3 ms·

Fuck investors. Set your product’s price and let the customers run away or stay. Then see if it’s enough. Then and only then, if it’s not enough, drop the produ

Fuck investors. Set your product’s price and let the customers run away or stay. Then see if it’s enough. Then and only then, if it’s not enough, drop the product. Or, if you really want to keep working on it even if it isn’t sustainable, find investors. In the meantime, kill off your cloud. Purchase or rent bare metal servers and stop paying per minute for shit. You can get 100+ of cpu cores, terabytes of ram, and disk space for less than a few hundred USD a month. Self-host the things you need (garage for s3, longhorn for persistent storage, harbor for Docker images, Loft for k8s management, etc).

octoberfranklin 4y ago
I must concur with this excellent advice. However I recommend undertaking it in reverse order: > In the meantime, kill off your cloud. This will massively reduce the price you charge your customers. Do it first.
- bradwood 4y ago
  If you are running k8s all day long, then sure, but if you run a slim serverless stack in AWS with dynamodb, lambda, SNS/SQS, eventbridge etc, you can stand up a substantially complex solution for next to nothing. Source: I am doing it right now.
presentation 4y ago
That depends on where they're spending money. If their cloud costs are 1/10 their labor costs, then doing that won't really move the needle.
- withinboredom 4y ago
  A ten percent reduction in costs is huge. In most businesses, you can only move the needle by fractions of a percent (without ruining people’s job security), so being able to get even a whole percent is nice, more than a whole percent is Must Do ASAP. Most of the time (in life, business, and marriage), you succeed/fail in “death by 1000 cuts”, not in huge gains or losses.
  - presentation 4y ago
    That's true for a more mature business that already has monetization figured out and is trying to optimize its operation, but sounds like OP is running an early stage startup that isn't even monetized yet. If they can't find a way to make solid margins on his operating costs, then it's a real question if the opportunity cost of being distracted by stuff like how they're hosted is worthwhile vs doing work to materially make their product better/more marketable.
    - withinboredom 4y ago
      OP has 100k potential customers. If they fail to monetize that, there’s no point in continuing. Beyond that, driving costs down before marketing anything and achieving sustainability should be priority number one. That’s the opportunity, and the only one that matters unless they get some other injection of cash. Trying to market on a short runway is suicide.

octoberfranklin 4y ago

I must concur with this excellent advice. However I recommend undertaking it in reverse order: > In the meantime, kill off your cloud. This will massively reduce the price you charge your customers. Do it first.

bradwood 4y ago
If you are running k8s all day long, then sure, but if you run a slim serverless stack in AWS with dynamodb, lambda, SNS/SQS, eventbridge etc, you can stand up a substantially complex solution for next to nothing. Source: I am doing it right now.

presentation 4y ago

That depends on where they're spending money. If their cloud costs are 1/10 their labor costs, then doing that won't really move the needle.

withinboredom 4y ago
A ten percent reduction in costs is huge. In most businesses, you can only move the needle by fractions of a percent (without ruining people’s job security), so being able to get even a whole percent is nice, more than a whole percent is Must Do ASAP. Most of the time (in life, business, and marriage), you succeed/fail in “death by 1000 cuts”, not in huge gains or losses.
- presentation 4y ago
  That's true for a more mature business that already has monetization figured out and is trying to optimize its operation, but sounds like OP is running an early stage startup that isn't even monetized yet. If they can't find a way to make solid margins on his operating costs, then it's a real question if the opportunity cost of being distracted by stuff like how they're hosted is worthwhile vs doing work to materially make their product better/more marketable.
  - withinboredom 4y ago
    OP has 100k potential customers. If they fail to monetize that, there’s no point in continuing. Beyond that, driving costs down before marketing anything and achieving sustainability should be priority number one. That’s the opportunity, and the only one that matters unless they get some other injection of cash. Trying to market on a short runway is suicide.