I feel like you're somewhat viewing this from the wrong direction. Let me paraphrase to make a point.
I went into my supermarket yesterday. They have thousands of different products on the shelves. Worse, the products run out and I'm expected to come back next month and buy more. Is this sustainable? Do I have to buy one of everything every month?
If you're drowning in new SaaS products then you're spending too much time looking at them. You're not expected to know most of them exist, much less subscribe.
SaaS is no different to anything else. They are businesses selling products to customers. If you're drowning in those wait till I tell you how many shops want to sell you bread... (or how many species of bread there are.)
There's no requirement to maintain a subscription. Each individual should be making this decision for themselves based on the value they get from a service vs the cost.
The broad "startups" in question are just businesses trying to sell a product. Not everyone is going to sign up, and not of those who do, most won't maintain their subscription forever.
> Is this really sustainable in the long run?
For who? For the startups? Maybe, depends on the value provided. For the users? Maybe, depends on if they feel they're getting value.
I'm kind of confused as to what the question is really asking.
Most people I know have a handful (making exception for things that have always been "subscription", like insurances): a streaming service or two, iCloud/Google One, and maybe a budgeting app or something more niche for a hobby of theirs. I'd argue that none of those are necessary, except maybe the cloud storage to retain information and backup their devices. User's can just vote with their wallets.
As for it being sustainable for a startup: who cares - a good product will be worth paying for an it'll survive.
(Hope the tone doesn't come off as rude, I'm just trying to shoot for brevity)
$10/month is a lot. You have to be solving a very critical need for me to sign up for that. $3-$5, not that big of a deal, $8 I'll think about it. $10 better be a major benefit
They're basically suggesting logging into a product with an OpenAI account and paying OpenAI directly for the product's AI usage. Presumably the product would get some cut of OpenAI's revenue.
Like AI is the gas that powers the AI product. OpenAI sells the gas, the product gets commission.
That was my understanding at least, since I've had the same thought.
I think you might be on the right track but the guy says that in his dream world the user only pays once. That’s the same thing that’s happening now. I neither know nor care how much OpenAIWrapperApp is paying for its API usage. I just get billed by OpenAIWrapperApp for the amount they charge.
Not the OP but I personally would be interested in people's perspective on the # of $10 subscriptions the average person can sustain.
Using Fermi estimation, is it: 1, 10, 100? I would say 1 because I don't believe businesses can create enough value (save them time or money) for the average person to justify spending $100/month on subscriptions.
Another thought is how CISO's found themselves with 20 cyber subscriptions and found themselves wanting to consolidated down. So I would make this the absolute upper limit.
So I have somewhere between 1-20 with the average across the consumer population being somewhere closer to 1 than 20.
The implication being that there is a lot of pain coming for the long tail of startups trying to make it with a $10/month subscription model.
I feel like you're somewhat viewing this from the wrong direction. Let me paraphrase to make a point.
I went into my supermarket yesterday. They have thousands of different products on the shelves. Worse, the products run out and I'm expected to come back next month and buy more. Is this sustainable? Do I have to buy one of everything every month?
If you're drowning in new SaaS products then you're spending too much time looking at them. You're not expected to know most of them exist, much less subscribe.
SaaS is no different to anything else. They are businesses selling products to customers. If you're drowning in those wait till I tell you how many shops want to sell you bread... (or how many species of bread there are.)
There's no requirement to maintain a subscription. Each individual should be making this decision for themselves based on the value they get from a service vs the cost.
The broad "startups" in question are just businesses trying to sell a product. Not everyone is going to sign up, and not of those who do, most won't maintain their subscription forever.
> Is this really sustainable in the long run?
For who? For the startups? Maybe, depends on the value provided. For the users? Maybe, depends on if they feel they're getting value.
I'm kind of confused as to what the question is really asking.
Most people I know have a handful (making exception for things that have always been "subscription", like insurances): a streaming service or two, iCloud/Google One, and maybe a budgeting app or something more niche for a hobby of theirs. I'd argue that none of those are necessary, except maybe the cloud storage to retain information and backup their devices. User's can just vote with their wallets.
As for it being sustainable for a startup: who cares - a good product will be worth paying for an it'll survive.
(Hope the tone doesn't come off as rude, I'm just trying to shoot for brevity)
$10/month is a lot. You have to be solving a very critical need for me to sign up for that. $3-$5, not that big of a deal, $8 I'll think about it. $10 better be a major benefit
My expectation is that the LLM providers will also start to bill for the AI startups through their APIs
They already do?
Yeah? Which one? You might not have understood what I meant
As far as I know, OpenAI does charge for API usage.
If that's not what you meant, clarify?
I meant OpenAI billing for me in the same charge, an external app integrating their API. So the user only pay once
I still don't follow you at all.
The way I understand it is that:
OpenAI bills people who use their API to build apps or whatever.
OpenAI also bills people who use its non-API products (like ChatGPT).
Am I missing something here or what is it that you dislike about this?
They're basically suggesting logging into a product with an OpenAI account and paying OpenAI directly for the product's AI usage. Presumably the product would get some cut of OpenAI's revenue.
Like AI is the gas that powers the AI product. OpenAI sells the gas, the product gets commission.
That was my understanding at least, since I've had the same thought.
I think you might be on the right track but the guy says that in his dream world the user only pays once. That’s the same thing that’s happening now. I neither know nor care how much OpenAIWrapperApp is paying for its API usage. I just get billed by OpenAIWrapperApp for the amount they charge.
Most charge way more than 10/month. Most startups also die.
They don't think about it that way. They only think about you signing up for theirs.
Are you concerned that startups will not be able to continue with these cheap subscriptions, or because you have to pay $10/product?
Not the OP but I personally would be interested in people's perspective on the # of $10 subscriptions the average person can sustain.
Using Fermi estimation, is it: 1, 10, 100? I would say 1 because I don't believe businesses can create enough value (save them time or money) for the average person to justify spending $100/month on subscriptions.
Another thought is how CISO's found themselves with 20 cyber subscriptions and found themselves wanting to consolidated down. So I would make this the absolute upper limit.
So I have somewhere between 1-20 with the average across the consumer population being somewhere closer to 1 than 20.
The implication being that there is a lot of pain coming for the long tail of startups trying to make it with a $10/month subscription model.
But hey, we're in a bubble so ¯\_(ツ)_/¯