Card image cap

Should You Code or Use a Crypto Trading Bot?

Cryptocurrency trading bots are the big winners in crypto markets. Their quickness and lack of emotions make them better than human traders on all points.

A simple turtle strategy can beat most manual traders easily (backtest results are in the link). There’s no doubt you can make much there.
Anyone can come up with a basic Python trading bot and have the same access to markets than big hedge funds.

There are even free open-source crypto trading bots for you to get started!

On top of that, non-coders can benefit from platforms like Kaktana to create their trading bot, without any coding knowledge.
Objectivity notice: you’re on Kaktana’s blog

Crypto trading bots are on the rise right now. To get started, should you use a pre-made bot or code your own?

We’re going to compare both solutions (as objectively as possible, promised) in terms of time and money needed. We’ll also the huge benefits particular to each approach.

Purely To Make Money

No matter what all people just “interested in it just by curiosity” say, the main reasons to run a cryptocurrency trading bot are pecuniary.
Seeing the rentability of such bots, the running costs of a crypto trading bot are marginal. A 2% trade on a small $1000 account would compensate for the price of running most crypto trading bots.

Human Time Costs

Your time is expensive, and your life is very, very short.
Coding a trading bot takes hours. You’ll need to code your live trading algorithm, hook it up to the exchange you want to use, ensure it will be profitable by running backtests (doing those fast is hard, Python is 400x slower than C++ and you don’t want to wait a week for a 1 year backtest to finish).

The slightest backtesting algorithm error and your godly-profitable strategy reveals to be unprofitable.

At a 25$ hourly rate (I bet you’re more expensive than that, it’s only for the sake of argument), spending 100 hours coding it will cost you 2500$.
That’s 100 months or 434 weeks of Kaktana subscription. (I repeat, this article is as objective as possible!)

Recurring Costs

To run your trading algorithms, you’ll need a server. To rent a 2GB, 2vCPU VPS on DigitalOcean, you’ll need to pay 15$/month. That’s without counting the time required to keep it up to date with changing APIs.

Why is just using one so cheap then?

Because you’re not alone, having a community of users all paying for developer time allows you to have advanced algorithms at your service. Would you code your own Google after all?
For instance, the not-python-doable super-fast backtesting service is the result of countless coding hours that are profitable only because they’re used a lot – economies of scale.

To Become a Better Trader

Backtesting your manual strategy might save you a few headaches.
If you want to become a better trader and learn how the big players are beating you on the markets then, coding your trading algorithm will make you understand many things about trading.

You'll be forced to learn about how the order matching engine works, how market-making and arbitrage bots are working. You'll understand who makes the markets and what makes it move.

Running backtests will show you what strategies would have worked and what would have not.

Even if using a pre-made bot might give you some of the lessons you need to become a better trader, coding your crypto trading bot will provide you with way more pieces of information on the inner workings of the markets.


If you want to make money easily, go with a pre-made bot! They’re super cheap for what they do and are an awesome tool to have.
If you want to learn more about the inner workings of the markets, coding your own bot might be a better option.
I advise you to try both, and make your own opinion!

Trade while you sleep

Skyrocket your returns on the first 100% configurable crypto trading bot.

No PhD required