>>Hi, everyone, welcome, thank you so much for joining this Session today. First, i would like to start by Introducing myself. My name is blanca, i work as a Program manager in the resource ops center people. We are going to talk about how to monitor your autopsy using App center. If i had to describe app center With just a few words i would say app center is a place for Your apps to get built, tested and received and the place for Your apps to get money turn. So it’s for continuous Integration and continuous pipeline as well as monitoring Services. I used to spend a lot of time Writing in labs, and we would come up with a lot of time Writing code. Then if we got something that we Thought was cool enough we would publish it to the store and then We would go to the store and take our reviews, our star Rating and so on. The feedback was not as good as We expected and we didn’t know why, we got a low star rating we Were like, what’s happening. We heard comments that the app Was crashing, and not behaving as supported. So for that situation, app center monitoring would have Helped me to understand what was going wrong and what was going Right and use that in order to improve my app. So app center accounts with two crucial services when it comes To monitoring your app. We have diagnostic services Which include crash and error reporting and we have analytics Offering. As part of the crash reporting, Crash would be those that are known as errors when your app Exits unexpectedly and we have error reporting which we believe those are situations that we can Handle. So for this session i just want To talk about error reporting. App center recently had this Feature announced back in march and we have error Reporting for the platform. Let’s say you are having issues With the app, there is a request timeout or server did not respond. By using error reporting handling these errors you can Make your app behave as exported and go to app center and realize The errors and see what is going wrong. With that i’m going to jump into the demo and see how errors look Like in app center. The first thing i want to show Is how do you monitor your app? We have our analytics model and then we have our crash module Which includes crash and error reporting. In this case this is my hotel App which you have probably seen before in the old app center. With one line of integration you can see the app center star Line. The sdk will get analytics and You will be able to track out of the box. In order to track errors we have Apa called track error and by doing that all errors will get Reported to app center. With that let me move into the Error side. This is a diagnostic service, Has crash reporting and errors. I’m in the error side. Here what we see is the error groups. As a developer we are finding those errors that we think are Important to track in our apps and there are going to be errors Happening continuously and we want to make it easy for Developers to manage those and make sure we fix them on time And make sure it’s be not too much to handle. We need to group those similarities, and they will be Counted as one error group. These are all the error groups That i have for my apps and i’m going to drill down into one of These error groups, for example, request timeout. I decided to track this error within my app and so this is What information i’m getting. First thing i want to know how Much is this happening? do i need to prioritize that? For that we provide information of analytics on your errors, Like how many users are affected and how much is this happening. That information should give me an idea of should i prioritize The error, is it important for my users or not? Also we need to understand what type of devices are being Affected by those errors so i can prioritize, do i care about That, if i do then that’s an error i probably want to tackle. So on top of that you can drill down into the specific instances for your errors so for this Group i have a bunch of instances happening and Organized by device nos. By looking at my metadata i know That i have s8’s being affected by my error and i’m going to Filter down by galaxy s8’s and the latest os so that way i Filter and decide to look at this and i’m going to click into One of these instances. Once i go into the instances, What do i need of in order to fix an error. I need to know where in my code the error is helping so it can Help me fix it. This is what we see here, we see It tracked by sdk, and we get information, such as os, Country, language of the device and even if you want some other Con physical examination wail information — contextual Information you can attach custom properties into it. So let’s say, for example, in this request timeout you want to Know what it causing this timeout or you want to follow up With users that are having issues so you might want to have Property call user id so you can follow up with them. So with that we saw how error reporting can help you fix Issues in your app and make sure users are having the expected Behavior, but there is more than that. We know when our app is going wrong but we want to know when Is it going right and what are users doing with the app so then We have analytics center. So i’m going to show the Analytics for my hotel app that allows you to book rooms within The app so i can see the rooms, make bookings, and on on. We want to know how popular is the app, how many users are Looking in the app and besides number of installs how much are Engaged with the app, how many times are they opening the app Per day, how long are they staying in my app so this is the Information we try to provide, also engagement of users. Also you might want to check the analytics to See for seeing and making sure that if there are Device errors i fix those and same for the oust s solution. Also for developers, pms, let’s say, we decide to run marketing Campaigns to make our apps more popular and we may want to Tackle those in a particular country and specific language And for that be understanding where my user base is located And what language that he speak i can make those decisions smarter. Last but not least we make our first version, then we make improvements, we have new Versions released and as developers we want to make sure That we are forcing in our latest versions however we know That a lot of users do not download the new versions Because they may think the newest version is worse or they Fail to update it. So by understanding the users’ Versions we can know whether we can stop supporting older Versions, in my hotel app no one is using 1.2.3 Version so we Don’t need to update that one anymore. All of this comes out of the box as analytics sdk. If you want more developer-defined behavior on Your app you can use custom events. I’m going to move for that. I see a bunch of events that the Developer defined as important in the app. In the app you have room suggestions, you can book rooms, And for example for us it’s important to know how many People are seeing the rooms, how many people are booking the Rooms, et cetera. So for each of them we know how Popular they are, how many users are seeing those events or Triggering them. Just recently we added a new Capability in our app which allows you to see when you Purchase a room, and i’m going to move down to that one. Oning see here by the end of april there was a big increase Of number of times that this happened and the main reason was We did a ui tent on our anticipate and moved the Functionality in some way that was more visible. We can see how that durable function a consequence in our Metrics. Moreover you might want to Understand how many people are using this feature but for this Example we want to know how much time our users are giving us, so We see properties and errors and there is an event where you can Attach custom properties into it. So this is what we see here, the property we filed here is the Amount and the value of the tip they gave us. So, for example, in our example we see most of our users gave us A minimum, $5 but it’s better than nothing, so we can see that Here. So the title of this talk is the Power is yourself and we really think it is. What we do in app center is allow you to explore raw data Into azure and we currently have support for azure blob sites. With just a few settings on the app you can explore all your data into azure. What i’m going to show today is how this looks like in Application insight. So i have set up my site and i Am pretty sure you are familiar with this and we show the same Information as we show in app center, app information, events Information, sessions, et cetera, but what’s Interesting besides is you can drill down farther. We have here information over the last three months but then i Might want to — like we care about this boom room event Because that’s what makes us money, when people book rooms. So what i want to see here is what happens when i look on my Users that actually book a room. How many of these book a room so I’m going to say give me everyone who booked a room. I can even, like, filter down to 24 hours up to days, and i can See that by one day or one week. It fluctuates, there is not a Clear pattern of growth and i can also here — let’s say my App runs in the u.S. So i want to see my u.S. Users, how can i Filter this graph to see my u.S. Users? I can say add filter, filter by country, country or region and Then i can say give me everyone which is in the united states. So once this loads — there you go. Then you see this new graph, showing 2 .6K users in the u.S. Doing the book room event. So application size also gives Us more advanced analytics, function, and today we’re going To show an interesting item in our app. As i said, book room is the most important event in our app and From the people that see my room suggestions how many of them go Through the flow and end up purchasing a room? I’m going to say my first step i want to see how many people are Seeing rooms, then i want to say how many of them actually Initiate the process of booking a room? So i see that about 50 of them that see rooms actually initiate The process of booking a room which actually i think is successful. Out of those i want to see how many of them go through the Whole flow and end up purchasing the room? I’m going to say first step, book room. I see that about 18 goes through the whole flow and ends Up booking a room. There is some sort of back or Something going on in the u.S. — Starting at book room All the way up to booking a room so we have designed the flow of Action that users are taking in the app. Also there is a way to look at your raw data and this is what I’m going to show next. For this same app, i have a few Queries i wanted to show you which use custom events and Properties inside of the events. So in app center i define an Event called rating and this tells me basically how good my App is and the rating is from 1 to 5 so i want to see how people Are rating me, how are we doing in the store so i say give me All the custom events, from those to the event called rating Which gives me a rating and then what i do here is select from The customs i mentioned, look at the properties and the rates which is the value of the rate And then i say give me how many users gave me a 1, 2, et cetera. So if i ran this query, we will See that here. I see here i’ve organized these By amount, most of my users are giving me a 5 star which is Awesome but then we have a few, pretty high number for rating 1 And the rest are kind of in the middle. I want to show similar, so in this case we have an install Option for those who book a room and you can basically download By different types of files. I wanted to understand what Files are more successful because i am thinking fished add Anyone or if they are good enough. I select my properties and then i select a property called file Type. Once i run these this is the Results i get. If i see on the chart i see that This is the most popular, followed by Jpeg. You can get query on your raw Data, however, you might want to have all of these organized in a Dashboard that anyone can look at it and you don’t have to run The queries all the time. So i can hit here and say Export, the file down loads and i can use export to create my Own dash boards. So i did that and i wanted to Show you other results i got here. For me it’s important to follow up with those customers that are Not happy with my app so i did a query that would give me user Ids that raid me with a 1 or with a 2 which i consider low Factors, so i can look up in database and i can email person And ask them, what are we doing wrong? why don’t you like app? And here i have number of users by different ratings, 5 most Popular, followed by 1. Similar to what i was showing in The query, file type, and then this is similar to what it Provides, it’s a different way of utilizing the data. So here i wanted to see which was more popular also included In the u.S. Information. So we saw basically to Summarize, our sdk, with our one-line integration we were Able to get out of the box errors, out of the box Analytics, being able to explore the data in blob storage and Access to our raw data and create our own visualization and Dashboard based on developer’s choice. So that’s free with app center. Everything you saw today is Completely free. We encourage you to fry it out And we have extensive awe mentation and we are going to be In the tools booth so please come see us if you have any questions. We are happy to hear that, thank you very much.