs a pretty penny per month.
This was possible since opus 4.6... Im probably an outlier though and I don't actually think the practice is very widespread.
Customer service is a moat that can justifies premium pricing.
Premium pricing segments the market for customers who see value in what you do.
And premium pricing allows you to afford customer service.
You could clone easily before AI. When learning a language, "Build a twitter clone" is a tutorial project.
That doesn't give you the business. It give you no users, no market, no branding. You don't have the operations set up to run at scale, nor the customer base to even need it. Not to mention the leadership to build out a vision. "Clone someone else" is actually a bit of a red flag in terms of vision.
If anything, the ease of launching an MVP should prove what many folks already know - coding is not the bottleneck that prevents business success. It is the tool that allows you to step up to the starting line.
It's not about cloning and selling your own app. It's about using app that you are paying for, by cloning it using AI.
An app that you now have to maintain. The point of spending $5 a month is not to have to deal with that hassle. It makes zero sense to do it to save money. SaaS customers really aren’t in the business of SaaS cloning.
The only people that might look into cloning an existing business are competitors, and by virtue of being competitors they already were a danger to the business.
Companies are happy to pay for SaaS products that are being run by a team of people who are experts in that domain and are also responsible for anything that goes wrong. Cloning something in house, no matter how it's done, doesn't bring you the subject matter experts and it shifts the responsibility/blame internally.
Those aren't serious users anyway, business users who are focusing on their business don't have the time to clone and maintain their own software. People who claim to successfully clone and replace SaaS are never good customers anyway, they will either churn after free trial has ended, or post on reddit searching for free alternative
That's true that businesses don't want to waste their time by maintaining that software but it's also true that a bigger number of users will start cloning in the future especially fable, it's very good at this thing.
Start? We're well past that. If your SaaS is just a fancy wrapper on some CRUD it's already been cloned by AI a dozen times. Even if it has some level of sophistication, unless there is a trove of data that only you have access to, it's already been cloned.
Have to train on top of the models or have some sauce not publically available.
Hard part has never been building software anyway ita getting clients and building trust with them.
You did not get my point. My point was about if user just start cloning apps they are already using.
This doesn’t work now unfortunately as described. Cloning is multi-months effort and it’s hard even when you know what you’re doing. If you want to have a sellable clone, it’s even harder challenge as users wouldn’t tolerate sloppy work when better alternatives exist.
You just discovered the SaaSpocalypse that began around January this year.
Is there an actual SaaSpocalypse, or just people saying it is going to happen? Because I haven’t seen it with my eyes, and it’s not like business owners are flooding to HN to complain how AI has killed their business.
What's going away is really shitty business models out of spite. I'd rather not build it myself, I'd rather just pay someone else to do it, but if the vendor is being a bunch of shitheads, I'll roll up my sleeves and replace them.
This is why investors ask what your moat is