Agree overall with above comments direction: by default be very careful when considering cloud mining.
I just wanted to add a bit of nuance:
Imagine you invested in a mining farm: land, warehouse, shelves, transformers. Your primary concern is to fill up 100% of your capacity as quickly as possible.
You have three options:
(1) Raise more capital and buy machines yourself => self-mining.
(2) Find clients who bring their hardware => hosted mining (or "mining hotels").
(3) Sell "virtual" machines, collect money, and buy machines => cloud mining.
Some operators might use a mix of all three options.
Cloud mining is administratively simpler than (2) since contracts are not based on specific units (which can fail or underperform), but rather on a theoretical hashrate. Some sell it as if new model "ASIC" while on the real shelves there are completely different models (who can check?). Also (3) allows to sell a "fraction" of a miner, making it more retail-friendly. The problem arises when the underlying hardware starts to fail. For instance, if you promised a certain hash rate to your cloud clients but the machines you bought are faulty. Many other issues can occur if risks are not properly managed, ie when the actual hashrate from operations is behind promised hashrate.
In theory, cloud mining, if done correctly, can exist (IMHO). However, it is challenging, which is why those who have resources prefer to go for option (1) or, at most, option (2).
To answer your question about why someone would choose cloud mining instead of mining themselves:
- In option (1), earnings are volatile since revenue is in BTC and operating expenses (OPEX) are in fiat currency;
- In options (2) and (3), earnings are fixed since contracts are usually in fiat currency per TH/s or kilowatt-hour kWh.
I hope we can add a section to our calculator (
https://aidala.uk/dashboard) that allows users to visualize and compare earnings from mining versus earnings from hosting over a selected period. This comparison can often provide very interesting insights.