With different adjectives = "good", "stable", and "supported" are three fairly effective ones.
Ask HN: How do you compete with “free”?
11–20 of 21 posts
Re: Ask HN: How do you compete with “free”?
#12I simply refuse to play their game (free/race to the bottom). This isn't always a viable strategy (see stable diffusion vs OpenAI), but when I find myself in one of these situations I shoot for quality instead. The customers will pay, but I will do everything I can to give them a higher quality experience, whatever that entails.
what are your takeaways from stable diffusion vs. Open I?
Re: Ask HN: How do you compete with “free”?
#13I'm not a "founder" in any real sense of the word but my feeling is that it depends on what market you're targeting. If your goal is to become a billion dollar unicorn, then yeah it's going to be tough to beat free because you're going to need a huge number of users. On the other hand if you're OK with building a small sustainable business to support yourself, family and maybe a small number of employees then I think…
definitely makes sense. even if you intend to become a paid product, it's hard to figure out the right pricing in the early days.
Re: Ask HN: How do you compete with “free”?
#14This is a bit business-model dependent. If you don't intend for your product to be free for the rest of its lifetime, don't launch it as free.
yeah, i generally agree with this advice. Although testing pricing is its own challenge, so can be hard to validate demand + pricing in the same launch.
Re: Ask HN: How do you compete with “free”?
#15I'm not a "founder" in any real sense of the word but my feeling is that it depends on what market you're targeting. If your goal is to become a billion dollar unicorn, then yeah it's going to be tough to beat free because you're going to need a huge number of users. On the other hand if you're OK with building a small sustainable business to support yourself, family and maybe a small number of employees then I think…
If ever there were a premature optimization... No one's calibrating their prototype roll-out according to how big or small they think they're going to be. That'd be a bit like turning down a soap commercial because it'll be embarrassing when you join the MCU, although to be fair, an unknown Gwyneth Paltrow did turn down "Cool as Ice" (1991).
> No one's calibrating their prototype roll-out according to how big or small they think they're going to be.
I don't think that statement is true at all. My impression (mostly from being a long-time reader of HN) is that VC's are primarily/only interested in investing in startups that have some chance of becoming billion dollar unicorns, so if you're looking to get VC funds you pretty much have to set your goals at that level. Some little web app with 10000 paying customers will never get you there, but it might be enough to make a decent living.
Re: Ask HN: How do you compete with “free”?
#16I find it is very rare that I actually complete a free trial, particularly for something I use for software development at work. (1) It takes a lot of effort, maybe one day to one week, to really evaluate a product and know if it is worth it. (2) That evaluation will feel like wasted time if, at the end, I find the product isn't worthwhile (3) It will also feel like wasted time if I decide I like it but my employer d…
If it takes you 40 work hours to evaluate a product then it must be a tool that is central to your workflow, but in that case you're going to have to evaluate it somehow whether, it's free, free trial or paid.
Yes, and that has all sorts of implications for why marketing can be so hard.
Imagine that somebody expects to evaluate 4 products, either because the first 3 won't make the cut, or because they won't feel like they did enough "due diligence".
It may be a safer bet to develop something in house than to spend a lot of time evaluating alternatives. All of the alternative products have a lot more development time in them, but odds are you don't need all the features those products have, UI refinement, etc. You don't have to worry that a vendor will be bought by Google.
Re: Ask HN: How do you compete with “free”?
#17Earlier quoted context omitted.
If it takes you 40 work hours to evaluate a product then it must be a tool that is central to your workflow, but in that case you're going to have to evaluate it somehow whether, it's free, free trial or paid.
Yes, and that has all sorts of implications for why marketing can be so hard. Imagine that somebody expects to evaluate 4 products, either because the first 3 won't make the cut, or because they won't feel like they did enough "due diligence". It may be a safer bet to develop something in house than to spend a lot of time evaluating alternatives. All of the alternative products have a lot more development time in the…
Ah, yes, I see what you mean. That's probably one reason software developers (at least experienced ones) tend not to make good customers.
Re: Ask HN: How do you compete with “free”?
#18Nothing beats free. Stallman wrote a great essay on this called "Who does that server really serve?", decrying the redundancy and ultimate futility of building software-as-a-service. You will always be at-risk of being undercut by a free alternative. Selling software is an anti-pattern of computing, you have every right to be afraid of cheaper/better products wrecking your shop.
https://www.gnu.org/philosophy/who-does-that-server-really-s...
Re: Ask HN: How do you compete with “free”?
#19The only thing that beats 'free' is 'quality'.
Your product has to be so good that people will clamour to buy it.
If your product's quality is 'meh', 'free' will beat that every time.
Re: Ask HN: How do you compete with “free”?
#20I simply refuse to play their game (free/race to the bottom). This isn't always a viable strategy (see stable diffusion vs OpenAI), but when I find myself in one of these situations I shoot for quality instead. The customers will pay, but I will do everything I can to give them a higher quality experience, whatever that entails.
what are your takeaways from stable diffusion vs. Open I?
In the case of Stable Diffusion vs OpenAI, a model was released to the public for anyone on earth to experiment with, creating a diverse ecosystem of ideas. There are thousands of use cases now that are being tested live on the open market -- a number that any single company would find incredibly difficult to compete with. Most ideas won't gain any footing, but some will. By the time those ideas gain popularity, OpenAI will only be left to copy them.
This is specifically with regards to Stable Diffusion vs DALL-E. I think OpenAI has plenty of breathing room when it comes to innovation in the broader spectrum of new AI applications. I think there might be some interesting discussions here regarding when it's a good idea to provide a cheap and available version of your offering vs when to provide a quality offering. Both definitely have their place in the market and can result in superior products, but right now to me it's not clear where that distinction should be drawn.