3 ms·

I used to write sneaker bots “professionally” (was pretty much my full-time job) back in college. Did it for a variety of reasons. The money in reselling is de

I used to write sneaker bots “professionally” (was pretty much my full-time job) back in college. Did it for a variety of reasons. The money in reselling is decent but you can usually make more by selling the software itself. I collected a few rare pairs and definitely found that fun. Also highly ego driven and there’s an entire “sneaker Twitter” community where people just want to show off. I was mainly into it for the software engineering side of things. You end up having to build really interesting/complex software. My infrastructure was basically half a dozen servers with a coordinator that would scrape websites all day long. Also had to learn how to exploit different features of websites to speed things up. Generally enjoyed it but the community is pretty toxic. Mostly driven by money and ego, no room for collaboration because it’s more profitable to keep things to yourself.

b_tterc_p 8y ago
Selling shovels to gold miners basically. Nice. Can you explain how you find people to sell software too and how realistic your value prop is? It doesn’t quite sound logical to me (and maybe it isn’t?). Your software finds margins of x, but surely only so many people can make use of it before it stops working for anyone.
- moate 8y ago
  I think the bots are being used to buy the sneakers from direct sales channels (Nike, Footlocker, primary market sources) and then you'd flip them on the secondary market. Basically the direct channels will sell the shoes at "reasonable" prices but well below what the secondary market is willing to support so you can walk out the door and immediately sell them for 2X+ what you paid. You see similar things in rare craft beer releases.
- kfichter 8y ago
  Mmmm you typically limit releases of your software for exactly this reason. Best model so far has been subscription based with a limited number of total subscription. I’ve seen it generate a huge amount of revenue (upwards of 300k annually). It’s in everyone’s best interest to keep the total number of bots in circulation relatively low. Helps that genuinely good software is difficult to write and most people doing that work are young/inexperienced - huge advantage to anyone with a strong softeng background.
ronyfadel 8y ago
Hey kfichter, care to share what your stack looked like? Did you use any public libraries (e.g. BeautifulSoup)?

b_tterc_p 8y ago

Selling shovels to gold miners basically. Nice. Can you explain how you find people to sell software too and how realistic your value prop is? It doesn’t quite sound logical to me (and maybe it isn’t?). Your software finds margins of x, but surely only so many people can make use of it before it stops working for anyone.

moate 8y ago
I think the bots are being used to buy the sneakers from direct sales channels (Nike, Footlocker, primary market sources) and then you'd flip them on the secondary market. Basically the direct channels will sell the shoes at "reasonable" prices but well below what the secondary market is willing to support so you can walk out the door and immediately sell them for 2X+ what you paid. You see similar things in rare craft beer releases.
kfichter 8y ago
Mmmm you typically limit releases of your software for exactly this reason. Best model so far has been subscription based with a limited number of total subscription. I’ve seen it generate a huge amount of revenue (upwards of 300k annually). It’s in everyone’s best interest to keep the total number of bots in circulation relatively low. Helps that genuinely good software is difficult to write and most people doing that work are young/inexperienced - huge advantage to anyone with a strong softeng background.

ronyfadel 8y ago

Hey kfichter, care to share what your stack looked like? Did you use any public libraries (e.g. BeautifulSoup)?