ALIS Update - Closed beta version usage report

This article was translated by Ambassador Jimmy and communicator Graham! Thank you for all the ALIS communities in the English-speaking countries.

Hey everybody! It has been about one month since the 23rd of April, the day we released the closed β version of our platform. It goes without saying that this has been an important time for the platform, which is all thanks to everybody using the platform and the data we have been able to collect thanks to this. We will present a large part of said data in this article. But before doing so, we would like to first thank you all for your continuous support, feedback and trust.

From today onward, we will be regularly sharing more in-depth information about what we are doing in order to improve the platform. In today’s article, the first in this series, we will present to you a part of our data analysis.

Closed β scope

The closed β version of the platform has been designed only to be able to test a specific hypothesis. If we think of the platform we are aiming to build to be level 100, the current product is somewhere around a level 2 or 3.

Trying to fill a leaking bucket is a waste of time.

The thing we are focusing on first of all is the closed β user retention rate. In other words, are people actually using the platform more than just once or twice before abandoning it? The reason for this is very simple: the ALIS platform is powered by its users, but if the users for some reason seem uninclined to use the platform more than just once or twice, this means it is not able to provide anything of value to new users.

With that being said, we will now share the results of the analysis of the first month after the launch of the closed β version, after which we will announce what we will be working on for the coming period.

ALIS Retention rate is very high Figure 1: Weekly retention rate of ALIS and other products (ALIS data from 2018/4/23~5/26)*1

In Figure 1, we see a comparison of the weekly retention rate of ALIS and other products. For now, ALIS’s retention rate has proven to be 4 times higher than that of smartphone applications, which have been claimed to be used more than web browsers. We were of course pleasantly surprised by this fact, but what is arguably even more surprising is the very high retention rate of the first week.

Figure 2: The retention rate of the first week of smartphone apps per category. (Taken from an article by Adjust)

The figure above shows the 7 day retention rate of smartphone apps of different categories. While apps in every single category on average lose between 80% and 90% of their users one week after downloading, ALIS, on the other hand, only lost 40% of its users after one week of registration.

*1 The data to which we compare our own data to is the retention rate of Android applications in 2016, published by Adjust. We have chosen this data as, while the numbers might not be a 100% exact, it is a useful representation of an overall trend, and presents an especially formidable challenge to ALIS: apps have a higher retention rate than web browsers; In this way, we have chosen data which errs on the side of caution, with which to examine and evaluate ALIS.

How did the Mikka-Bōzu* Blogger enjoy ALIS?

*Someone who does not stick to something for long

As explained above, we have learned that the retention rate of the closed β version is very high. However, this alone does not suffice. Next to numbers, we have to look more in depth to how the platform is actually used. For now, we have focused on trying to look for signs that show that the Mikka-Bōzu Bloggers (people who tried writing blogs before but quit) were inclined to keep using the ALIS, where they had previously given up on other platforms.

At the time of the designing stage, we have created two personas, who we have called Hajime and Ken. The former is a Mikka-Bōzu Blogger and Ken is a seasoned blogger.

For more info, see: https://medium.com/@alismedia/design-concept-a4c37816dfd4

■User division Total registered users: 2,285Hajimes: 1,618 (71%)Kens: 667 (29%)

The data on the persona division was acquired through a questionnaire filled in by the users upon registering. One of the questions in the questionnaire was: “Are you currently writing a blog?”. We have labeled people who answered “I have never written a blog.”, “I have set up a blog, but have never actually written an article.” and “I have written a blog, but am not anymore, currently” as Hajime. We have labeled people who answered “I am currently writing a blog” as Ken.

Figure 3: The horizontal line was set based on the average monthly articles written by Kens

Figure 3 shows the % of users of both personas wrote more than the average amount of articles per month written by Ken (seasoned blogger). The y axis represents the two personas and the x-axis represents the threshold of an average of 7.7 articles per month written by Kens.

The reason for setting the threshold at 7.7 articles is because in order to not end up with a half-baked conclusion, we tried to be as strict as possible regarding this matter. In other words, instead of looking simply to see if the users labeled as Hajime wrote multiple articles every day or every week, we believe a better indicator is to compare their data to that of the seasoned bloggers.

The data shows that, 26% of the users labeled as Hajime, who ended up not continuing with their previous blog, exceeded the threshold of 7.7 articles, meaning they write more articles than the average seasoned blogger. Also meaning they did not just simply write multiple articles in one month.

Percentage of article creators in the media

Although the data is a little dated (2009), according to a survey done by the Japanese Ministry of Internal Affairs and Communications, only 2% o people using blog media are people who continuously write blogs. In other words, 98% of the people do not consequently produce information on current Japanese blog media platforms.

See (Japanese): http://www.soumu.go.jp/iicp/chousakenkyu/data/research/survey/telecom/2009/2009-02.pdf

Possible reason for ALIS’ high retention rate

As for the reason why ALIS has been able to maintain such a high retention rate, we think this might have to do with that, compared to existing services; it is easier for users to acquire some form of feedback on their published articles and establish relationships with other users through the ALIS platform.

Generally speaking, on existing platforms, it is relatively difficult for a person who just started a blog to get any form of feedback or any form of compensation. On these platforms, this is only possible after writing multiple articles, paying attention to SEO, and also, relying on access through other media outlets.

We believe that with regards to the high retention rate, this is an important difference between ALIS and other platforms. The token system of the platform and the logic behind it motivates users to read and rate articles, which results in an environment where even people who have just published their first article are also able to gain some form of feedback.

As far as this phenomenon goes, at the current stage we would like to wait on sharing any specifics, but in analyzing the personal connections between users on the platform, we were able to conclude the chances our assumption is correct are relatively high. We will share more on this once we have gathered more data.

Next Action

A summary of what we have learned after one month of the closed β release.

The retention rate of the ALIS platform is 4 times higher than that of smartphone applications, which are said to be more used than regular internet browsersWhere with regular smartphone apps, 80%~90% of the users abandon an app within a week, only 40% of the users left ALIS within a week25% of the Mikka Bōzu Bloggers exceeded the threshold of 7.7 articles, the average amount of articles published monthly by seasoned bloggers. This is 12.5 times more than the comparable rates on existing Japanese blog platforms, which enables us to assume we can reach that many more potential usersWe have argued that a reason for this might be because people on ALIS are able to receive feedback and rewards much easier and earlier than on existing platforms Functions we are planning on adding For the reader: search functionFor the writer: comments, notifications,

✳These plans might change without prior notice.

Apart from these steps we will of course also be working on the logic that determines the trustworthiness of articles, and a solution for the current trend where the platform tends to lean to articles about specific coins. And, we are also planning to work on effective platform evaluation protocols.

Before we wrap up this article, we want to thank everybody using the platform and others that have been continuously helping us to improve the ALIS platform. We will continue to do our very best to answer everybody’s feedback on discord, and to live up to your expectations. We are looking forward to working together with the community in order to create the best platform possible!