Measure people instead of devices

[Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] [Music] hey everybody welcome welcome to experience league live glad to have you with us uh this morning this afternoon good evening good night whatever time frame you're on uh glad to have you with us this this fine day and uh happy holidays and we're glad to have you here on experience league live uh today we have a great show we've got three guests that we're going to be talking to as you probably saw in the lobby that was running there that today's uh show is measure people instead of devices and so we're looking forward to that and again this is all brought to you by experienceleague.adobe.com so if you don't know anything about experience league then head on over to experienceleague.adobe.com and get everything self-help we have the documentation we've got tutorials and free courses and we have the community so you can talk to your peers and uh there's lots of good fun stuff over there for you so hopefully you'll come and over there and join us on experience league so awesome so without further ado let's meet our guests first we have matt freestone a principal product manager hey everybody i don't know yeah you are kind of there it's kind of kind of odd but i'm here so we'll look through that you shouldn't have put that that film in front of the camera then exactly but uh principal product manager yeah that's right yeah principal most people aren't that actually excited to meet the principal but there you go principal product manager so um tell us about in 20 seconds tell us about your day what does a principal product manager do so primarily i am working with an engineering team to build products that customers love so i work a lot with customers to learn about their needs my primary area of responsibility is around cross-device cross-channel stitching so it's all about measuring and improving the customer journey good stuff all right so bring in our second guest we have alex ivana also product manager wait a minute alex you deserve also the uh that was a bigger crowd than yours matt i'm sorry alex is more popular than you what can i do yeah so alex you're also a product manager as well is there anything else did matt get it right is that what you do as well or is there anything you'd like to add about your normal day-to-day work well he got it right i might still add that our primary job is to be sure that our customers are getting what they need and that we are also close to the engineering team so everything goes smoothly and our products are successful nice nice yeah that's great trying to trying to juggle everything keep everybody happy i like it good luck with that so all right okay and then certainly not least uh we have nils angle who is a principal solution consultant another principal yay oh hang on this is there you go tell us about a principal solution consultant and what you do uh yeah so i'm um primarily customer facing so working with the customers who are using the technology that matt and alex are building and really trying to make sure that they understand what the capabilities are and that they're seeing the value and getting to use it and apply it against their data awesome okay making it real making it real yeah okay i like it okay if you guys haven't noticed certainly you noticed when nils came on but uh if you haven't noticed yet we've got our yeah we've got our holiday sweaters on here you know i i will say that i i have like holes in my body you can't see here because you can hardly with my green screen you can it's hard to uh miss green around christmas time so you can see right through me to that wall and matt you've got uh what i understand is an award-winning yes that's right this was one of our early christmas parties when we moved into the new building for adobe down in lehigh and uh my wife and i had matching or mostly matching christmas sweaters and yes we won first prize that year yeah don't get it wet i mean while we no if you drool then i i'm thinking there's going to be some electricity there it's probably not good to do snow angels with it either that's right alex so show us a little bit more of yours we're waiting a second for nails here well i have my naughty reindeer oh nice this is the one santa uses so yeah that's right uh and nils you gotta show us this one in its full glory yeah thank you all right i'll give you the full view here oh my so this is uh homemade if you can't tell and um yeah i don't know we showed up at a christmas party where nobody else was wearing ugly sweaters so that was interesting oh wow yeah if only you could have come in you know like on some wires and flown in that would have been have you been invited back no surprisingly that's so weird that's very weird well let's let everybody get to know you just for you know another minute or two um let's go through your your uh your fun facts they might have might have read them on the lobby part there right before the show but matt you are a photographer i know you're an avid photographer and i i follow you on the instagram so the instagram on the instagrams and so uh what is what what is your favorite thing to photograph probably my favorite thing is landscape and nature but i've also done a lot of photography for like local theaters and productions so like um arsenic old lace at a local theater or um some local choirs when they've done concerts i've done photography for them so i really like event photography as well okay you need those action shots at the choir yes yeah good free freeze the action yeah awesome okay a lot of fun stuff and alex you are a traveler uh said that you uh like to travel and get out and about probably been a little hampered in the last couple of years on that but what is your what is um what is your favorite place that you've ever traveled let's go there well definitely to be florence in italy i mean that was the place i proposed my wife so it was a magnificent holiday i will not go there without any hesitation again i definitely recommend it yeah but you're yeah yeah and that's not that far or maybe it is but i mean you know from you know relatively speaking from romania or where you're joining us right it's it's probably just a hop and a handspring oh my goodness yeah that is nice yeah yeah florence florence is awesome okay i love it and then nils you are um a spartan racer right you've done uh how many you said over 50 is that right spartan races yeah this year i crossed a 50 race threshold wow that's crazy so um yeah so where i was going to ask you where is the what is the hardest one you've ever done and and what was it about it that was so hard that was difficult hardest one um i forgot what year was i think 2015 or 16. um i did lake tahoe it was the world championships for um spartan racing and uh it was really hard um i expected it was going to be warm because it was in california um it ended up being 30 degrees with ice on the ground close to 16 miles of running and i think probably the hardest part was that it was cold and that the elevation was significantly higher than uh new york which is where i'm from um and i didn't have time to acclimate to the elevation so i really struggled with you know both the temperatures and the elevation there was a swim in about 54 degree water as well oh man and it was windy when you got out so it was a rough race wow that's craziness okay while you're talking about that and we're talking about obstacle courses so i have to know if you've ever tried out for american uh what is it called gladiator ninja warrior ninja ninja warrior i have not but i would i don't know that i'd want to try out but i'd love to play on their on their uh you know course for sure yeah nice yeah well my my foray into american ninja warrior is watching it from the couch so same here so far it's it's yeah it's quite it's it's it's pretty hard you know to get up and get to the fridge and get some yeah so awesome so thanks guys okay so let's dive in we're going to talk about cross-device analytics very excited about this um you know and i'll let we're going to turn it over to alex here for a second but um super excited about this because people for so long you know have have had to kind of you know we we call it unique visitors and things like that but but at the end of the day we've been dealing with devices and and now you know these days obviously with people having so many devices then um we really want to know more about people so very excited to hear you guys i don't mean to give away the ending there so let's turn it over to alex and we should be able to just go here to your screen share hopefully there we go and so alex why don't you take it away for a minute and uh and matt and you can jump in as well and have you guys kind of tell us a little bit about like what is so cool about cross-device analytics inside of adobe analytics and how that all works thank you thank you hey everyone i'm very excited to have the opportunity to present cda just to answer very quickly why it's so cool it's because i'm sitting like alex i'm not sitting as my laptop right now so that's what customers want to see they want to see the person they want to target the marketing campaigns they want to understand the truth behind the activity so that's why we built this product a few years ago because our customers said we are living in a more modern way in a more modern world and we need to understand the person not the device so i will just give a bit of words about cross-device analytics basically in analytics any customer who goes analytics and wants to do reporting it will see those visitors basically it will be a cookie a device you name it and once you purchase cross device analytics you will change the reality to a person's centric view by stitching data across all those devices for example i have a laptop and a phone and i identificate on both of them i want to be sure that the customer can understand the truth about what i did and this is possible by choosing two stitching options which i will go in more detail in the next slides meaning field-based stitching and private device graph just to give a bit of insights on what this product does we are processing more than 4 billion events a day from our customers we have over 69 paying customers and we are geo distributed in across all regions just to get a bit of understanding the scale of this product which is growing year by year quite a lot so let's understand what analytics does right now we have here some visitors you will see pictures with devices but yes especially their visitors cookies or devices and for example isabel has those devices and she browses differently the search box product and eventually she will purchase well the customers won't understand what was the real uh desire and what made isabelle purchase because she they can only see what they did at the beginning so they searched for that product and then they purchased moving in the stitched world where we stitch all those events for those devices we understand the full journey of isabel meaning she started with the search box moved to the product review asked the community which was the main driver for the purchase then she moved back to see the product again on the website and she purchased and now we understand like a customer i can understand where i can put my money if i'm talking about marketing campaigns or any others to be more in-depth if we take an actual use case based on what we just described if we take the same path meaning search box product review as community product view and purchase so we are now talking about the devices if you look on the left side we are talking about the vista based only about what it is does right now and on the right side we are talking about cross device analytics if you just make a small comparison the thing that will win the money in the analytic is what is the search because isabel went first on the search then she eventually moved on the phone to do the purchase so the search was the reason why she bought that item but that's not a reality because if we go in the cda world we managed to understand that the key driver was the fact that she went to the community has the question as we all do when we want to buy a phone for example we start to browse different uh blog posts and to understand what that phone can do and then she moved back to purchase and this has a great value for customers and basically gives you the reality because you need to understand the real activity on your website so that's the way your customers they have many ways to do it but that's one way to to actually use it perfect you've heard that can i ask a question sure let me just jump back for a second to that last one so in this case where where where the um attribution is is shifted maybe to to the question that was asked in the community i guess that means probably that you are populating some some kind of like an evar some kind of a variable then that that shows whether they searched or whether they reviewed or whether they had the question am i understanding that correctly because we're those and then and then maybe we're looking at the latest thing that was populated into that variable and without the cross device analytics that it was never populated there in the middle and and then with cross-device analytics we see that it is populated that way am i thinking about that right yeah i think the way to the example in this slide is using last touch attribution so what we're talking about is what's the most recent thing the person did before they viewed a product or they purchased and the important thing here is that when you're looking at legacy visitor based analytics the the phone and the desktop device and the laptop are separate entities right so if you if you want to understand in the legacy analytics world what was the last thing the last test last touch attribution what was the last thing that happened before the purchase was made the only other thing that happened was the search box at the very beginning right because it happened on the same device right so the attribution for that visitor is that the search box got credit for the purchase and same thing for um the the product reading the product review was the last thing that happened on the desktop before viewing the product view right so it's three visitors we're talking about here with three different attribution one where somebody searched and then purchased another visitor where somebody read a product review and then and then looked at the product and then the last one where they asked a community question and did nothing else so it's it's like fractured that customer journey into multiple pieces whereas if you're looking at it from a person perspective with cross-device analytics all one person isabelle did all of these things right and the last thing that she did before she viewed the product and before she purchased was she asked the community question right so it's a different mindset you're you're not looking at attribution on individual devices you're looking at attribution for the entire customer journey yeah awesome yeah and maybe i if you guys don't mind me adding one more comment which is yes we're using a last touch model right now but we could very easy easily use like a linear model or even a participation model that would then give credit to all three of those touches for their influence on driving that conversion yeah that's right that's right nice thanks alex thanks for letting me thank you for a second yeah okay perfect thank you guys uh you heard about the stitching options at the beginning of the presentation so currently cba has two options in order to do stitching and as you can see on the left i'm talking about field-based teaching on and on the right private device graph what the difference between them is first of all field-based teaching does teaching based on a field like dogs just said anivar for example on a prop so the customer comes he said okay i want to get cda and i store my identities when i cut when a certain visitor logs in in a bar perfect we'll be using that one and the investigation will work perfectly another interesting thing about this option is it's fully transparent what the customer has in that field it will be the source of truth for everything we do across the journey and of course we can do backfields what backfield means if for example i have a customer in analytics for like two years and right now he decides to uh get in there like his ultimate and get cda he has the option to start fresh like it has in the private device graph where backfill doesn't exist well in field-based teaching if he has the historical data we can bring it over and stitch it just for the customer to get more insight with what happened in the past not just go from here to the future and of course those two options can only work if users are authenticating so it's a deterministic method that needs some input from the users for example myself i will go on the website and i will login then i can understand that my device belongs to me same goes for the private device graph the main difference between them is if you have multiple fields for example i store my email in one field one er i store my crm id on another email and so on in other fields sorry then i want to use all of them you need to go to the private device graph option because it gives you the option to combine them based on the graph algorithm the algorithm is a very straightforward thing which doesn't give much transparency but at least gives you the option to get advantage of all your fields and of course you still need identification that's a required thing no matter what you choose for keywords cross device analytics because that's the world we live we need to respect our customers privacy we can just get there and uh take advantage of their um those informations moving a bit more you have oh sorry i'm not i'm not talking fast enough but so yeah i just wanted to i just wanted to comment on that um but yeah we need to know who people are obviously as they authenticate um and you've got a you've got a a link down there that you know it's prob probably pretty small for uh for people but but if they if you want to know more about not only um you know customer device or cross-device analytics but but also get more information about set customer ids and and and setting that up um yeah this will be my my mid you know uh mid meeting plug here i suppose for experience league because you can go to the documentation there and you can see uh you know how to how to run set customer ids what information you need to provide in there and how you hook that up you know to to the login and those kinds of things like that so that they can um you know make that private degree uh private device graph work i think we even have a video on that so yeah you know feel free to go over to experienceleague.org.com to get more information on um setting these up uh including again the you know the the format that you need set customer ids to to run in so anyway just a little plug there for where you can get more info on that thank you maybe i'll maybe i'll follow on to your comment there doug and just say so really uh both of these accomplish similar goals field based stitching and private device graph they have slightly different capabilities but in the end um the the main difference is the way that you communicate to adobe who the person is right in the case of field based stitching you stick that identity in an evar or prop in the case of private private device graph you call set customer ids and you send the identity or identities as alex mentioned because there could be more than one type of identity you send those uh to the experience cloud via the set customer ids call then the graph takes care of relating all of the identities together so from an implementation perspective that's the that's the main difference in one case you use a field in another case you use an api call to send the identities to the graph awesome thanks thank you great insights yeah i think it's always important and nils will show us a bit what you can do the implementation can take a while and having that in mind i can give you a bit of insight what actually cpa means so i have a question we have a question coming in that i think is an important one i'm going to pop it up on the screen here and is there anything that can be done about you know not logged in non-authenticated users how would you do this without any id from from the customer is there anything that can be done uh for uh for those people who aren't logged in because that's gonna be yeah you know some of the end users so i'll take that and and i think bills could probably comment as well and alex but so both of these require that you know who the end user is in some way it doesn't necessarily have to be a login that's probably the most common way to do it you could use like a click through email address for example if somebody has read an email address and clicks through and you store that email address in a proper and evart or you send it to the graph that's another way that you could do it but but all of these both of these capab forms of stitching require that in some way you know who that person is in the past there have been ways to try to identify people like um as an example we had or we have a an adobe co-op you may have heard of where we use third-party cookies to try to help figure out who a person is like if somebody logs in on one website then on for one of the companies who's using the co-op and then they go to another uh brand that's a part of the co-op and they don't authenticate then we could use third-party cookies to help figure out who that person is without them actually being required to log in but as we all know third-party cookies are going away that's just a thing of the past the world of being able to rely on third party cookies to help identify end users is just a thing that you can have to work around it doesn't it's not going to exist anymore very soon as soon as google pulls a plug on third-party cookies um all those type of methods are done so so really the important adobe strategy for identity is is you should get to know your customers and and build relationship with them such that they're willing to identify themselves in some way yeah and i appreciate that question by uh by arun and he had a follow-up one which is you know what about email being pii right so do we do we just uh make sure that we hash that or or what is what are your thoughts there yeah i can't i mean the determination of what your company views as pii is something that you're going to have to work with your attorneys on but if it is a concern yes you could definitely hash the email and and use that as an identifier yeah thank you arun i appreciate those thank you for the questions for our team yeah okay so yeah alex alex uh go ahead continue thank you was a great comment so a few a few items that i would like to touch is because everyone asks and customers are very curious about what how can you actually get cda so the first thing is crosstalk scientist is part of energy's ultimate subscription so if you get into the ultimate subscription then you automatically have the entitlement to your cda but this is not added automatically because as you heard from matt and nils you need to specify a few things like fields or you need to implement customer ids and those things require some help it can be just automated so in order to get it once you are entitled you need to submit a request via customer care and then customer care we take those information like the field for example from you and will take in touch with engineering team to provisioning which typically takes between two or four weeks depending on how complicated the implementation is it can be even faster or in some cases we need to talk more and understand the use case in order to make it more efficient but basically once everything is submitted and customer care can touch base with the engineering things will get in a straight line another important thing you heard about stitching but analytics can organize data in different ways you have the org level and you have the report suite level well what stitch is doing is stitching events multiple devices all together in a report so we can't combine multiple reposits that's a different use case which is not covered in this product and matt can give us some insights about another product who can do things like this and of course once you want to do reporting everything is set up everything is working we have the green light from engineering and customer care then you have to go to analytics workspace create a virtual report suite with stitched option and then just play with it because it offers a lot of capabilities and of course you can see full details in the experience cloud documentation with everything that you should know all right and those i think yeah i think nils is going to show a little bit about the virtual report suite here in just a minute correct yep so i think one of the key points from alex is there is that with stitching's done inside a report suite so you really should the report suite needs to be cross-device in nature right you need to have data from desktop and mobile in order to stitch desktop and mobile together so we sometimes sometimes we refer that to that as a global report suite which is uh it doesn't necessarily have to be global from a geographic perspective but it needs to be global in that it contains data from multiple different types of devices if those are all within the same report suite then cda can be used to do the stitching great yeah very important there awesome okay alex any any more words of wisdom for us yes everything will be given by nils in his awesome level you go ahead and share and we'll get we can switch this over to you that's that's your best words of wisdom yeah that uh please handle the next part uh nil yeah so let me put this over and look like this all right doug there is another question i don't know if we want to address it now or two sure sure let's uh here let me pop it up on the screen here sora's can we let me see how can we stitch uh customer id when the user navigates from a native app and some link takes them to the website which opens up in a mobile browser ooh that's a good one yeah that whole uh from app to mobile browser bills yeah sure i mean the reality is we just need to capture that id right like we still need that common id that exists across those data sources so that person needs to be known or authenticated somehow we have to capture whatever id we're using um in the mobile app and then when they're in a mobile browser we would need to it either needs to be passed to the mobile browser when they click on the link from the mobile app um and capture it in an eval or a prop in a mobile browser um or in a mobile browser um and to be clear the analytics data from both the app and a mobile browser would need to be in the same report suite for that to work yeah yep yep not sure if you want to add anything to that map no i think that's good and there's one question before that as well doug oh okay sorry i'm slow here okay yes so you want to hook your credit ready yeah cross-device analytics work if customers only log in at the end of the visit um or will that visit be split into two once they log in good question so stitching has kind of two i would call parallel processes we stitch the data as we go so as soon as we know who that person is then we begin stitching from that point forward but there's another process that's part of stitching that's called replay and today there are two options we can we can replay once a day with a one day look back or we can replace seven days with a seven day or once every seven days with a seven day look back and the purpose of replay is to go back and help uh if we have learned in that within the replay window who the person is then we go back that period of time and we restate the data within that window as belonging to that person so one of the key advantages of cross-device analytics is that ability to go back and replay some of the history to try to bring in that pre-authenticated behavior awesome thank you matt let's let's move for uh for now let's move uh let's go ahead these notions appreciate yeah nils why don't you uh show us some stuff here as well happy to show some stuff okay so um you know once the data has been prepped uh so so basically you know the the report suite has been defined that the stitching will be based off of and the appropriate ids have been defined and submitted to the to your csm and the stitching has been applied then we can go into analytics and we can actually define the virtual report suite that will be able to take advantage of cross-device analytics so i'm in i clicked on components and i'm in the virtual report suites interface you can see i have a bunch of them here and i can just click on add to create a new virtual report suite when i do i give that virtual report suite a name i'm just going to call it test because i'm not going to actually save it i would give some sort of description for that virtual report suite and then i need to define the source of the virtual report suite and this is where you would point to the stitched report suite which would have been created as part of the process of reaching out to your csm your customer success manager and so you would point to the appropriate source you could then do anything that you can within a virtual report suite so define a specific time zone that you want to use you can even apply segments against the virtual report suite to only include certain data into that virtual report suite and then lastly there is also well a little bit more than lastly there is the visit definition um tab within the visit definition tab you have the ability to enable report time processing and cross-device analytics relies on this report time processing capability when you enable that it then allows you to define a standard visit timeout which is 30 minutes or you can define based on any kind of preset or you can even define a custom visit timeout if you want i'm going to leave mine at the standard 30 minutes i can also define that key event may start a new visit so think about maybe an app launch event if you have app mobile app data in here an apple launch event could kick off a new session or a new visit if you wanted to but the key thing for cross-device analytics is to turn on the stitch user visits across device across devices this is basically going to now glue a person together based on that id that we've captured and store the data by that level and really what ends up happening is where we have an id for a person will store the data at that id level even if it's just if we only have one id one device associated with them where it gets interesting is if somebody has become known across multiple devices we're now storing this data at that person id level or that customer id level and that now allows us to stitch them across the various devices within here now there's a bunch of other options you have within virtual report suites for example i can say i want to bring in all of my variables all of my evars my props my custom events automatically or i can say i want to select only certain ones that come across into this virtual report suite in this case i'm going to just have them all pop over if i were to click on save now i would have a new virtual report suite that's using cross-device analytics and it would be called test once you have that virtual report suite built you can then go into the analysis workspace interface and you can start performing analysis on this data you can see i'm in a virtual report suite that i've called cda vrs that stands for cross-device analytics virtual report suite you could have called that anything you wanted and we actually have a template that we provide out of the box that allows you to perform analysis and has a whole bunch of pre-built types of analyses i've taken that template and just refined it down to just some key areas that i'm going to be focusing on today but you just know that you could you know open create a project off of that template and you'll have a whole bunch of pre-built views that allow you to understand you know how cross-device analytics is stitching people together so some of the key things we're looking at here in this cross-device analytics report suite i can see that 59 of the people are unidentified that means that we did not capture any type of id for them their customer id or email or whatever it might be that we use to stitch them by 41 of them are identified which is kind of interesting i have a donut visualization that's allowing me to get a breakdown of how many people are unidentified how many people are identified on one device and how many people are identified on two or more devices within here i also have some segments that have been built and these come out of the box which are how many people are there that use mobile devices only how many people are there that use desktop devices only how many people are there that use both mobile and desktop devices and then i also have some other filters here such as how many people use one device two plus devices how many people convert from phone to desktop so on and so forth so there's a lot of different segments that you can apply against this data to start understanding who these people are as we go on further down here i can actually start measuring my identification rates this is kind of interesting as you start thinking about maybe you have campaigns that you're trying to use to drive people to authenticate to log in well you can now start understanding if you have higher authentication or identification rates at certain points within um within you know that campaign for example in addition to that we can now start breaking it out by device type so on desktop how many people are there total and the people count here is a new metric that's going to come out of here and this one encompasses both anonymous visitors so this would be people stored at the cookie id level or the experience cloud id level as well as people who have been stored at the customer id level level and this may be a person who has multiple devices within here so that's what that people metric represents i can see what percentage of them are identified across each of the device types and you can see certain device types have higher authentication rates than others and there's another is that sorry is that is that people metric then is that out of the box in that uh virtual report suite that has been that has been set up like this yeah are you saying are these do these visualizations come out out of the box that's in the template i know that's in the template i'm asking about the i guess the people that people metric oh yeah yep that comes out of the box yeah this is automatically a new new metric that's available for you in this in this yeah and in fact in fact there is no no visitors or unique visitors metric in this virtual report suite that's interesting yeah so the people metric comes out of the box it basically replaces the unique visitors and it's it's a i think i'd like to think of it as a unique visitors on steroids metric which is basically it's it's better than that right because it's going to identify a person where we can and an a anonymous visitor where we can't we also provide another metric within here called unique devices which which is pretty close to unique uh visitors and that would be a count of the actual kind of experienced cloud ids or cookies okay i'll throw in there that this concept of people is spread throughout this virtual report suite environment for example within segmentation if you're familiar with creating segments in adobe analytics you have hit visit visitor containers as part of a segment but now it's hit visit person right so again we're swapping out that concept of visitor with a new concept of person okay another interesting method that comes with this visualization is identified people for example if you want to see only the identified people you want to get a hint of how many i managed to identify you can see that out of the box with the identified people metric and you can understand where you are at a certain point is nearly just showed yeah yeah nice right and yeah that was all right that was right up there in the um in the segment drop down right that you showed nils correct yeah yeah the hit visit person yeah okay great and as we continue on here right we can do other interesting analysis for example we might want to understand how many people are there in this virtual report suite that only have one device how many people are there that have two three four right so we can actually get a breakdown of not just how many people are there but how many people are there that have certain counts of devices which is an interesting view into the data we also using event visualization can now start understanding what percentage of the people in this data set are actually uh across device right multiple devices within here so this becomes interesting because it's not just at a device level but it's a device type level so what i'm looking at here is how many people do i have that use desktop how many people do i have that use mobile phone what's interesting about that is you can see that there's a cross section of people who use both desktop and mobile phone and then i also added in a tablet within here as well so i can actually see the breakdown of people who use tablet and in this case there are certain people that use tablet mobile phone and desktop at any point if i wanted to i could easily create a segment of those people to understand how they look compared to for example maybe just people who use desktop for example nice yeah so that's a really powerful way of being able to start understanding the behaviors of those people that use various devices as we continue on we can actually see the flow across mobile device types right so as somebody for example comes in for the first time using desktop we might want to understand what percentage of them stay on desktop what percentage and move on to mobile phone and what percentage of maybe even come back to desktop or some other device a really interesting way of being able to kind of see that cross-device behavior and understand how people move across devices if we haven't seen it sorry if we haven't seen it scroll back up a little bit um the the whole other yeah i mean we'll just kind of we didn't really say it in here but this is a mobile device type so other means desktop right this means non non-mobile so just point that out we've been not not dancing around that better just we haven't pointed that out so i just wanted to say that cool yeah good point i probably should have said that thanks for saying that doug um great okay so as we continue on here right we can even use the fallout visualization to now start really getting counts and seeing how many people there are that actually go from the one device type to another in this case i wanted to understand how many people start on a mobile app and move to a desktop app now what's interesting about this is this can be done at a person level or it could very easily changes to be done at a visit level are there people who are starting a process potentially right on the mobile app getting frustrated not able to complete it and then pop into the desktop to actually complete it right i'm guilty of that for sure and then the next event we wanted to look at is some conversion event in this case it's an an order event but a really interesting way of being able to analyze that to see you know is there potentially something in our process that people don't feel comfortable completing that in the mobile app but move to the desktop to complete it now going back to the slides that alex showed earlier if we were to analyze this in standard analytics outside of the cross device analytics report suite it would look like some some people came to the site and potentially started a process and application process on their mobile device and gave up that's it right and you'd see that some people came to the desktop started an application and completed or sorry they started an order and they completed it what the part that you're missing without having cross-device analytics is that people started here for whatever reason they moved from that device to the desktop device and then they moved into the online order now we could do some further analysis on here to start understanding what were those pages that they gave up on in the mobile app or those screens that they gave up on in a mobile app a really powerful way of being able to analyze that we can even very actually understand yeah in addition we can go even deeper right from a cohort visualization perspective i want to understand people who have a mobile visit how far into the future do we see them having a desktop interaction right and this is just starting to kind of understand what those cohorts look like this also allows us to start in understanding engagement of a person right so engagement is a very powerful thing depending on the industry that you're in i want to see people who come in for example this week how many of them come in next week in standard analytics without cross-device analytics right i might come on my mobile device this week and i might come across on my desktop device next week i look like two different visitors this cohort visualization wouldn't show that but because we're in cross-device analytics i can now actually see regardless of device they had a visit on this day and then they had another visit on a separate device next week we can actually see that engagement across week over week which becomes really powerful here's another example that kind of takes us to the next level it uses a very powerful metric that we could probably spend a whole other session on called the life cycle index and really what it's doing is it's using our attribution capabilities our attribution iq capabilities to give credit and typically we would be doing this at a uh like a marketing channel or marketing campaign level but what we want to understand is in this case we want to give credit to devices based on when a device is the first device that somebody interacts with on their path to a conversion so what's the first device somebody interacts with i like to think of this as the acquisition device right so what's the first device somebody interacts with when they on their path to a conversion what's the last device that they interact with on their path to a conversion and we're able to take those two metrics and build a life cycle index off of that that can then easily be visualized in this visualization over here that allows me to glance at it and say devices that have a bar that move to the right are better from an acquisition perspective getting people uh to the site or to the or to the app for example bars that move to the left are better from a closing perspective or the last touch perspective these are this is where the actual conversion occurs and this becomes a really powerful way of kind of understanding that at you know kind of a device level if you will now we're showing you a lot of things that are showing you how people are navigating across devices and kind of proving that we are um we are kind of gluing a person and people together across devices but what's most powerful about this in my opinion is all of the other analysis that you can do all of the standard analysis that you do measuring your marketing campaigns measuring how people navigate across pages all of that analysis that you're doing today in adobe analytics becomes that much more powerful because underneath the hood we're storing this data at the person level wherever possible which means that this data is that much more accurate for people who are becoming known across devices yeah so again that's the most powerful statement right there right do all of your when when we say when we say hey what are your favorite use cases uh for cross-device analytics it's like everything it's all of your use cases because this this this paints all of your use cases that whatever it is whatever your you know whatever your conversions are whatever your goals are it paints them across that uh that that tapestry of of all their devices so that's that's the really powerful part isn't it and and sometimes it takes a little bit of um uh getting used to in a way i guess that everything in in this environment is a person what a person-level thing like even the concept of a visit right if if i start a visit on a mobile device i can finish that visit on a different device in this world it's if it's within the minute timeout or whatever i've specified for my visit definition so segmentation metrics um visits they're all now person-based concepts yep 100 yeah that's so cool i mean but but you're right come on when you say i think that you'd kind of have to almost get used to that if you've been doing uh you know if you've been doing analytics for years or something like that you're yeah you'd you wouldn't think that uh it's a little bit mind-blowing right that you could actually start a visit on one device and complete the same visit on another device and so yeah it's something to get used to but again uh just so much more powerful and and accurate for knowing what is going on with your customers right yeah so cool so cool all right so i i'll add that we've had a guest person answering questions for us anka balanelle who's the engineering manager for cda also in romania she's been helping us out answering a couple of questions so thank you anka for your help there yes appreciate it thank you so much well that is awesome guys i know um we could probably go on and on about this and everything but i think that's that's a great um you know summary of what what is needed what you know kind of how to get it going appreciate that all that great stuff from alex and matt and and mill's showing us how to uh you know set up that virtual report suite and then some of these great reports um the whole cooking show approach here i happen to have this one right here in the oven all completed so um so that's that's very helpful i hope that everybody um you know is is seeing what they what they can get out of this great feature of cross-device analytics before we leave though before we leave you today we have uh we do have one more segment for you wait for it [Music] [Laughter] [Music] unrelated cool tip yes we have our unrelated cool tip of the day and uh nils i believe that you have our unrelated cool tip for today is that right i do yeah this is gonna be my second demo actually for the day yeah that's right yeah let's see let's see demo number two yeah here's what i'd say you know my sweater has nothing to do with what this cool tip looks like the tip is way better all right so i'm going to give you guys a cool tip on how to make a healthy tea this is something that we actually my wife and i saw at a restaurant that we were at actually very recently um and they happened to give us all the directions right in the menu of how to make it we had it it was delicious so we've been making it ever since so healthy tea how do you make it so for starters you need an orange um take a fairly thick piece of orange i typically would cut that into three pieces pop it into your uh your mug i only have a glass here but pretend this is a mug and the benefits of orange and actually as we're while we're at it i'm also going to be adding a piece of lemon i would also cut that into about three pieces pop that into the drink benefits of those are they both have vitamin c in them which are good for your skin for healing on your skin and also an immune system booster on top of that we'll take some ginger so i've cut up a couple pieces of ginger not too not too big but taken those pop them into the drink those have anti-inflammatory properties they're really good for calming your digestive system and a host of other things but um plus they taste really good i think um then next one take a piece of cinnamon just one of those little cinnamon sticks pop that in there right cinnamon is also an anti-oxidant and also an anti-inflammatory among a number of other things and it tastes really good um we have some i'm just waiting for you to put something in there that just tastes terrible but i'm hoping that everything it's delicious in my opinion uh so uh mint so mint i typically would rip this into about three or four little pieces mint has vitamin a which is um well it's good for you uh but in addition to you it helps with bad breath if you have it um and then also sage so again once the one leaf or two leaves of sage but depending on how big they are uh again i would rip that into about two or three pieces that has oral health benefits also helps with brain function so pop that into the tea and then lastly extra i'm gonna have to add some extra leaves then on the brain function the last face would be uh some honey honey is great so you would just pour a little bit of that into your drink that has it's good for your heart health it's a cough suppressant and it's really good for allergies if you can find honey that's local to where you live um it can actually really help with with allergens in that case and the last thing water which is good for hydration so pour some pour some hot water in here and you've got yourself a drink i let it steep for maybe i don't know two minutes stir it you're off to the races don't add too much honey it's really sick wow sugary nils i like how you brought that back to your off to the races [Applause] oh awesome thank you great unrelated cool tip i love it oh man awesome thanks guys so this has been super fun hey i really appreciate you guys coming on to show us this great great feature uh in adobe analytics of cross device analytics so thank you to you guys appreciate your time thank you to everybody who is watching and those who have commented and if you're if you're watching the recording of this uh then then thank you for watching that as well uh again this is all uh brought to you by experienceleague.org.com uh we will have this recorded um and and showing it on experience league as well as again the pieces of this so that you can get to know how to set it up and and those kinds of things all available to you on experience league so thanks everybody for coming today and we will see you next time bye all thank you you

Channel: Adobe Experience Cloud

Views: 1,399

Keywords:

Id: tQXJ7eA4zWY

Length: 62min 54sec (3774 seconds)

Published: Wed Dec 15 2021