Robot representing automated market maker
Cryptocurrency Finance

What is an Automated Market Maker (AMM)?

Complicated term right? Well, the job it does isn’t easy either. So buckle up and catch hold of the nearest chair. As always, I won’t be using any fancy jargons to explain this to you. Apart from that, once you wrap your head around this technology, it would feel no less than magic. A part of the complexity in the blockchain world is also because our mental models aren’t equipped to come to terms with these new developments in Decentralized Finance (DeFi) world. The very fact that a piece of code is replacing an entire industry is just so overwhelming. Jumping right into it.

Centralized Exchanges and Order Book Model:

I am sure you must have traded stocks or crypto by now on your favorite exchanges like Zerodha, CoinDCX. These exchanges are centralized in nature. This means that companies like Zerodha and CoinDCX store your stocks or cryptos with them.

These type of exchanges work on an order book model. That means if you plan to buy a specific cryptocurrency, say BAT, you place an order at a specific price. The order book matches the price quoted by you to all the bids raised by the sellers. Whenever there is a match, your order gets executed.

Supply and demand is automatically taken care of as the users themselves are quoting the best price of their assets. In other words, higher the demand, higher the price of an asset.

Decentralized Exchanges or DEX and Automatic Market Maker (AMM) Model:

Now the entire narrative of blockchain is developed around decentralization. But then what is the point of using a centralized exchange? Agreed that you are getting rid of governments snooping into your account, but what about the exchange itself? It is like placing trust in a different entity if not the government.

So the question that arises is that can we do it without anyone’s involvement in a truly decentralized and trustless way? Well, the answer is yes. Enter Automatic Market Maker or AMM in short.

AMM helps individuals exchange one cryptocurrency with the other using smart contracts and liquidity pools. Too much to digest? Let’s take it from the top:

How do AMMs work?

Imagine it’s year 1850. You are a farmer, living peacefully in a village. All of you grow potatoes for living. You can make fries, mashed potato etc. from that potato but now you are kind of sick. Your spouse is demanding something else to eat.

At the same time, a trader comes from the other side of the world. He mentions that he knows of a village where people are sick of eating apples every day. He offered a deal where he could facilitate a deal to trade 50,000 apples against 50,000 potatoes. Since that’s exactly you were looking for, all your villagers agree.

The trader continues: he has a magic warehouse where all these apples and potatoes will be stored. Since there should be a balance between the value of apples and potatoes, the product of both these entities would be equal to 2.5bn (50,000 apples X 50,000 potatoes).

This way, if there are more potatoes than apples in the warehouse, each apple would cost more potatoes to buy. If this is clear, congratulations, you just learnt the basic principal of a constant product automated market maker.

It is based on the formula X * Y= k. In our case, X is apples, Y is potatoes and k is always a constant at 2.5bn.

In the beginning the price of apples is equal to the price of potatoes because each of them is kept at a Re. 1 mark. However, as the trading (exchange) begins one might become more valuable than the other thus pushing it’s price more than a dollar.

Calculations in Automated Market Maker:

Let’s continue with our example. Our magical warehouse has began trading with 50,000 apples and 50,000 potatoes. Comes along a farmer with his produce of 7,000 potatoes and wants to exchange it with apples. So let’s calculate how many apples would he get in return:

(Mathematics Time)

Number of potatoes in the magic warehouse: 57,000 (50K+7K)
Number of Apples in the magic warehouse: 50,000

Using ‘X’ (apple) * ‘Y’ (potato) = k (2.5bn)

X * 57,000 = 2,500,000,000

This means that number of Apples that should be there in the warehouse to maintain a balance should be:

X = 2,500,000,000/57,000 = 43,859

So how many apples should he get in return?

50,000- 43,859 = 6140 Apples

Since the number of potatoes in the warehouse was more than number of apples, farmer could fetch only 6140 apples for 7000 potatoes.

Does our constant product AMM hold true? Yes it does (43,859 * 57,000 = 2.5bn)

Rupee Cost of Potatoes:

Since we started with the cost of Re. 1 per potato, the cost would be constant at Rs. 50,000 for the entire lot. Now that 57,000 potatoes could be bought in Rs. 50,000 the cost of each potato would be 50,000/57,000 = 0.87p.

On a similar note, the cost of apple would be 50000/43859 = Rs. 1.14

Its basic supply and demand you see? Since someone brought more potatoes to the magic warehouse, the price of potato fell under a rupee and apple observed a contrary effect.

Example 2:

This time another farmer comes along and tries to add 10,000 more potatoes to the magic warehouse. Also, let’s call this magic warehouse what it actually is: Liquidity pool. Let’s quickly run through the numbers:

Number of potatoes in the liquidity pool: 67,000 (57K+10K)
Number of Apples in the liquidity pool: 43,859

Using X(apples) * Y(potatoes) = K (2.5bn)

X * 67,000 = 2,500,000,000

The number of apples that should be in the Liquidity pool = 37,313

Apples he would get in return: 43,859 – 37,313 = 6,546

Can you calculate the rupee cost of each Apple and potato now? Drop your answers in the comments section.

I hope it’s a lot clearer now. The rupee value of potatoes and apples would always be equal to 50,000 each in the liquidity pool. The pool will always maintain the ratio of value to 50:50.

The Real Liquidity Pool:

Jotting down a couple of differences from the above example when you are dealing with real liquidiuty pools:

  • Unlike apples and potatoes, real liquidity pools can have millions worth of cryptocurrencies locked in it.
  • Also, more the value locked in a liquidity pool, stable are the prices. When you are dealing with 5,000,000 apples and potatoes the impact of 5,000 may not be much right?

Who is Adding Value to the Liquidity Pool?

In real life, who is bringing those initial 50,000 apples and potatoes to the magic warehouse to get started? Well, its investors like you and me. And why would one do that? Because they are rewarded with a fee during each transaction. Even if this fee <1% but the numbers add up really fast.


Pay yourself on the back. It took me a while before grasping the concept of AMMs and articulating in a blog. Also, please don’t judge me next time when I refer to any such technology as magic. Hopefully now, you know why!

What do you think about AMMs?

Let me know in the comments section below. Hit me up on FacebookTwitter or Instagram. If you find this article interesting, please consider sharing it on social media using the links below:

Until Next Time. . .

For our beloved “non readers”, I also do quick carousels on these topics over Instagram. Come join the fun. Hit me up here.

A techno manager by profession and a hardcore geek at heart. I love to poke my nose into tasks where other usually gave up on. My hobbies include, reading about latest trends in tech industry, playing guitar and yes, memes!

One thought on “What is an Automated Market Maker (AMM)?

Leave a Reply