Decentralized applications (dapps) have gained significant attention from the public since Q3 2017. According to our Q3 2018 Dapp Market Report, dapp users have risen from less than 20,000 in June 2017 to over 950,000 by September 2018. The Ethereum dapps had a good run in the first 3 quarters in 2018 with a total transaction volume of over US$ 2 billion worth of ETH. Although the industry is still in its early stages, it demonstrates the potential of being a market with hundreds of billions of dollars.

We are witnessing the massive landing of dapps here.

However, how to determine whether a dapp project is good or not? Most of the people consider only daily active users (DAU) as the sole measure of the excellence of a dapp. Yet in many cases, it can be inaccurate and misleading. Therefore, we would like to take this opportunity to introduce you a more scientific way of evaluating dapps — Dapp.com Ranking 2.2.1 — and explain how it works.

Why “DAU” Is Misleading?

First, let’s start with the definition of DAU.
Daily active users (DAU) is a way of gauging the success of an internet product such as a social networking service, online game, or mobile app. It calculates the number of users visited or interacted with an internet product or service on a single given day.”[1] “Usually the only requirement for a product user to be considered “active” is that they somehow view the product or engage with the product.”[2]

Introducing DIU
The DAU data is recorded in the product backend that only operators can get access to. Some website traffic measurement tools such as Alexa and SimilarWeb provide estimated data, which oftentimes is missing and incomplete. For most dapp data sites, they measure a dapp’s DAU via monitoring its contract address to find out the number of transaction. The number here is usually referring to “DAU”. Dapp.com ranking, however, takes one step deeper and looks at the unique wallet addresses that interact with the contract on the blockchain. The data extracted according to this criteria, instead of calling it DAU, we’d like to call it DIU — Daily Interactive Users.

Then why do we use DIU instead of DAU?
Users who have logged into or browse the product are counted as DAU, but only those who have interactions with the blockchain are DIU. For example, a dapp game ABC might have 10,000 DAU, but those people just explore the marketplace with zero purchase and transaction activity, then they are not DIU.

Is DIU Important?

Back in 2017, Crypto collectibles were quite popular and we call those dapps “Hot Potato Game” — buy items at low prices and flip at higher prices later. Then many idle games and Ponzi schemes launched in 2018. The daily participated users number became the one and only index — just to make sure a large number of new participants join the game every day to be able to secure higher dividends. A lot of investors studied the game mechanism and the DIU stats to decide whether to invest in newly launched crypto collectibles or other dapps with investment attributes. So people refer to DIU (previously called DAU) as the sole criteria for evaluating a dapp.

We all know that transparency is the key feature of a dapp, and every single detail of wallet addresses (accounts) interact with the smart contracts are recorded on the blockchain. However, people still find a way to cheat and here’s how: That dapp game ABC has 1 user with 1 ETH of transaction volume when that 1 ETH was transferred from an address to the smart contract. But it also could be 100 users have only 1 ETH of total transaction volume if that 1 ETH was split into 100 different wallets and transferred to smart contracts. (Think of this way, 1 apple pie for 1 person vs. 1 apple pie for 100 people.)

Also, some people even build services to provide higher “DAU” (actually DIU) for dapps to get on the top ranking on websites like DappRadar. One site called EOS DAU, which can help dapps create huge amounts of EOS accounts to interact with its contracts and be able to list on the top 3.

So, is DIU really important? Yes, but DIU is NOT the sole criteria for dapps.

How to Identify a Good Dapp?

Well, it is a daunting challenge to pick a good dapp. We at Dapp.com have been researching and analyzing thousands of dapps over the last year. And here are some key metrics that we believe you should look into:

1. Daily Interactive Users (DIU): The number of accounts (users) that interact with the dapp’s smart contracts.
2. Quality of Users: There’s a huge difference between a user with 0.01 ETH in the account and the one 100 ETH in the account, and we consider the one with 100 ETH is a high-quality user.
3. Transaction Amount: It includes non-zero transactions (usually purchases) and zero transactions (actions).
4. Funding Sources of User Accounts: It purges bots and fake users whose funds are from the dapp’s contracts through multiple transfers.
5. User’s Attention: If a user account has 5 out of 10 contract-related activities with a dapp in one day, we consider he/she has high attention to this dapp.
6. Timeliness of User Actions: If a user’s last contribution to a dapp was a few months ago, then his/her recent action(s) to the dapp should reflect differences in its dapp ranking.
7. Community: It includes the community size and activeness. A community is formed by its loyal supporters and users who join the entire journey, from the dapp development to the expansion. It not only shows the potential to grow but also reflects the team’s capability of operations.

Dapp.com Ranking covers all the 7 key metrics in our algorithm and we’ve also added the community data in the new version 2.2.1. The higher the ranking score, the better the dapp is. Also, the growing tendency of the ranking score reflects the potential of a dapp.

In a nutshell, the new Dapp.com Ranking considers the contract performance, user quality, and community performance. Since the Dapp.com Ranking is a linearly cumulative continuous function that decays exponentially over time, the ranking scores are continuously distributed from 0–100. The following scale can be used as references:

• Ranking score 80–100: Top tier dapp. It has a solid user base and “high-quality” users, as well as an active community with high engagement.
• Ranking score 60–80: Good dapp. It means the current dapp status is generally healthy. There may be differences in the scale of effective user transaction and community activity compared to higher-score dapps.
• Ranking score below 60: The dapp may have just released or currently still under development of major functions. Its user base and the number of valid transactions are relatively small. And also the community development is in an early stage. It might be risky sometime when the dapp is categorized as betting or built as Ponzi scheme.

Dapp.com Ranking has its limit too — it can only reflect the current status of the project, yet factors such as the growing tendency of the dapp should be considered into the grading system, too. But as for now, we hope it can help you cut through the noise and evaluate dapp more precisely, and we will continue upgrading along with the growth of the industry.

We will make Dapp.com Ranking open-source at the right time, eventually, we will invite our dapp community to participate and contribute in the near future.

— — — — ——

[1] Henry, Theresa F., et al. “Socially awkward: social media companies’ nonfinancial metrics can send a mixed message.” Journal of Accountancy, Sept. 2014, p. 52+. Business Collection, http://link.galegroup.com/apps/doc/A381838689/ITBC?u=clackamasccl&sid=ITBC&xid=81e00621. Accessed 25 Oct. 2018.https://en.wikipedia.org/wiki/Daily_active_users#cite_note-1

[2] Seif El-Nasr, Magy; Drachen, Anders; Canossa, Alessandro (March 30, 2013). Game Analytics: Maximizing the Value of Player Data. Springer. pp. 56–57. ISBN 978–1–4471–4769–5. https://en.wikipedia.org/wiki/Daily_active_users#cite_note-2

— — — — ——
Original post on https://www.dapp.com/article/can-we-finally-stop-talking-about-dau