Flurry: The Key Metrics You Should Be Tracking in Your Mobile App

awesome welcome to the mobile app chat podcast I am your host Steve p young and I've got two awesome guests for you today I've got Alex fredman who's a senior manager of analytics and Mike Roland who's the lead engineer of partner Integrations of flurry and I'm super excited to have you both on because we're going to talk all about mobile analytics and what you should be doing to make sure you're properly tracking the right elements the right analytics within your app Alex Mike welcome to the show thank you Steve it's good to be here thank you cool so just for everybody I think we know who's Alex and who's Mike but this is Mike Mike you gotta say something Mike this is oh hey guys I'm waving at the camera hi there's a video element as well for those who are listening cuz we're going to talk a lot about you know how to integrate and there's going to be some codes there's a video El if you

want to check that out you can go to mobilea chat.com flurry and you'll you'll find the the show notes to it and here's Alex hi everyone cool all right Mike Alex I'll I'll leave it up to you guys all right thank you guys um so um as Steve said my name is Mike Rollins and with me is Alex fredman we both work for flurry um Alex I let introduce yourself sure um so this is Alex I am responsible for developing and retaining key clients for flurry and I work with some of our biggest clients on their mobile business strategy um and down to the nitty-gritty of their data to the high level questions that they have in Mobile sure and um my name is Mike Rollins um I'm a lead engineer with the partner integration team um I am responsible for getting people integrated into the flurry platform that can span from just uh integrating the SDK all the way up to integrating systems so it's a a fun job to have cool so today we're going to take

you through a presentation but um before we get into uh how to acquire the best users and how to retain users we wanted to just give you a very brief overview of what is flurry analytics um just to set the tone for today uh so we have um one slide up for those who can see it you can see a couple of interesting stats about flurry we're um the number one player in Mobile Analytics today uh we're tracking 1.2 billion monthly unique users on our platform which basically means that we're we're able to see data on 90% of the world smartphones and tablets um we have about 155,000 app developers um using our analytics product 500,000 app total um and we're tracking 1.3 trillion monthly events and events are essentially actions um that you can track within an app for example if a user makes a purchase or gets to the next level in a game and you can see some of the some of the brands that we're working with on

the right side there um we're working you know because we're working with 500,000 apps we're working with from the smallest um one or two person app shop to some of the largest Brands cool yeah it's great to see some past guests of mine on the on the logos as well Shazam and I think the other one that I saw Smule was a past guest as well so good stuff good both good partners yeah um so we're going to go over a few topics today um one of the things the first thing we're going to talk about is you need to know your data um your data is very important and it can tell you a lot about um the second topic that we're going to lead into which is how to identify your most valuable users um now your most uh valuable users might be people that you can monetize or your most valuable users might be people that bring more people into your ecosystem um and using the data that you set up um in the first section you'll be able to

capture who are your most valuable users and then the third thing we're going to show you how to do is how to measure success so once you've identified your most valuable users um we're going to take you into what are the ways that you can continue to capture them um using some of the other tools that FL has um so right now we're going to hit know your data and so um you know what are the basic metrics that every app should track right so if you're going to use analytics what do you need to know what are the table sticks what are the things that have to come right out of the box well you need to know new users you need to know sessions you need to know active users and you need to know retained users um those are those are very basic to being able to do anything so um you know for instance let's take sessions um what if you release a new version and as a result um you introduce a crash and

you you may or may not know about it but you can watch your sessions and all of a sudden you notice a session did um well that you know could be an indication that there's a problem and so that kind of data can be very valuable or conversely um what happens if you uh launch a new feature and all of a sudden your number of new users go through the roof um well that's a great indication that maybe you're really on to something hot and you should continue doing what you're doing um but Mike yeah go ahead so all this stuff is in flurry right once we get that integrated all this mobile user sessions all these four things that you kind of laid out they're all part of it already yes as a matter of fact all this is is is a basic um function of the actual flurry integration so just by including the SDK you get this right out of the box um and the the basic integration is extraordinarily easy in some cases it's

just a couple lines of code um so yeah that's a very good point you do get this right out of the box with flurry and and quite a few other things as well so cool yeah I I'll also jump in there really quickly so the Met that Mike just walked through are some of the basic usage and Technical metrics so all of the new user session um active user data um are just the first section of of various different groups of data that you can have in flurry we also have audience and performance metrics where you can understand what users are doing um cross app what categories that they're spending time in what their age and gender is um geography language in addition to other features like crash Analytics um and user acquisition analytics um but that is just the basics you probably want to know things that pertain specifically to your um application um and so we uh in flurry parlance call these events and I think

that's a pretty common term throughout the industry um so you know when would you want to capture an event well you want to capture an event based on the business goals of your application um so let's say you're interested in when a user makes a purchase you probably want to capture that as an event maybe you're a game you want to know when a user completes a level because if the user completes a certain level you're going to consider them to be very engaged uh maybe you're you have a very strong social component so you want to know shares on Facebook um well you can log an event for that um there's all kinds of things that you can log events for um it's it's pretty wide open um with the flurry system um you know you can not only log vents but you can capture parameters around it and later slice and dier metrics based on those parameters um so uh and that leads us into how granular can events and

parameters be um well it turns out that you know not only can you have an event and so with this particular event we're looking at an episode watched um and and it turns out that you know we're going to be watching an episode of Nashville so we're going to log the event um that we're watching Nashville um and then we're going to capture two different parameters um we're going to capture the episode number which could be the first episode of Nashville which is fantabulous and very good um and we're gonna we're going to log the duration of that watch and that's one of the things you can do with flurry is you can track um durations of events and so um each parameter can have multiple values it could be episode one two three five seven eight 10 um or maybe you know most people are going to watch for the full I guess it's what 40 minutes um but you know maybe some people only watch for 15 minutes maybe that wasn't an interesting

episode or the music wasn't quite as good um but you can parameterize your events such that you can capture a lot of extra information um and so you can also reuse events as a result um over and over which is important because in the flurry system you do get a total of 300 events um once you hit that 301st event um you know you're you're no longer able to capture data VI um so what are some sample events that you might have in game apps um so if you're in a game um it it you know the level that you're play um and so you know as soon as the level starts you might trigger an event that says user started level one um the parameters might include the level played the level started or the level restarted you might um want to know timing um how long does it take the user to to play level one um things like that or um you know what about navigation Events maybe um you've got a tutorial in your system in your in

your app and you want to know which um Step they're on um and how long it's taking them to get through each step and so you might log an event that says um you know they're in step three and the duration is five minutes um and in fact you designed it so that that step three would take a minute and so now maybe you know that you've got a problem on step three or something that's you know concerning or confusing to the user um mon ization is a really big one uh you know you want to know every single step of the funnel around when your user is entering a monetization thing um so you want to know that the user Taps on the buy um the the buy button um you want to know that the user has selected the the number of coins or the level or you know the thing that you're interested in you want to know that the user has initiated the Ina purchase and then you want to know that the user got to the end of the

Ina purchase and um as we'll we'll show an example later um that was you know based on a real life thing um tracking every single step of the way can actually show you some really valuable information um when it comes to monetization and then like we mentioned earlier social um you you know uh being having your app spread by word of mouth is pretty important um and maybe you want to know when people share you on Facebook or when they pin you on Pinterest um and you know that that would be a great type of um event to uh to capture so um Mike so say we're we're gonna integrate flurry we've kind of built out the game already do you kind of recommend kind of going through all the important events and laying them down like maybe you know if I was doing this on the web I would kind of look at the funnel and be like all right here's our funnel here are all the pages that we want to track should we do that

before we even like you know should we plan that out before we integrate any of this stuff beforehand you know it's it's quite funny I normally talk about this ler in the presentation but um you know in my position um with what I do I hear a lot of this hey we really want to put flurry into the app we're about two days from submitting to the App Store can you help us figure out what to track and you know you can do a good job in two days but if you think about it from the beginning um if you think about the things that you know are going to be important to you or know are going to be valuable to you and you implement those in the beginning not only does it help you with your analytics implementation but it's my firm belief that that helps you with your app development and your product development because now you've identifyed the very important points within your app that you need to know

about so you might even recommend somebody who's mocking things up they've got some the flow already done say you know like maybe highlight certain things that they want tracked as event okay cool I like that love it um so okay um you know maybe you're a product manager and you've just heard about this great thing called analytics and you've heard about flurry and you hear that they do great analytics um and now you've got to go convince the developer to put it in and you know I'm an engineer I'm a developer I know that we're all pretty much kogin at heart um and you bring me something and you're like well what's it gonna cost me to put in um and it turns out that the basic integration for flurry um is typically about five minutes um it's a little bit quicker on iOS it's a little bit longer on Android um and we'll we'll delve into that but you know what we're showing right here is that um a typical integration on

flurry is you know you import flurry Doh um and your application did finish launching with options you call flurry start session along with the API key which you get off the dev portal um and you've integrated flurry every time that app launches flurry is going to be initialized flurry is going to start collecting analytics and flurry is going to be available for you to throw events at um you know if you want to get into a a full full Advanced implementation um you know it can take a few days not necessarily weeks and if you've got your your road map laid out in advance if you know what your business goals are you could probably get it knocked out in about an hour um and so you know the other thing to remember is that you know flurry analytics is basically software as a service um model and it doesn't require support from your it Department your it Department doesn't have to stand up a server we maintain all that stuff for

you and we Prov provide you the tools to get access to the data um so a basic iOS integration we kind of went over this a little bit on the other one on the other slide um you know if you look up at the top here basic integration application did finish launching with options um you get your API key you call start session with the API key and you're done um of course you got to import flurry Doh but that's not a big deal um you know for vin it's a little bit more in-depth but it's still pretty simple um you know you basically have the event itself and you can pass in parameters to that event the parameters are collected within a dictionary and so you create the dictionary that's your parameters and then the next step is you log the event with you know article red and with parameters and you pass in the dictionary that you just created um there's a few caveats here um the dictionary must contain only strings um

and you can have no more than 10 parameters um and and you know the reason for that is that it at some point you've got to kind of limit the data that you collect and it has to be easy to work with and obviously if we let people log 100 parameters um there would be people that would do it um and that that tends to make the processing pretty complex on the backside so you only get 10 parameters um and they have to be strings hey Mike how many platforms do you integrate with like you know I know you guys are on Corona and in in The Mastermind we were kind of going through flurry integration and with Corona it was super easy too it's like one line but what other platforms are you you you guys integrate with well let's see um we're integrated with most of the major platforms we've got Unity Integrations um we've got Corona um we've got um Titanium it's accelerator titanium um I know that there are some Coco 2D modules

out there um typically uh if anybody wants to integrate on just about any kind of a modern platform um you can do it um there's you know monotouch there's there's all kinds different things um so um now the the thing that that's worth noting is that typically those efforts are community supported um flurry you know we we're you know we're only like 120 people um we can't necessarily write an adapter for every platform that's out there because there are quite a few platforms um so what we'll typically do is provide resources and capabilities to um the communities themselves um in order to have them build it for themselves cool um so what does an Android integration uh look like now Android is you know the the the function the the structure of an iOS um app is that you always enter and exit the application through the same point which is the application did finish launching with options method um

on Android because of the activity structure that's not the case on an Android application you can actually enter the app on any of the the number of activities that are within the application itself and so that changes the um the integration just a little bit um the basic integration on Android is that on every activity's onart method um you make a call to onart session with your API key um and on every activities on stop methods you make a call to on end session um now one real uh successful approach we've seen people take to doing this is that they create a subass of the basic activity and they call it something like flurry activity um and that handles the onart and onstop Method calls for them um and that tends to be a really good way uh to make sure that your flurry integration is cented um now what about events um you know it's it's almost exactly like uh the the iOS um events themselves you know you

basically log an event you give it a string name um and you pass in a hashmap instead of a dictionary but they're basically the same thing um and you know you can put up to 10 parameters in there and all the parameters and values Must Be Strings so any questions so far all right um and so now we're GNA talk a little bit about identifying your most valuable users um do you want to take kick this one off Alex sure okay um thanks Mike um so one very important question that we get asked all the time by app developers and lots of big brands that we work with is how to identify and acquire valuable users and um there are a lot of different ways to do it um we have two in mind for today to really understand the different segments and personas or behavioral groups within your audience and be able to um effectively use that understanding to acquire more of them so I'll start with segments um so segment in flurry

allows you to analyze how a different group uh that uses your application varies in usage and behavior so for example if you have a game uh you may want to understand how users who reached level three in the US differ differ from users who reached level three in another country for example Germany and with segments you can actually um create segments in flurry for both groups and then then examine all of the metrics um in flurry analytics for each of these segments independently and be able to compare them and so you know you're you're able to create um a whole variety of segments within flurry some powerful examples that we've seen customers use recently are paying versus non-paying customers or casual versus Power users or demographic defin segments like men versus women and then on once you've built those segments in flurry um you can use this information to uh plan your next acquisition campaign and try to

acquire users that are most engaged in your app um or package key parts of your audience to sell directly to advertisers and you know there are a variety of different ways that you can segment your audience and flurry um one way is cohorts so you can understand a group of users based on their acquisition date or when they first started using your app h a second way is by usage level so you can define a group of heavy regular or infrequent users and really Define that however you want you can the Third Way is you can define a segment based on events uh so as Mike mentioned an event is any action that a user takes in your app and you can have a segment of users that completed one two three um events or even parameters of the events for any value you specify and have you know a group of users that have made a purchase of x denomination um the fourth way to build a segment as I mentioned is using

demographics and geography so you can have an age gender location uh language based segment and then the final way is by acquisition source so if you're tracking um user acquisition channels and campaigns in flurry you can actually um see how those users acquired in a particular Channel or campaign uh perform and engage in your app hey Al yeah how do you how do you get demographic data yeah um so so we actually um we uh we use a couple different methods um there are some self-reported uh demographic apps so you know you know typically a dating app might be interested in reporting that you're a male or a female or what your age is or things like that um and we are able because of the data that we have and the data store that we have and the fact that you know um we can we do see so many apps um to start taking this core set of data these small you know um you know a particular dating app that passes up age and gender um and

start extrapolating that using some um pretty complex models over to other users and so we're actually able to give you pretty reliable estimates um BAS based on you know the apps that you use and the other data that we've collected from various sources so um it's a it's actually pretty cool and I remember I was around when we first started developing it and um I signed up to say okay you know you think you can Peg me um it was kind of a challenge from Dave leam who's our chief architect you know let me see if I can Peg you and um so I I gave him my information and he put me into the system and it popped out that you know here I am um I'm a a male uh between the ages of 32 and 36 uh that's likely you know um I think it slid me into a couple person as it were pretty accurate so um I think in general this is this is pretty accurate information in the models that it's based on are pretty complex so okay yeah one one

quick data point to offer there is that um the the size of our data pool is that we've collected um age and gender information from 30 million users which is an enormous sample compared to um how a lot of other companies do it so I think that this is by far the largest um mobile data sample for agent gender data that we're using um to base the estimates on cool thanks yeah um so the next the the next key area that uh we typically look to to Really identify and acquire valuable users and understand which segments of your users um are the most engaged Andor uh the most valuable to Target going forward are personas um and flurry personas are basically behavioral profiles of your user base um that allow you to understand more about your app audience and Target them and we have over 40 personas you can see some of them listed here we have business Travelers new moms fashionistas um hardcore and midcore Gamers and personas

are based on actual mobile usage across um nearly all of the apps that we track on over a billion devices so um we're really able to see um you know for a specific user which group they fit into based on a number of different apps that they use and um you know we get asked a lot how how do personas work so basically each Persona is based on a basket of apps that we're tracking um that are representative of the interest and behavior of that audience group so for example the sports fan Persona um includes apps that let users check sports scores and read sports news and engage in fantasy leagues and that kind of thing and users of a given app are added to a Persona when their most recent activity indicates a very high level of usage across one or more apps in that basket of apps so for example for me to fit into the sports fan Persona I would have to generate more than 25 times as many sessions as the

average mobile user in that type of app so we're really looking at the whale users here love that yeah it's it's a a very very popular feature um on the advertising side and we see we see folks using Persona data um both as a way to understand more about you know the interests and needs of their users um and also to improve monetization in the app and increase Roi and marketing spend because you really know who's in there and what they're doing um and so now we're going to talk about how do you measure your success so you've gone through you've instrumented your application um you've you've you've put analytics in you've put events in um and you've taken that instrumentation and You' started to identify your most valuable users within uh your application um so now how do you measure success and how do you start acquiring more of those users um and we actually have a tool that we call user

acquisition analytics um and so this this is typically a tool that would be used in some kind of an acquisition campaign so the basics of it work like this um you identify a an acquisition channel so that acquisition channel can be somebody that's predefined um such as you know a jump tap or a millennial or a flurry um and you uh can also identify by a custom Channel such as a web campaign or an email campaign um and each of these channels gets their own unique link um that allows you to uh basically capture the click and the install because your app has flurry um flurry analytics implemented um and when that install occurs we can tell you who to attribute that install back to based on last click um now the great thing about this is that it you know can tie into all of your demographics all of your flurry information um and it allows you to determine who bu who delers the best um the best quality now one thing

that you can do with user acquisition analytics is very interesting is that you can set up a campaign um and you can go in and say I don't know if you guys can see my mouse but you know here we are in Jump tap um Sirius Spar is Q3 V1 and they've got 683 th000 clicks 15,000 installs which is an install rate of 2.3% but we have this quality metric over here and that's pretty important um because you can set a segment or an event or a funnel um to highlight that a user is a quality user and so you know just getting to the install is one thing getting beyond the install is something else entirely um and so let's say that you're a game which is you know you're the serious sparrows game um and you want to know who makes it to level three and that's who you've determined as your quality user because in your analytics um and in your segments you may have determined that if anybody gets to level three they're actually very likely to

conduct an inap purchase um and so what you might see here is that you know with the volume um that you're looking at it looks like this web campaign right here um has the best conversion rate into your quality so maybe you want to focus on your web campaigns a little bit more and reduce spending on some of the the other campaigns that you're that you're using um you know it's it's it's a pretty straightforward product um you set it up in the dashboard you get a uh link that is used as part of the click um the user installs the app the the the link takes them to the App Store um they install the app as soon as they launch the app flurry records the app install um and then flurry goes through a matching process depending on whether it's device identifier based or whether it's um uh device fingerprinting based um and we match up the install with the um the the click um and then we report on it in the

system and that reporting is generally available within two hours Alex sure um so another key way to measure success is by using the funnels feature in flurry and so um for those who aren't entirely familiar with funnels funnels track users as they execute a defined series of steps allowing you to basically see how many users that started a given process ended up completing it um and so for those that did not complete the process then I'll show you at which Step they exited and you can use that as a feedback loop to understand how engaged users are in the events that matter to you um with funnels and flurry any event that you have set up in the app can be used to define a step and then you can answer questions like you you know how many users that viewed the tutorial ended up finishing level two um how did our last um change to the confirmation process in our app affect the number of people that

successfully registered or how many users make an inapp purchase you know within an hour a day or a week after launching an app and you know this the process that we're showing here um is a hotel app and you can see that you know in the first step um 75% users were served a list of hotels 35% then clicked um to select a hotel 20% entered their payment information and then 177% in the final step confirmed the hotel and so you could use the funnel feature um to understand you know how many users are progressing along the different paths which segments of users are progressing in different ways and also really dive deep into particular steps and say okay why did 20% of people people confirm their payment information and 177% um end up getting actually confirmed by the hotel where did that 3% go and what can we do to prevent that in our app um and then one final way to measure success is retention and we have a

variety of different retention metrics in flurry um one very commonly and new ret commonly used and new retention feature in flurry is called return rate which is essentially static retention and this feature measures the percentage of your users who return to your app on a given day week or month post install and we look at this value by cohort group um which means you know based on when users first started using your app and so this is a a very important metric for a lot of folks we're working with um we also have a rolling retention metric uh which allows you to group together a set of users by date and measure turn um as well as some other retention metrics and uh you know we wanted to just point to these here because this is a very important way um to really measure success in your user acquisition efforts in your engagement efforts um and everything you're doing how how long are users staying with our app one one thing

I'd like to point out and we actually have a blog post about this that's um posted up on the flurry blog is that um this is actually a pretty important thing to to measure um we did a we did a good cross-section of apps in an investigation um and we found that apps that can maintain over 50% of their users past month four have a much longer shelf life than those that lose that can maintain less than 50% um through month four um and so that that blog post is definitely well worth a check okay I'll link that to the show notes Mike if you can send that over to me I'll put it in sure can great um so some concluding thoughts um you need to collect the right data early on um don't and you know this goes back to the point earlier um don't let this be an afterthought um don't you know don't come to me two days before launch and say hey Mike we'd like to do this um if this is part of your thinking in

advance uh then you're going to have a much greater rate of return on this investment um than if it's an afterthought um so think about it upfront um you know the other thing is that use the data that you collect to know your most valuable users and then focus on acquiring those key users um you know that's going to differ from app to app app but in general you should know who your most valuable users are based on the data that you're getting um and then you know make it a cycle um conduct ongoing assessments of goals and progress um you know you may find that what you thought were your most valuable users to begin with or not um and the data could tell you that you know you may think think that anybody that makes it to level three is going to be my most valuable user but in fact it's people that make it to level five um so you know you can adjust based on that and um you know start uh letting the data tell

you how to build further into your app to further please your users um and so you know we we feel that if you can follow these key steps uh you will in fact grow very well so any questions mik I got a few and and for those who are listening live please put it in the chat if you got some questions but thank you so much Mike and Alex this was this was awesome the question that I have is when you're going back to the Quality metrics and you said you know you can say once a user hits a certain event they become a qualified user and you kind of see and that I forget what the slide is but how many events can we have that say Hey this is a quality user yeah go ahead no go ahead sure so so basically um the quality metric allows you to Define user quality your way so you know as as Mike mentioned Beyond simply tracking The Impressions clicks and installs and UAA enables you to measure the quality of acquired users

in that table um and you can measure based on engagement demographics inapp actions um any segments or funnel of your choice now within the current flurry user interface um you can only put in one custom quality definition at the moment we have a couple of others in there like um you know three and 7-Day retention um but for now you can only put one one metric however um because UAA is built right into the flurry analytics um interface you can actually deeply measure your required users um and analyze acquired cohorts any which way you want throughout the rest of the analytics product so just because it's not in that table you can get to any answer you want throughout the rest of the analytics product got it well was going to say it's also worth mentioning that you can have different quality metrics for different campaigns too so it's it's not just you know one quality metric across your whole app um as

you're setting up each acquisition campaign you can specify well for this acquisition campaign I want this to be my quality metric I want to know who gets to level three got it so it's campaign based yeah it's campaign based that makes a world and so for the for those who are listening or even watching UAA stands for user acquisition analytics within flurry the other question I have and I love this dashboard because you know yeah I've seen some other people's dashboard it's insane where you can measure but when you have a certain campaign that you set up can that drill down to personas too um you know that that I don't believe that it can I know that for like an event for instance you can see um how Persona shake out in in an event um so personas are actually available throughout um the dashboard that you can slice and dice based on personas um so you I can't answer that definitively right now but I can get you an answer

later okay great and I think Margaret has a question Margaret I muted you oh thank you just on that personas that's really interesting I've got a travel app and my f personas tell me I've got 55% business Travelers and zero Leisure Travelers so what's the difference between a business and a Leisure Traveler um the question is you have um a very high percentage of business Travelers and a low percentage of leisure Travelers and what drives the difference yeah um so basically you know the reason for that difference is the definition of the two personas is based on different apps and flurry so um business Travelers I see um based on apps that are typically only used for business travel um versus Leisure Travelers have different different apps in the basket and it really is just the math of how many sessions users are logging in each of those types of apps and do you publish that anywhere so that I can see the

sorts of apps that a Le traveler uses um we have we have information on the types of apps and um events and actions that can be taken in them but we don't share the list of names of apps um I'd be happy to send that along to you though so you can see what types of actions uh those users are taking I'd be really interested in that thanks if you can get saved huh sure cool Eric you got you got a question yeah I have a couple uh real quick uh first one is if uh we've talked to a couple other companies we do use flurry that's our main for analytics and stuff but we talked to some other ones and they talk about being Facebook preferred vendors these other analy does that mean anything do they have any advantage or do you guys have the same exact stuff but my assumption is it's just an advertising ID that's that's put into the advertisement and then you match those up once the the install is made is that

correct yeah so that that actually pertains to user acquisition analytics um and whether or not you can use user acquisition analytics to track um a Facebook campaign um and you know if uh if we're not currently a um certified vendor for Facebook um and you know but there are plenty of uses for UAA outside of Facebook itself and we actually have a lot of very large names that use user acquisition analytics um for the majority if not all of their um third party tracking needs okay and my other question was um update times there's particular times in the day when you guys update data I I look a lot on the FL website I didn't know if it's set in stone or maybe I I'll stop checking every hour it is it is not set Stone um you know based on the architecture of our system there's a lot of batch processing that goes on so we'll tend to um you know collect a bunch of data into one place and then run jobs on top of

that um to process that data and typically you know for a piece of data that comes in um it really just depends on the job that's processing it for when it's going to be available so as a for instance um if you're looking at your uh if you're looking at your dashboard and you want to know if you've integrated correctly um the quickest way to find that out is to actually log an event and go to your events page and the event logs um and typically within five to 10 minutes you'll start see the event logs rolling in um but now the job that actually breaks those event logs up and parses them out and shoves them into the other dashboards you know those can take um anywhere from you know three to four hours um and and this the same story goes for uh you know your new users your active users things like that um it it's job-based and it's pretty much just dependent on when the job is finished um a typical safe answer is that the delay

is going to be anywhere from three hours to six hours okay thanks no problem cool I think we got a couple more questions before we hit the big finish Margaret you had one question just a quick one I can see that my apps had a few errors and they're all coming through as uncaught is there any way to get get more information on where they happened or or what they could be um we we actually do offer crash reporting now is this on iOS or Android iOS iOS so um the the thing about Crash reporting on iOS is you know you could definitely use flurry's crash reporting um if you do not have currently have a crash reporting system in there um then uh you know I would recommend flurries for certain um on iOS unfortunately though you can only have one crash recorder um and so uh you could um if you've already got one integrated you wouldn't be able to integrate flirties um on Android that's not the case you

can have multiple crash reporters so um but yeah we do offer crash reporting option but sorry I've got I've got flurry it's telling me there's no crashes but it's telling me I'm getting errors are you logging errors I'm not the developer I paid someone else to do it so there they are that's the thing I've so don't talk Tech to me I won't get it but I can see I've got five errors Apple hour simulator iPhone 4 and iPhone 4S so okay what do I say to my developer let me let me back it up just a little bit um there is so there's a couple ways that you can catch err in general in in the SDK um you can specifically log an error similar to how you would log an event um and that is you know it's it's it's um kind of the most rudimentary basic way of doing it um you can also set an uncaught exception Handler which kind of logs a default error um if uh one bubbles up to the top and and it's not caught um or

you can implement the full-fledged um crash reporting option um and so it sounds to me like maybe you you're not using crash reporting and either the the developer has logged specific errors or has um allowed them to Bubble Up to the top and be caught by the uncaught exception Handler um in which case I would recommend going back to the developer and saying you know could you tell me where these are coming from um and if you know you're looking to take it a step further a really good idea might be let's integrate some crash analytics um so that we can see the exact details the only problem with an error is that you really get minimal information um because it's it's more of an event kind of a thing you're limited to about 256 characters and so unfortunately you're just not going to get a lot out of an eror the error structure was really um more geared towards quick short messages whereas

crash reporting is very comprehensive okay so I need to ask him to set up crash reporting thank you no problem well Alex Mike this has been absolutely amazing but let's go to the big finish what is your favorite mobile app Alex will let you go first thanks Mike um ladies first right so I have a lot of favorites but um um I guess you know something that I miss in my daily life is a little bit of creativity and and the app that I've been using the most when I'm on the road or you know on a plane traveling and looking for something to do quickly on the subway is this um sketching app it's called the tyu sketch app um and you can make uh beautiful drawings you can do watercolor painting um and I really just like it because I find it kind of fun and relaxing I like that never heard of that what is it called again Alex it's called tasu sketches tusu sketches okay I'll have to get that from you I gotta

see some of these sketches Alex yeah I will not be sharing my sketches they're really not not skillful but they look better than they otherwise would if I did them by hand that's awesome Mike you got one yeah I do um so right now I find myself using my fitness pal a lot um and you know one of the things I really like um I'm kind of on a health kick um is I like the fact that it it integrates a lot of external sources almost a a huge ecosystem and puts it into this one view um that you can kind of look at to get a almost a dashboard view of what your you know your daily exercise and calry intake and things like that has been um and I think it's a really powerful idea um and I I kind of Envy them in that position I think it's a great place for them to be great Alex just for fun give us a half-baked idea for an app something you wouldn't mind someone else working on sure so I'm at the very beginning of

a very stressful New York City Apartment hunt and I'm going from website to website to realtor to I mean I'm all over the place and if somebody would just create a Tinder for Manhattan apartments that would make my life easier like it Mike um yeah so um you know I'm a father and I got an eight-year-old son and he just finished up a um a baseball season um and we would get to the baseball field and I've got two other kids and everybody be like we want some big leag chew which is a gum and if you're not familiar with it big leag chew comes in these pouches that are about the size of a child's head and you tear the top off and then it it's made the look like chewing tobacco which is really the signal you want to be sending your kid um but at the end of the game all the kids are asking for Big League Chew and we get to the end of the game and there's like seven or eight kids that have like taken

two mouthfuls of Big League Chew and they've all got this Big League Chew left over um and so I would love to see a sharing app so that you could get to the baseball field and log that you have big league CH and be able to share that with other people so that all the parents are not going out and buying a big league ch very Niche app Mike I like that what advice would you have what advice would you give anyone looking to build a mobile app Alex you want to start my my very quick advice would be track from day one um I think that you know even even though analytics may not be the the main thing on your mind when you're launching your app get something in there and even just passively track your data uh if you're not looking at it for the first six months so be it but have it so that when you have a question and you want to make a change in the app or know who are your valuable users you have that in

your back pocket and then you can use it it's great advice Mike yeah I mean that's that's my answer too okay um you know you're GNA put all this time and effort in heart and blood and sweat and tears into an app um and even if it's just for fun um you know you never know you might just be the next uh Flappy Birds um and if you are and you don't have a view into your Analytics um you know you're you're going to be hampered in in what you can do and how you build things out going forward so that's mine too is track things from the beginning yeah and you can't iterate if you don't have any data because I I mean I did I made this mistake and so I'm kind of going in blindly revising my app a little bit but not not anymore Alex Mike thank you so much if the audience wants to learn more about flurry and say thank you for coming on where would they go um they could obviously go to flurry decom um or they could tweet um at

flurry mobile um and yeah that'd be a great way to get in touch with us great Alex Spike thank you so much for taking the time to do this thank you Steve thanks you thank you all for listening we'll see you at the next chat and that's a wrap thanks guys that was awesome