Data is the backbone of any successful business, and the web3 ecosystem is no exception. As DeFi continues to gain popularity, lending protocols are constantly looking for new ways to optimize their products and improve the user experience.
Customer data analytics play a crucial role in this effort by providing deep insights into user behavior and preferences. In this article, we’ll introduce customer data analytics and how it can be used in the lending ecosystem to drive growth, increase revenue, and ultimately provide a better experience for borrowers and lenders alike, using the example of the Morpho Protocol.
Morpho is a peer-to-peer yield optimizer protocol deployed on the Ethereum blockchain that operates on top of lending pools. It enables lenders and borrowers to match their liquidity on-chain, resulting in 100% capital utilization and superior rates for both parties. If liquidity cannot be matched, fall-back liquidity is provided through Compound or AAVE pools, with the same rates as these protocols applied. Let’s see what KPIs could be monitored on-chain to better understand customers and improve products.
User Weighting: The key to understanding Customer Behavior
When looking at the growth chart of the cumulative supply and borrow volume in the protocol, we see that Morpho is experiencing rapid growth with over half a billion dollars locked. However, the weekly number of transactions and weekly number of users remain stable at around 120 weekly active addresses. This suggests that users are either highly engaged with the protocol and do not withdraw, or addresses do not have equal importance in this growth.
To better understand user behavior, it’s important to identify the most valuable users of a protocol. At Aleno, we understand that not all customers interact with the protocol or dApp in the same way. The most valuable customers may hold the largest share of tokens, stake the most, or interact the most with the protocol. To map user behavior, we use a user weighting system that takes these factors into consideration.
3% of addresses account for more than 89% of the protocol’s volume
In the case of Morpho, the most valuable customers are those who transfer the most volume to the protocol’s contracts. An analysis of the volume distribution per address that interacts with Morpho reveals that 3% of addresses account for more than 89% of the protocol’s volume, while 66% of addresses that generated less than $10,000 in volume only make up 0.1% of the protocol’s volume. The top users are therefore vital to the protocol.
User weighting allows us to establish a clustered user leaderboard for improved attribution. It helps businesses keep a list of their most valuable users, who can then be rewarded with incentives, airdrops, or in-dapp bonuses. By recognizing that a single user may have multiple addresses, the risk of giving the same bonus multiple times is reduced and the mapping of balances and activity is improved
Additionally, user weighting will help us to map the overall protocol’s behavioral usage.
User Segmentation and Behavioral Analysis
The key advantage of user weighting is the ability to draw a map of user behavior and gain a general overview of their activity. The following table shows a behavioral map for Morpho.
64% of the volume is generated by leveragers
Upon examining the prominent labels sorted by weighted users, we can observe:
- 64% of the volume is generated by leveragers, who are using liquidity loops (re-supplying what has been borrowed). Improving the user experience for this group by facilitating liquidity loops on the Morpho app can potentially attract more volume. (This is now possible with apps such as Instadapp or DefiSaver)
- 48% of the volume comes from clustered users who have connections to other users, highlighting the importance of treating these users as unique entities in the user leaderboard.
- 34% of the volume is generated by lending degens that use multiple lending protocols. More volume could be attracted to Morpho if these users consolidated their liquidity on the protocol.
- 33% of the volume comes from token degens who might be exposed to high market risk.
- 33% of the volume comes from contract degens that may be exposed to high contract risk.
- 32% of inactive addresses (16% of volume) indicate a churn rate.
This user segmentation shows that different groups of users have unique needs and behaviors that impact the protocol’s success. This information can be used to tailor products and services to meet the needs of these key user groups, and ultimately drive growth, increase revenue, and provide a better user experience.
Additionally, it is possible to gain a deeper understanding of user behavior, by examining the evolution of each behavior over time. For example, let’s look at leveragers, to see how traffic has evolved.
For this chart, we divided the leveragers’ volume into two parts: the stETH-ETH leveraged strategy and the rest. The stETH-ETH leveraged strategy enables users to leverage their stETH yield while earning rewards from the Morpho Token. The others, who are only engaged in farming the Morpho Token, can be referred to as farmers.
The chart reveals a decline in farming volume at the beginning of Morpho’s Age 2 (when $MOPHO rewards started for morpho-AAVE users), leading to an increasing share of organic volume. We can also note a shift from farming to the stETH-ETH leveraged strategy.
Traffic analysis is crucial to monitor organic volume and user activity. Its comparison with business milestones or marketing campaigns helps to assess how a specific event impacted the protocol’s usage.
Acquisition and Retention
Let’s dive deeper into traffic analysis to monitor user acquisition and retention. Here we redesigned the classic chart as seen on google analytics to differentiate new users from returning ones.
Here we mainly see that each week there are usually more than 50% of new addresses using the protocol, except during the 3rd epoch of each Age where there are mainly old users. This shows that changes in token incentives are powerful to attract new users.
To monitor engagement, we use cohort analysis, as it helps to distinguish growth metrics from engagement metrics as growth can easily mask engagement problems. In reality, the lack of activity of the old users is being hidden by the impressive growth numbers of new users.
A cohort matrix is a powerful visual representation of user behavior over time. It shows how a group of users (a cohort) interacts with a product or service. In this case, the cohort matrix for opened positions on Morpho displays the number of funds deposited by a user cohort, and how this amount changes over time.
The Y-axis represents the cohort date, which is the date when the users in the cohort first deposited funds on the Morpho protocol. The X-axis represents the number of months elapsed since the beginning of the cohort. The matrix displays the initial value of funds deposited by the cohort at the start of the cohort (X = 0), and how this value evolves. Summing the diagonal would result in the total amount deposited in the morpho protocol. On the following matrix, we see that the biggest cohorts joined in July and November 2022 with 297M$ and 664M$ supplied.
NOTE: This matrix doesn’t consider clustered users yet and only displays the retention per address.
In the previous charts, all metrics pointed to the fact that the protocol is experiencing tremendous growth. However, this can be nuanced by the fact that when looking at cohorts’ retention matrix for opened positions on Morpho, we see that cohorts usually withdraw their funds rather quickly. In fact, less than 20% of liquidity remains after the 3rd month, and few are returning (only June and July Cohorts). To improve engagement Morpho could try educating on the benefit of monthly savings to get recurring deposits on the protocol.
Competitive Analysis
Another powerful feature enabled by on-chain data is user profiling, which could replace Web2 cross-origin cookies. It helps us understand users’ habits through their transaction history. When examining the activity of Morpho’s users, ranked by the value transferred to each contract (Uniswap, Aave, Curve, and Compound would come out on top if aggregated), we can see that after Morpho’s own protocols, the most frequently used contracts are Balancer Vaults and Euler Protocol. Euler Protocol is a lending platform that allows for the lending and borrowing of almost any asset and can be considered a direct competitor of Morpho.
This analysis helps us understand that 26% of Morpho’s users are also depositing to Euler. These financial flows account for 830M$ that Morpho did not manage to attract to the protocol. A deeper analysis will help us understand which tokens were mostly deposited, which could improve the token offering on Morpho, but this is outside the scope of this article.
The beauty of on-chain analytics is that it is transparent and the data is not proprietary. What has been done for Morpho can easily be applied to any other protocol. It enables any business to conduct a competitive analysis for its product research to understand their competitors’ strengths and improve its own weaknesses. It helps to target potential customers, understand their habits, and attract them to the protocol through strategic partnerships with communities or better-designed products.
Conclusion
Customer data analytics is an essential component in enhancing products and optimizing user experience.
Using the example of the Morpho Protocol, we have seen that user weighting and segmentation provide valuable insights into user behavior, preferences, and value. Our study showed that a small group of addresses drive the majority of the protocol’s activity, with a behavior dominance towards leveraging. Although the Morpho Protocol is experiencing rapid growth with a growing number of users, there is still room for improvement in terms of user engagement.
On-chain customer data is a vital tool for product research and customer acquisition in Web3, with the potential to drive growth, boost revenue, and improve the overall user experience.
Don’t miss out on our next post! Follow us on LinkedIn and Twitter for updates. If you want to fully leverage on-chain data, leave a message at contact@aleno.ai & visit our website to learn more.
Written by Angelo Canesso & Vincent Martin
