DeFi Concepts You Need to Know for 2026 (Part 1)
Jan 9, 2026
Are you exploring decentralized finance (DeFi) and overwhelmed with all the new concepts and jargon? Well we’ve written this plain-english guide to take you from the most basic concepts through to some of the most advanced ideas as quickly as possible. Let’s dive in:
DeFi 101: Smart Contracts, Gas & Tokens
If you’re already comfortable with the concepts below then feel free to skip this section.
Smart Contracts
Smart contracts (SCs) are the backbone of most blockchains. They underpin basically all functionality and are what DeFi is literally built upon. They were introduced into Ethereum by the core development team and became the chain’s main differentiator against the Bitcoin protocol (more on this in an upcoming post).
But what are they? Simply put, they’re computer programs that are stored on the blockchain. These smart contract blockchains, like Solana, Sui, Near etc., let you build anything, as they are (usually) Turing complete languages, which is a computer science term for being able to generate any output from any input, unlike Bitcoin which has a much more limited set of functionality.
You can think of blockchains as basically a global, decentralized computer. There are multiple computers all running at the same time, with the same software, with the goal of making their record of all events stay in sync. This means you can run a program and have all the computers on the network see that action and update their version of the blockchain to match the result of that program’s output. On Ethereum, or the Ethereum Virtual Machine (EVM), all smart contracts are publicly viewable too so any malicious code can be seen and independently audited.
Gas
For running these programs and updating this across all the nodes across the world, users are required to pay gas, denominated usually in the native chain currency (like ETH or SOL), as a way to pay the nodes for running the computation (and to prevent spam too!). The other fun part about SCs is that they can be read and executed by other smart contracts. Imagine having a computer program that calls other computer programs - that can be very powerful as you’ll see below.
Tokens
A token is just a smart contract that lists out all of the holders and their balances. Typically, most fungible assets are represented as ERC-20 tokens on Ethereum (EVM) or SPL-20 tokens on Solana. These labels are called standards and describe the inputs, outputs and functionality of the smart contract so they can all be mutually exchanged. These fungible token SCs outline information such as the token ticker, the initial supply, the total supply and includes functions to transfer balances and mint new tokens, for example.
Decentralized Exchanges (DEXs) and AMMs
Swapping (or trading) one asset for another is one of the earliest and main applications of DeFi. Uniswap is one of the oldest protocols in DeFi and its sole purpose is to allow a person to swap from one token into another. Fun fact, it was actually proposed by Ethereum’s co-founder Vitalik in a Reddit post and has since become one of the most successful projects in the space.
These systems are called decentralized exchanges (DEXs), and they work differently to their centralized counterparts (CEXs), like Coinbase and Binance. There is no central party (or parties) making these exchanges run, like via an order book for example, it’s all just a set of smart contracts running independently on the blockchain. Uniswap’s team (the one that oversees the protocol) does have some control over certain parameters, like revenue splits to the $UNI token, but there’s nothing that can prevent you from swapping if they wind down as an entity.
There are different algorithms that various DEXs use, like Uniswap’s model, which at it’s core uses the ratio of token A to token B as the price, versus Curve’s model where their algorithm is more stable at a specific point as to allow higher volumes of stablecoins to be swapped without impacting the price too much.
We’ll go into this in detail in a future post, but for now the main way these DEXs work is that liquidity providers (LPs), who are separate third parties to the DEX developer and the swappers, provide liquidity on these exchange contracts. They deposit their funds into these DEX contracts to allow other people to swap with them.
For example, If a user wants to swap from USDC to USDT, there will be a LP that provides some of both tokens into the DEX smart contract. As the swap takes place, their USDC balance goes up and the USDT balance goes down. For providing this ‘service’ they are rewarded with a swap fee (as configured by the DEX protocol itself).
Lending
In a similar fashion to traditional lending, there are products or protocols (we’ll use these terms interchangeably) that allow a user to put up collateral and get a loan against that collateral, but all completely on chain and without needing to deal with an actual person.
Note, lending markets are also different to collateralized debt position (CDPs) protocols which mint net-new tokens based on assets supplied as collateral. These have fallen out of popularity in the last few years in favor of conventional lending markets like Aave and Morpho, which we’ll be discussing here.
There are three main differences between lending in DeFi versus traditional markets:
DeFi lending is Overcollateralized
Unlike traditional lending, all current (heavily used) DeFi lending protocols require users to overcollateralise their positions so that the maximum loan to value ratio (LTV) is roughly between 80-90%, and is configured at the protocol level. This means that there isn’t a conventional way to get more than 200% exposure on your principal like in traditional finance, but there are workarounds to get close (coming up in part 2).
This would mean that for every $100 deposited, the borrower would get roughly $80 back out. This is the case because without any ability to know who you’re dealing with, as users are mostly pseudonymous, these protocols require overcollateralisation to pay for borrow fees and to prevent malicious users clogging the system with bad debt. Undercollateralised lending is still a holy grail area in DeFi and being explored by companies like 3Jane, which aim to create more of a credit history and profile around users.
Borrow Rates Are Set Globally
In the case of a protocol like Aave, their borrow rates are set by the DAO implementing an equation which is used to derive the borrow rates. Once the utilisation goes above the 80-90% threshold, it then becomes aggressively more costly to borrow. This is designed intentionally to encourage lending supply to expand as demand increases.
In the case of the next most popular protocol, Morpho, it is set on a market-by-market basis, where each market tries to target a utilization of 90% (borrowed assets / supplied assets). Morpho’s mechanism tries to target 90% utilisation, but it also actually adjusts the borrow rates in a non-linear manner every 5 days the longer the utilisation rates stay below the 90% target. There are many other lending protocols out there, but these are the biggest as of writing. We may write about this more in future posts.
Liquidations Are Automatic
Given the fact that lending protocols are used as core primitives, or building blocks, in other products in DeFi, timely liquidations are important to ensure bad debt doesn’t affect the broader ecosystem. These protocols implement real time liquidations by allowing anyone to monitor these positions and if the LTV threshold has been passed, a liquidation event can be independently initiated.
This can even occur without the borrower knowing because the lending positions and the debt risk metrics are all public and transparent on chain. In the case of Aave, they’ll liquidate your entire position, but Morpho is designed to only liquidate portions to keep you within the loan to value (LTV) target, also called the liquidation threshold value (LLTV).
Stablecoins (and Yield Bearing Stablecoins)
Given it’s such a hotly discussed topic, it’d be wrong of us not to cover it in the first post of this series. Stablecoins are actually a sub-category of real world assets (RWAs), which we’ll cover next, and are assets that in theory are pegged 1:1 with an asset in the ‘real world. Therefore not a crypto-native asset like Bitcoin, Ether, Solana, and so on. Typically, the most common stablecoins are US dollar (USD) based given their broader world-wide appeal, acceptance and perceived safety versus other fiat-based stablecoins, but there are those that exist for HKD, AUD and AED, to name a few.
Why stablecoins? For consumers it provides an on chain native way to exit from a risker position, like when you want to reduce your Bitcoin exposure, without having to off ramp onto an exchange (which can be clunky). For the issuers, it can be a massive revenue-driving opportunity as they can take a float or clip from the assets they’re managing. Some products opt to pay this yield to their users, but given the regulatory uncertainty (and better business opportunity) most stablecoin issuers keep that difference.
For example, Circle (CRCL, USDC) doesn’t pass on any yield to users, so Circle the company now keeps all the interest they earn on their fiat. These stablecoins also can generate yield independently from their issuers, or directly as part of the protocol (known as yield bearing stablecoins), which we’ll cover in part 2.
There used to be a wider variety of USD stablecoin designs, however over time the Darwinian evolution process has weeded out the weaker designs, like algorithmic stablecoins, which would use a fully speculative token as the backing of the stablecoin. The main stablecoin designs are:
Fiat collateralized: Where there is a 1:1 backing of dollars in a bank account and stablecoins minted.
Asset or NAV collateralized: Where a basket of instruments or assets are acquired and the dollar value of that basket is used to mint the same dollar amount of stablecoins. This could be commodities or securities, or even derivative positions, like in the case of Ethena.
Algorithmic: Where the underlying asset is a token with inflation controlled by a smart contract and where that supply modulates based on the current market price to keep the stablecoin pegged.
The only real designs that have long-term sustainability are fiat collateralized and asset/NAV collateralized, and of these fiat-collateralized tend to be the preferred one by institutions given their simplicity and basically zero risk. There was a major incident in 2022 where Terra (UST) suffered a massive de-peg which sent shockwaves in DeFi as their native backing token, LUNA, suffered a massive price drop and left the whole system to unravel until its demise.
Real World Assets
As mentioned above, RWAs are the result of representing an off chain, or real world item on chain via a token, like fiat, real estate, equities or even bonds. The process is simple: take an asset, literally anything, stick it in a bank or brokerage account controlled by a trusted party (like the project or their custody provider, like Anchorage) and then mint 1 token per 1 asset in that off chain account. Typically they don’t use ERC-20 token standards, but rather ERC-1400, 3643 and 7943 (in case you wanted to do some further research).
For example, if you wanted to tokenise the S&P500 ($SPX) you could buy some, stick it in a brokerage account, and then mint 1 SPX token onchain for every SPX share you have in that account.
You probably can see some of the potential issues just from that one example, like ensuring the project doesn’t sell their backing assets, ensuring they’re minting 1:1 with their purchases (and vice versa), making sure they don’t get hacked off chain etc. There’s a suite of tools and numerous projects now focused on solving these problems, like Dinari and even tokenization products for institutions to do it themselves from the likes of Anchorage.
At the moment RWAs are exploding in popularity, mostly as a way to allow crypto-native users to hedge riskier assets, like via holding stablecoins, but even institutions are using RWAs as a way to access a larger audience outside of their initial markets and to allow these RWAs to be used, or composed, with the rest of DeF. For example, issuing loans against your RWAs or to allow looping, like in the case of Apollo and their ACRED token. But don’t worry, we’ll go into this more in our next post.
——
If you’ve made it this far, thank you for your time. We’ll be posting part 2 and 3 in the next few days. In the meantime feel free to get in contact if you have any questions about applying DeFi, RWAs and tokenization to your business.