Data and analysis by Danning Sui
The DeFi User Base Overview
By August 10th, around $5B USD had flooded into DeFi, with a high correlation to the recent introduction of liquidity mining/yield farming mechanisms. In the following chart, we see that cumulatively, DeFi user base has grown to 289,072 unique addresses (excluding the double counting of overlapped-users, which otherwise makes the chart total read as ~350k) among just the 14 protocols listed.
Figure 1: User Growth on 14 major DeFi Protocols (by Aug 10th, 2020)
Among the 14 monitored protocols, Uniswap leads the board with over 100k unique Ethereum addresses interacting with them cumulatively to date. The second one is Augur, though without much growth recently.
The protocol with the highest growth in the past month was Balancer. The daily active user chart below gives a clear picture of it.
How Many Daily Active Users Are Seen Across Platforms?
Figure 2: DAU of 14 DeFi protocols. April - August, 2020
Using DAU metrics, we see a clear trend on Compound, Curve, and Balancer demonstrating the recent frenzy of liquidity mining in the community. On June 20th, the number of active DeFi users addresses doubled from ~4k to ~8k, with major rises attributed to Curve (1,098 addresses on the 20th) and Compound (2,877 addresses on the20th). In mid-July, Balancer was the dominant protocol (3,041 addresses on the 21st) with regard to DAU, and made up nearly 40% of all active users. This dominance had continued through early August when we collected this dataset (Aug 10th).
Note that this data only tracks Uniswap V1 contracts, and that may be the main reason why we see a drop in the pink Uniswap section.
In the following section, we would like to discuss the impact of liquidity mining for DeFi user growth. DeFi, yield farming, memecoins, gas fees, and token prices have dominated online discussion on social media and community platforms. But we wanted to answer the question: has the recent frenzy in DeFi attributed to more people using Ethereum? Looking at the data, has DeFi so far proven to be the ‘killer app’ that brings more users into the Web3 financial ecosystem?
We will look at Compound, Curve.fi, and Balancer to answer two questions:
Looking at daily active users (DAU), how many new users were brought into each specific protocol during the yield farming frenzy?
Among those newly introduced users, how many are new to DeFi entirely? If they are not new, but rather existing DeFi users, which other platforms did they come from?
By answering these questions step by step, we can channel down the user cohort and gain a better understanding of how DeFi frenzies like yield farming actually impact the user growth of blockchain technology, and take lessons from this bull market and apply it to the future..
Approach: The number of new users brought by liquidity mining campaignsSince we are trying to identify the newcomers brought by the yield farming events, the “number of daily new users” is a more reasonable metric to look at, rather than “number of daily active users”.
In the following discussion, we assume that all the new users coming to platforms during our analyzed time windows are brought by liquidity mining, though in fact some may have begun using DeFi with different needs or intentions. We acknowledge that more sophisticated work needs to be done to determine the specific intentions behind a behavior of users. In other words, we need to combine other behavior data to determine accurately if an address (seen as a DeFi user in this article) interacted with the protocol for pure lending needs, or just for liquidity mining rewards. However, for the purposes of this article, we can identify a window of time during which Compound’s new users skyrocketed right around the release of $COMP, and accurately assume that the spike in new user activity was due to the attention to the token. We apply that same logic to the other protocols as well.
The time windows when campaigns are effectively bringing in new users are marked in shaded sections in the following charts. We label the first date when there is an abrupt growth with a significant higher level of new users, compared to previous average.
In figure 2, we see that daily active users on Compound increased significantly starting in mid-June and dropped slightly after mid-July. During that time period, DAU was ~3x compared with the time before yield farming.
When we look at the number of daily new users in figure 3, we see that on June 15th, new users started entering Compound for $COMP. The highest peak appeared on June 21st, when there were 1,319 new users in one day on Compound.
Figure 3: Compound’s Daily New Users in 2020
Counting from June 15th to the latest date (Aug 10th), the total number of new addresses was 20,350, while the total number of unique active users is 25,589. That means 79.5% of the recent active users on Compound was successfully acquired from the yield farming campaign.
We want to track further down to answer the question: how big was the impact of Compound’s liquidity mining on the whole DeFi ecosystem? Did they indeed bring in over 20k new DeFi users, or are those addresses just existing DeFi users from other platforms that happened to use Compound for the first time once they heard about yield farming?
From the monitored list of 14 protocols, we found that of the 20,350 new users Compound had, 2,462 had been seen on other 13 platforms throughout Ethereum’s history. That means 88% of their newly acquired users are also new to DeFi (this is assuming a new address is a new user; some common DeFi users, of course, may use different addresses for different on-chain activities).
Tracking down the source of those existing users from other platforms, we see the overlap and user flow segmentation in figure 4 below.
Figure 4: breakdown of compound’s new users coming from other DeFi
With a similar approach, we looked at Curve.fi’s data. The hype trend started on June 15th, which is consistent with information we saw from the DAU chart. Different from Compound, however, Curve show 2 spikes in daily new users - one in mid June and then in mid July. On June 20th, Curve had 523 new users and then on July 21st, it had 628 new users.
Figure 5: Curve’s Daily New Users
Since June 15th, there were 13,207 unique active addresses seen on Curve and 11,306 new users until now (Aug 10th) acquired. It means the mining effectively helped them attract 85.6% of their existing users.
8,091 of those 11,306 new users were seen before June 15th on other DeFi, meaning that 71.56% of their acquired users are existing DeFi users, while 28.44% are new to the DeFi community.
When looking at the pie chart of new user source breakdown, we notice a big shift in the overlap with AAVE, which appeared as the 2nd largest source for new Compound users, but had a much smaller overlap with Curve (and Balancer, discussed in the next section). This is likely because AAVE and Compound both serve lending purposes while Curve and Balancer are DEXes.
Figure 6: breakdown of Curve’s new users coming from other DeFi
Balancer shared a similar pattern with Curve; both had 2 spikes in new daily users. On June 23rd, Balancer had 817 new users; on July 18th, it had 810 new users. With Curve, the new user growth trends both times abruptly hit the highest point and then gradually vibrate down. On Balancer, waves came as a stepp, high peak and then a steady series of lower peaks, which helped it become dominant among DAU in figure 1.
Figure 7: Balancer’s Daily New Users
Since May 29th when Balancer began its liquidity mining, there were 19,300 unique active addresses seen and 18,515 new users acquired until August 10th. It means the mining effectively helped them attract 95.93% of their existing users.
6,892 of those 18,515 new users were seen before May 29th elsewhere in the DeFi network, meaning that 37.22% of their acquired users were existing DeFi users, while 62.78% were new to the DeFi community entirely.
When we have all 3 pie charts compared, an interesting finding is that Uniswap makes up such a stable portion of overlap across different platforms. Its interoperability and adoption are undoubtedly why Uniswap is a rising star connecting all DeFi wheels.
Figure 8: breakdown of Balancer’s new users coming from other DeFi
All together, we conclude the following findings:
Timeline-wise, Balancer launched the liquidity mining mechanism as the first pioneer, as early as May 29th; while only until June 15th when Compound launched their campaign, liquidity mining was brought to the majority’s sight, due to Compound’s wide adoption and user foundation.
Due to yield farming, 88% of Compound’s, 28% of Curve’s, and 63% of Balancer’s new users were also entirely new to DeFi. Comparatively, Compound and Balancer had a larger impact for expanding the community’s user base.
After a short hype of a few days, among those 3 protocols, Balancer showed the most sustainable user growth with a steady trend going forward.
This is a critical dataset to have available as we continue building out the DeFi ecosystem. It suggests that novel financial mechanisms such as permissionless yield farming aren’t just serving a niche community of hyper-knowledgeable individuals (though certainly the ability to maximize these opportunities is helped by pre-existing DeFi knowledge). The DeFi community is growing. Not just the activity within the existing network, but from outside the DeFi world as well. This is a positive message for the future development of DeFi. As novel mechanisms are released, we should expect the DeFi community to continue growing, and for the pie to grow larger as open financial protocols grow in prominence.