Yes, we reached 13 thousand users thanks to the Prankota channel (we’ll tell about it later) And we decided to summarize a little, and at the same time show why it’s important to merge into a common channel catalog (one site-project-group is one channel), and also merge devices shared user account.

This all goes against the concept of integrated browser push notifications on websites, or notifications in applications, because with this approach, you can receive / send notifications from many different third-party sources to various connected devices. A device means any method of receiving notifications, whether it is a normal browser, phone, tablet, even email, or a telegram - this is a “device” i.e. This is a method for receiving alerts. Perhaps in the future we will be able to send a push notification to your coffee maker ... wait.
')
Many say:

Why do we need a universal service at all, make a subscription through the web push on the site and everything, why do we need this login through Google + ...

Here you can use 2 tactics - let's start with the fact that users choose several devices, web pushes work only in certain browsers, and they need to be signed on each device separately.

3 million alerts? Anyway, no one else can ...

No one can even make 10 thousand alerts per day, so we decided to remove the restriction altogether and see if someone could send as many alerts as possible. We are waiting :)

Device distribution

So, we have 13,000 users. Find out how many of them have devices and how many.

Users who have at least 1 device (which means they most likely use the service) - 6805 people. Yes, a little sad, but only 52% of our audience has connected devices. The rest either did not find an interesting channel, or did not find support for their device (for example, iOS or WP), or simply left earlier. But they can be returned if you give them what they need later.

By the way, now you can count - we have 6805 devices, and we have 9264 devices
Total leaves that on the average 1.36 devices are the share of one user.

More than 1 device (2 and so on) - 1483 - 21%

More than 2 - 478

More than 3 - 184

Then I think it makes little sense to continue. But by the way, there are 10 users who have more than 10 devices.

It turns out that, at least for 21% of users, it is worthwhile to implement a common place for aggregation of devices. With the addition of iOS and WP support, this number should increase significantly.

Channel Distribution

Of the 13,000 users:

9834 is greater than 0 - that is, there are more signed ones than those who do not have devices. Almost 75%. Probably about 3000 did not find a device :)

3753 have more than 1 subscription - 38% of active

2263 more than 2x - 24% of active

1321 more than 3 - 13% of active

570 more than 5

107 more than 10

18 more than 20

There is a greater spread, and a sufficiently large number of users have more than 2 or 3 subscriptions. It is for this reason that we say that other subscriptions may attract a user to subscribe to another channel and yours too - as a result, the user is more motivated to use the service. he is “held” by several sources.

What is the share of devices in PushAll now?

The topic that users do not particularly like Web Push-notification.

On 9265 devices in the database

Native Android application

3032 devices - 32%

Chrome add-on

2893 devices - 31%

WebPush (Chrome + Yandex)

1236 devices - 13%

Telegram

1107 devices - 12%

SocketPush

783 - 8.5%

Email

168 - 1.8%

Safari

38 - 0.4%

Firefox

1 - 0.01%

Results

As you can see, WebPush occupies only 13% of the audience, which has quite a lot of people subscribed to several channels and use several different devices. So aggregation in this plan is beneficial, and many leave positive feedback about the work of our service and the possibility of receiving alerts through a single bridge connecting hundreds of services and thousands of users together.