Back to Basics: Account Activity

    Terra - 11

    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...


    The question of “active wallets” faces every ecosystem. Let’s investigate this question for Terra — and go beyond the “one transaction in 30 days” approach!

    In this dashboard we will first discuss the methods of measuring active wallets including my proposed method, called diversely active wallets. Then, we will go on to comparing the three method of measurements. And finally we will dive into the the proposed measurement and discuss its pros and cons.

    In a nutshell, the proposed method broaden the definition of activeness by taking into account the types of activities such as swapping, staking etc.

    db_img

    Measurement a: Basic



    Measure b: Timeframe is taken into account



    Measure c: Diversity of Transaction types is taken into account



    Comparison



    The effectiveness of Measurement C


    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...


    The graph below shows the daily number of active wallets based on the measurement a. All wallets that executed at least one transaction have been considered ‘active wallets’.

    If we want to see how this measurement is effective, meaning to what extent the measured active wallets remained engaged in the chain, we can look into the fee that they pay in a finite time period.

    To be able to compare different measurements I focus on active wallets (with different methods) on a certain date (30 days before the current date) and examine the tnx fee payment of those wallets in a certain time period (past 30 days).

    We can see that the average fee paid in the past 30 days by ‘active wallets’ (defined by the basic approach) 30 days ago is about 2.5 LUNA.


    Active wallet measurements


    The following approach will be discussed in this dashboard:

    • measurement a: The basic measurement approach defines active wallets as any wallet that executes a transaction regardless of past and future activities.

      \

    • measurement b: The time period-centered approach defines an active wallet as a wallet that is not inactive in a defined time period. In fact, active wallets have at least two transactions in the defined time period (e.g. 30 days).

      \

    • measurement c: My proposed method add the diversity of activities into these approaches. It lies in the fact that the more a user is embedded in a network the more likely it is to be active in the very network. The degree of embeddedness can be measured as both the strength of relationships and the diversity of relationships. Here we measure the degree of embeddedness as the combination of the number of transactions (strength of relationship) and the types of transactions (diversity of relationships).


    The graph below shows the daily number of active wallets based on measurement b. All wallets that executed at least one transaction on a day with a condition that their previous transaction was not completed longer than 30 days ago have been considered ‘active wallets’.

    As it is done for measurement a, we will calculate the fee that active wallets of 30 days ago, by measurement b, have paid in the past 30 days.

    We can see that the average fee paid in the past 30 days by ‘active wallets’ (defined by the second approach) 30 days ago is about 10 times higher which is 25 LUNA.

    We can understand that the second approach grab users with a higher engagement in the chain than the first approach.

    Now we measure not only the number of activities but also take into account the type of activities. Ten types of activities has been distinguished:

    • Swapping
    • Staking
    • lp activities
    • sending a transfer
    • receiving a transfer
    • minting an NFT
    • selling and NFT
    • buying and NFT
    • voting (taking part in the governance),
    • and other activities

    The number of different activities that a wallet takes part in makes up the diversity score of the activities of that wallet.

    The barchart and donut chart show the distribution of diversity score among wallets that have been active in the past period.

    past period can be selected from dropdown ‘past_period’ above. Available selections are: 1 day, 1 week, 1 month and 1 quarter.

    We can see that almost half the wallets have the diversity score of 2. It implies that in a period of 30 days they engage in two type of transactions.

    About one third of wallets stick into only one type of transaction. Very tiny share of wallets engage in more than 5 types of transaction.

    If we filter only active wallets with the diversity score of more than one we can see that the average fee paid by active wallet can be the same as measurement b showing that the average active users by definition of measurement 2 engage in more than one type of transaction in 30 days.


    Introduction


    The line chart below compares the daily number of active wallets by measures.

    The barchart focuses on the past period and shows by different measurement to what extent the number of active wallets varies.

    For the measurement c the diversity score can be selected using dropdown ‘diversity_score’. For instance, when diversity_score equals to 3 it means that active wallets are selected with more than three type of transactions involved in in the past 30 days.

    We can see that how the number of active wallets can be different when we use different measures.

    Measurement c considers the fewer number of wallets as active compared to measurement a and b.


    The scatter plot shows how the number of wallet’s transactions correlates with their diversity score.

    Interestingly, we can see that the wallets with the greatest transactions have diversity scores equal or less than four.

    The barchart below shows how the average transaction fee paid by wallets varies depending on the diversity score. We can see that the higher diversity score, the higher degree of embeddedness, leads to the higher tx_fee paid by the wallet.

    We can say that taking into account the diversity of type of transactions that wallets are involved can lead us to the wallets with the higher engagement with the Terra chain.


    Discussions


    The dashboard explored measures of active wallets in the Terra ecosystem.

    The basic approach (measurement a) considers any wallet that executes a transaction as active.

    The second approach (measurement b) considers an active wallet to be one that has executed at least two transactions in a defined time period (e.g. 30 days).

    The third and proposed approach (measurement c) takes into account the diversity of transactions, such as swapping, staking, etc., as well as the number of transactions.

    The dashboard shows the average fee paid in the past 30 days by active wallets defined by each approach. The second approach captures wallets with higher engagement in the chain than the first approach, and the third approach captures wallets with the highest engagement on average. The third approach considers fewer wallets as active compared to the other two.

    Now the question is if diverse engagement is crucial how we can increase the active wallets with the highest diversity scores.

    The barchart next to this box demonstrates the what type of transaction is more likely in what steps of wallets onboarding in the Terra chain.

    We can see that most of the wallets begin become engaged with the chain by receiving or sending a transfer. Also staking is one of the early actions on the Terra network. We can see that in the fifth and sixth actions, it is likely that wallets become involved in lp actions. in the seventh or eighth action wallets become involved in the NFT market.

    To increase the engagement of users on the Terra chain one approach can be broadening the type of activities that users are involved in. As the analysis and the last two bar charts show, involvement in NFTs is not the early actions of active wallets on the Terra chain.

    We can expect that onboarding through NFTs can bring about more diversely active wallets. This is the same for being involved in governance. Any strategy that pushes wallets to be involved in a diverse type of activity and even diverse platforms and tools can increase the engagement of active wallets on the chain.

    Loading...