Opinion Post: Thoughts on Cardano and its Future in the Smart Contract Platform Space

A detailed analysis of Cardano's architecture and smart contract platform and implementation, conducted by Qluster's head fundamental analyst.

Note: This article includes a lot of technical jargon, in order to better understand the concepts discussed, it’s recommended you have our “Cryptocurrency List of Terms and Definitions” open to maximise understanding.

Before starting, I need to preface this post by stating that I am an investor. Although I enjoy the cryptocurrency ecosystem and its technology, my ultimate goal is to profit from playing this market. Hence, I don’t have a particular bias against any singular project. This is just the conclusion I came to after doing my due diligence.

Before I dive into the issues regarding Cardano, there are some fundamental concepts I need to cover.

First, many people do not understand the software lifecycle. Any good modern application or Dapp that has ever been built has gone through this cycle, ultimately resulting in a bulletproof, secure and robust product.

The cycle is as follows:

Empathize, define, ideate, prototype, test and finally iterate. This is how developers take an idea and turn it into a real-world product.

More importantly, this is how good software is built. Unfortunately, it's almost impossible to predict what users will do with a specific product or network, so the best thing is to continuously release new versions and create new updates that optimise the overall user experience.

This has an upside as well. For example, I'm sure many of my fellow Software engineers have heard of "agile" vs "waterfall" software development styles. I'm not about to hold a lesson on the topic, but a basic summary is that agile is adaptive, constantly changing requirements and specifications to meet a changing environment and changing user needs. On the other hand, the waterfall style has a predetermined set of defined user requirements, and ultimately this is what is built in the end with minimal deviation from the defined path.

You can surmise which method would be the most efficient for an extremely adaptive and versatile platform such as a Dapp ecosystem. Agile.

The next concept which needs to be understood is risk and reward. Fundamentally, risk and reward can be defined as the level of risk of every opportunity to gain a certain reward. Put simply, the higher the risk, the lower the reward is worth, and hence, if the risk is high enough, abandon the trade/investment.

Guest writer Vu Nguyen did a great job covering this concept in the article What to Do and Not Do when Trading. Specifically, point 5:

Establish a profit target risk-reward ratio before you enter a trade       

  • Decide on the potential of the trade versus the size of the investment

  • If it is a significant investment with a small potential return, then pass 

It’s important to note that as of today, Cardano has a market cap of 75 Billion dollars, with only a functioning testnet after 6 years of development. Therefore, you should be asking yourself: Does Cardano have any further room for growth, and will it be able to compete with other rapidly growing Dapp platforms?

I’ll dive into the product below to help you make your decision 👇

Now that the context has been established let us dive into Cardano and its ecosystem.

ADA has been in development hell for 6 years. As a software engineer myself, it's obvious that they are suffering from severe scope creep due to making big promises. As a result, it's tough for them to change their underlying software architecture without literally ripping out years of work and starting again with a more robust codebase that can support modern Dapps and Dapp development. This is also a flaw in not adopting agile software development and changing the software to meet new user requirements, as will be shown with Cardano’s overall network design.

First, let’s take a look at the core issue that instigated the recent FUD wave against Cardano.

Cardano is a UTXO (Unspent Transaction Output) based blockchain, which utilizes a different programming paradigm for Dapps than on account-based blockchains like Ethereum or Solana.

The UTXO model is a verification model. This means network users submit transactions that specify the results of the state transition, defined as new transaction outputs spendable by the recipient. Nodes are then responsible for verifying if the consumed inputs are unspent and if the signature satisfies the defined spending conditions.

An example of this is as follows. A database is used to store the changes made from cryptocurrency transactions. As more transactions are made, the database is filled with records of the changes created from the transactions when all the transactions are finally completed, if leftover outputs are stored back in the database and can be reused later for new transactions.

More specifically, Cardano uses a derivative of the UTXO model, eUTXO. This specific model offers greater security, allows for fee predictability, and offers higher parallelisation levels than the standard UTXO model.

The account model, unlike the UTXO model, is a computational based model. In this model, users submit transactions with instructions to nodes on what the state transitions should look like. The network then computes the new state based on the instructions given.

Hence, it quickly becomes clear why the account model is far more superior for building a smart contract platform. This is simply because it is more efficient at dealing with the more complex logic required in a Dapp platform. Furthermore, the account model has more efficient memory usage because transactions only specify the sender, receiver, the transfer amount, and a single digital signature.

Moreover, the account model is more intuitive, allowing developers to create complex transactions requiring multiple parities or state information, allowing for complex use cases like flash loans, which cannot be executed on the UTXO model.

This brings us all the way back to the original piece of FUD. When one of Cardano’s first decentralised exchanges (DEX) went live, users quickly figured out that the UTXO model isn’t great for building an AMM type exchange like Uniswap.

Users in real time discovering the limitations of the UTXO model.

Many people were quick to point out that this was pure FUD, and Cardano can, in fact, allow for multiple transactions per block. According to an article written by SundaeSwap Labs addressing this exact issue:

Instead, it is accurate to say that Cardano allows a given transaction output to be spent a single time, by a single transaction, so protocols that give multiple people access to the same UTXO might face contention issues.

As many engineers and users pointed out, this is still an extreme limitation and will not scale.

To build a Dapp that can scale on Cardano, a very high level of centralisation is required, which goes against the core ethos of the cryptocurrency ecosystem.

The SundaeSwap team does provide some “methods” for solving these issues. However, non will work efficiently in a real-world system due to the methods of implementation.

Their first solution involves fracturing liquidity across multiple pools, so there are more ports for users to interact with, avoiding the error in the above Reddit screenshot. However, as they admit, fracturing liquidity significantly impacts capital efficiency and overall liquidity.

Even implementing an “order book” style decentralised exchange comes with extreme trade-offs on Cardano. For example, due to contention over the orders closest to the current price, a third-party aggregator would be needed to match orders on-chain, essentially meaning a centralised entity is needed to pair two orders together to get around the unfavourable system architecture.

SundaeSwap does state they do have a different “solution” to the stated issues. However, as of writing this article (08/09/2021), no code has been made public for the community to audit, as well as SundaeSwap not providing specifications on how their solution works.

All these issues stem from a concurrency problem Cardano is facing.

To keep it simple, In a multi-user system, concurrency allows multiple users to access and use the same database at one time, which is referred to as the concurrent execution of the database. In a multi-user system like a DeFi Dapp, concurrent transaction processing is absolutely crucial to allow for significant scaling of a Dapp. Otherwise, users run into the issue mentioned above by Minswap and in the previous Reddit screenshot.

Although Cardano claims eUTXO solves this issue, there is no real-world application, which shows that the concurrency issue can be solved, allowing for complex DeFi Dapps.

Furthermore, other issues with Cardano’s architecture and overall design include:

  • No priority fee gas model (Users still need to wait for extended periods of time to get transactions included in a block during times of high demand), which can significantly impact transaction times, delaying the time it may require to replenish margin on loan, for example, if using a lending protocol, resulting in liquidation.

  • State channels cannot support composable and ownerless smart contracts without relying on stake pools and making liveness assumptions.

  • A severe lack of testing and iteration in the 6 long years of development, resulting in outdated software architecture, although it claims to be a “third-generation blockchain”.

  • A minimal amount of smart contract developers use Haskell, making building on Cardano difficult due to labour shortages in an already small market. Instead, most smart contract platforms use Solidity or Rust, both having significantly more developers versus Haskell. This also allows for interoperability as developers can more easily deploy their Dapps on other smart contract ecosystems that use Solidity for smart contracts.

Now that the fundamental design flaws of Cardano have been discussed, the root of this entire problem can be found by linking to the beginning of the article and discussing Cardano’s chosen software methodology.

Unlike many other smart contract platforms like Ethereum, Fantom, Avax or even Solana, Cardano has had the longest time in development without releasing a reliable product yet. Essentially, all one can do on Cardano is send and receive transactions, and up until recently, stake, which does not justify Cardano’s massive valuation and number three spot in market cap.

Other smart contract platforms took to deploy a minimum viable product as fast as possible to allow users to begin tinkering with the system. Then, iteration can begin, implementing newer technologies, creating robust scaling solutions and developing more complex methods of executing smart contracts.

As Cardano loves to proudly state, they are “research focused”.

However, research does not translate to real-world results. One must actually test designed systems to check if the research is actually viable in the real world, which is an approach Ethereum has taken since the launch of the network. Researching new methods of doing certain actions on the chain, then executing and iterating to remove flaws, improves network function and makes the overall user experience better.

Furthermore, Cardano clearly has followed the waterfall methodology for developing the system due to the outdated technology and an extreme lack of adaptability to modern user needs. Not following the software agile model resulted in developing software that is not competitive in the market current state and creating something with a massive lack of user feedback, resulting in a fundamentally broken product.

Thus, it’s clear that Cardano spent more type researching potential solutions and new methods of creating and executing smart contracts but did not actually test any of this in the real world, resulting in the failure and premature shutdown of one of the first Dapps available on the test net.

Now, let’s look at the usage numbers around Cardano to see if it’s actually gaining adoption versus other smart contract-based chains.

First, let’s look at active addresses:

Both Bitcoin and Ethereum have far more active addresses, with Ethereum having almost 5x more active addresses, interacting much more users with the network.

Network fees paid:

Both Bitcoin and Ethereum have significantly more fees paid on the network, simply due to higher real-world usage.

Fees paid on Layer 1 networks:

As stated above, the protocol fees paid indicate how healthy and utilised a network is. Although being number three in market cap size, Cardano is not even in the top 10.

Furthermore, a recent tweet by Cobie accurately highlights through search results how speculation instead of fundamentals massively fuel ADA.

“Cardano is over-valued and overwhelmingly a marketing-driven and retail-hyped project. It’s reflected in lack of fundamentals/on-chain activity and value settlement vs valuation and retail YouTube behaviour”

Search engine data further highlights that point:

Looking at the TVL (Total Value Locked) across different networks also indicates a massive market disconnect.

Cardano currently has a TVL of 0 because it has no functioning Dapps on the main net.

Ethereum TVL:

Solana TVL:

Avax TVL:

Both Solana and Avax, which are ranked far less than Cardano in market cap size, have billions in TVL, indicating that they are networks with actual real-world utilisation. Cardano, on the other hand, does not even have a functioning DeFi ecosystem.

Many investors are buying Cardano based on extreme speculation, as the data clearly does not support its current valuation, making its current market price severely dislocated from its fair market price.


Cardano is a cryptocurrency that fails to meet current market standards to justify the extremely high valuation. After 6 long years of development, it has no smart contract integration, no decentralised applications and minimal real-world use cases. The recent test net deployments of various Dapp’s show the extreme limitations of their eUTXO, and a lack of user feedback and testing, resulting in a fundamentally broken network with poor software architecture and design.

Many Dapp platforms such as Ethereum, Solana, and Avax are making considerable strides in both technology and user experience and gaining billions of dollars in total value locked due to having hundreds of thousands of real-world users. And yet, Cardano is worth more than both Solana and Avax, which indicates that its current market price is propped up by speculation and not real-world functionality and usage.

Ultimately, Cardano’s research-based approach resulted in developing software that is not only outdated but poorly tested. Researching new methods to implement smart contracts is fine. However, spending only a fraction of that time testing and iterating results in a flawed and unusable network.

If Cardano cannot implement a robust smart contract and Dapp ecosystem soon, I predict a large majority of its investors will move their funds to networks already being used and experiencing massive user adoption, like Ethereum, Fantom, Avax, Solana and so on.

Finally, the cherry on top…

I guess DeFi isn’t so easy, is it Charles? 😏

See you again for the next update.

- Kanj

Leave a comment

Follow us for more detailed analysis on all markets, including Decentralised technology. Check out the links below

Join our Facebook group and speak with likeminded traders:


Like our Facebook page for future updates:


Follow us on LinkedIn for future updates:


Keep up with us on Twitter:


Share Qluster Research

The information on this website is for general information purposes only. It is not intended as legal, financial and/or investment advice and should not be construed and/or relied on as such. Before making any commitment of a legal and/or financial nature you should seek advice from a qualified and registered legal practitioner and/or financial and/or investment adviser. No material contained within this website should be construed or relied upon as providing recommendations in relation to any legal and/or financial product. Qluster does not recommend and/or endorse products and does not receive remuneration based upon investment and/or other decisions