PurePerformance - 041 BizOps, Digital Customer Experience Monitoring and Digital Transformation

Episode Date: July 31, 2017

It sounds like 3 buzzwords, But there is more than that. We were intrigued by the Digital Mastery & Joy ( https://info.dynatrace.com/apm_wc_panera_na_registration.html ) webinar Klaus Enzenhofer @kenzenhofer ( https://twitter.com/kenzenhofer ) did with Panera Bread. In his introductory statement, Klaus cited a recent study from IDG on Digital Customer Experience. The biggest challenges are data silos, poor data quality, redundant data, and missing coordination between departments that manage the individual digital touchpoint channels (Mobile, IoT, Web, Physical, …). In our discussion we find lots of parallels between the problem that DevOps tries to solve and which challenges digital transforming businesses face: Silos! Disconnected Silos! But instead of Silos between Dev & Ops its Silos between your Business Teams that are all strictly focusing on their slice of bread (to reference some great stories from Prashant Karre, Director of Performance Engineering at Panera)Listen in and join our conversation. Make sure to check out the webinar recording Digital Mastery & Joy ( https://info.dynatrace.com/apm_wc_panera_na_registration.html )

It's time for Pure Performance! Get your stopwatches ready, and exciting episode of Pure Performance. As always, my name is Brian Wilson that has not changed yet, and I think your name is still Andy Gravner, is that correct? Well, on my passport it says Andreas Gravner, just to be sure, but everybody please call me Andy. Can we get your passport number and driver's license number and some other information while you're giving out?

Yeah, well, you don't want to see my driver's license, especially not the Austrian one, because it's an embarrassment. First of all, the picture is old and doesn't look good. And second of all, in Austria, I still have the old pinkish

looking piece of paper that they call driver's license. And people typically laugh at it when they see it outside of Austria. I hope you have long hair in your old Austrian driver's license and people typically love edit when they see it outside of austria i hope you have long hair in your old austrian uh driver's license picture that'd be awesome a mullet i look like 16 and uh anyway so andy why why are we recording another we just so obviously these are aired weeks apart and everything and maybe even months depending on what we're doing but in reality we just finished wrapping up our recording with klaus about visually complete

and speed index so anybody who's listening like a long time ago is like wow this is all on the same day but we were so excited to do another one why are you so excited what are we gonna we're recording another one back to back why yeah so we still have klaus here with me why have him with me here in b. And the reason why we did another recording is Klaus earlier today did a webinar with Panera Bread and you did the intro section to that. And what I was so excited about was some of the messaging he had about, I mean, he talked about digital mastery. I will let him talk about what this means. But he brought up some topics that just so much reminded me of what

we have been talking to when we talk about DevOps and breaking down these silos. And he just has a different angle on the whole thing. And I was just fascinated. And actually, without further ado, Klaus, first of all, welcome back to the show. Hey, welcome. Wow. Love to be back. We won't tell everybody what happened in between these shows no no no um so klaus i was excited about the webinar you talked about digital mastery and the challenges and can you just fill us in a little bit on what you actually presented so i was talking about yeah digital transformation is a topic that everybody hears depending on which company you are in a little bit more a

little bit less but i spoke about not about the transformation itself i picked with panera a very good example they already went through the transformation. And I picked in the intro a couple of examples to just let people know that, yeah, digital transformation is something that happens. Other than the climate change, where people are still debating digital transformation, there is a common sense it is going on right away. And we've seen there so many changes, like the biggest sales event in the world the single day over in Asia on 11th of November is today the biggest e-commerce day. It's within just 24 hours, close to $18 billion of revenue. And 82% of this revenue is done via mobile devices.

So mobile is just a fact. And we have very, very good examples, like I'm traveling a lot like you, Andy, and the Marriott app is helping me a ton so it's really focusing with what they can on the frequent traveler to help him booking stuff getting

people to the hotel getting into the room even with keyless just hold your phone to the door it really helps a lot and now we keyless, just hold your phone to the door. It really helps a lot. And now we are actually expanding here massively.

Think of airline industry. You have kiosks on the airport. You have your mobile phone. You have maybe the app on the tablet. You have maybe your smartwatch with you. It helps you, again, while you're traveling, from the booking all the way till you're in the airport and up in the air and then back down. It adjusts what are you delivering to the consumer.

But there's so many different digital touchpoints. And this is now the thing, even getting worse. Digital touchpoints are talking to each other. Banking industry. If you go to an ATM, you don't insert your card anymore. You just hold your phone next to it because you already pre-arrival, you said, hey, dear ATM, I'm coming by. Give me $100. And it just gives you this money.

So you have digital touch points communicating with each other and if we look at that we have those I call them digital Raptors we have them already out there and they are just eating away the good simple nuggets from all those that are not willing to transform so So they are losing money already. And the thing is, if you look at digital transformation, a study that I was involved in, the result was the top four reasons why people cannot transform are data silos. Data silos meaning different databases that say you have data in Google Analytics and you have data in Dynatrace, you have data in your CRM,

you have data all over the place in multiple databases. And this is something that is actually hindering people to move forward with the digital transformation. The thing is, yeah, you will say like, man, the DevOps, didn't we just get rid of all the silos? Yeah, we got rid of the silos between the developers and the operators. But if you look at a company, there is so much more around. There are people like marketing division, salespeople, executives. Everybody wants to know something about these digital touch points. And since digital transformation is all about digital, you put all IT up front, really up front and on the spot. spot because now all of a sudden in order to get the answer hey is somebody let's say our virtual

my my example virtual consumer max is he actually transacting in a positive way with our company so that we are making money is he buying our services is he consuming our services? Is he buying our product? That's a key thing, but how can you find out? Max might be using at home his tablet on the way, the mobile phone. Yeah, then he's maybe in your shop, if you have a shop. He's in your shop and interacting there with your point of sale systems, a kiosk or something like that. How do we find out? Is he interacting with us in a positive way? What are the flows in between?

What is the user behavior behind that? And where we've been always really good is like DevOps takes care of, of, of each of these digital touch points you have a pipeline per digital touch point by the pipeline the mobile app developers want for the web site development another one for the kiosk app development and yeah whatever you have in house you basically always have these DevOps pipelines but this is basically a silo then. Because per pipeline, we probably get out the business results

because they are the most important thing that we have to look at anyway. The second thing is, is this digital touchpoint available to a consumer so he can interact with us? Is that whole thing going on without any failures? We have to look at the errors and customers, consumers today are expecting a certain performance and this is what they look at. They'll really like per silo everything's covered but now makes multi touch point how do you do that and this is where i think we with dynatrace did a great job and panera is this was was kind of an awesome example for that

we had we we can bring break down these silos help breaking down these silos, because we have this enormous amount of information. We can give you the few mechs used your digital touchpoints from the certain locations with whatever device, give you the deep technical insight. And also the business relevance, saying, hey hey how often did he interact with uh with us in a positive way for the company how often did he complete the transactions and on top the cherry on top of the cake is uh alerts for it used to be okay I go out there and fix it. But now within Dynatrace, we changed it. We changed this whole alert to also see what is the impact. So let's say I have there a box, a server somewhere in my backend sitting, and I would get an alert, disk is full.

Is this disk is full just because we are writing too many log messages and, yeah, who cares about the log files in this case? Or is disk is full means, okay, out there my users can no longer interact with my website because we cannot store a file or whatever this user might be wanting to upload. And this is something that I get out of the box. And then it makes it for operations people much more real. If I have all of a sudden an alert that tells me, hey, I have a thousand user interactions per minute being impacted by this particular problem or hey this is just a problem that I can take care of whenever I have

time because nobody there is no user input out there and this is a yeah but I think it's digital experience monitoring and it's at the edge, because having this central view, because it's not stopping here. In the webinar, I used the example with Alexa, where we have now user interfaces that are no longer depending on clicks, and still you want to know, did this user get a right answer from Alexa?

Was he, was this user able to buy something by Alexa or yeah, stuff like that. That's, that's what I'm passionate about. And I think I hit a good point. Yeah. So,

so what I, what I liked also so much from the guy, what was his name again? Uh, Prashant. Prashant from, from Panera, what he actually said, and you brought it, right?

You have the teams that are focusing on the kiosk. They're focusing on the mobile app and whatever else they have. And it's similar to what I see now in the microservice world, where we have microservice teams that are having their own pipelines, but they only focus on their microservice world where we have microservice teams that are having their own pipelines, but they only focus on their microservice and they make sure that they have all the monitoring correct, but they don't know how they potentially impact others. And so the same way is with here. So what he actually said, well, they figured out that in the morning people use a lot of

the mobile app to, but not to order order something online but just look at the zip code where is the next location so if I would be the person or the group that is responsible for the mobile app I would probably freak out because every day in the morning I see a lot of users but nobody of them converts and buys something whereas at noon they all use the mobile app

to actually put an order in, an online order and then pick it up at the store so conversion rates go up. But if I would not know, if I don't know that the people in the morning actually then go physically to the store and buy something through the kiosk or through the point of sales, if I don't know that, I may make the wrong decisions even. But if I have this consolidated view that I know, in your case, Max,

Max is using his phone first to get to the next digital touch point and then actually purchase with us. If we have this information, it's much better to actually understand customers moving through the system. Right, right. Absolutely. And then also, if you have the same data set and not need to compare log system one versus log system two versus CM system one, CM system two. I mean, we all know that there's the challenge, even we within Dynatrace, within our organization, we're using Salesforce and we're using this tool and this tool and sometimes it's a little hard. So that's why using actually Dynatrace on the whole thing and putting the user in the center and then knowing all the interactions, it just makes it so much much makes more sense and actually breaks down these silos absolutely absolutely and you know i can think of another example too um if you think about let's say any brick and mortar store who maybe has a website

and then they start doing a mobile app and maybe have a kiosk in the store doing the electronic point of sale systems in their store these These all might have been developed separately, right? And because of that separateness of development, they might each be even storing data differently into different databases or different file formats, or, you know, there might not be common data points through them. So that when, let's say, the mobile app gets developed, they have to pull data from one of the other existing resources, convert that data to their format and change it over. And instead of taking an approach of saying, all right, now we have all these different touch points.

Let's look at this, you know, not in terms of from a monolithic app point of view, but let's standardize across all of it. So that means we have to have our mobile app development team talking with our kiosk development team, talking with our point of sales development team, talking with our standard website development team so that we're all using same formats, data points,

all the same kind of formats for the services so that we don't have to take and do special conversions to make it work with our component. Bringing those all, you know, when you were talking about the idea of silos, Klaus, that's kind of where I was thinking more in the terms of, in traditional DevOps, of bringing that single line of team together, bringing the multiple lines of team together into one, and really just sharing all that data. Besides that awesome, that example, Andy, you brought up about the mobile development team in the morning, the location being used, but then nothing else until the afternoon.

And they don't know that those conversions are going on because they might not be, you know, there's so much here. And it kind of goes back to, I forget which episode and who he asked. I think it might have even been Gene Kim when we were asking, like, what's the next step in DevOps? What's the next big breakthrough? And I don't know if I would qualify this as quite a major breakthrough, but it's definitely a big leap in the idea of, you know, breaking down silos within the teams from dev to production, but also then within those separate tracks into bringing those together. I think this is just a fantastic idea here. Cool. Yeah, we have to do it because I just pulled out the study again,

and it's like the top four things were kind of interrelated. So the number one item was data silos department specific databases number two was poor data quality number three was missing coordinated processes between departments and number four was redundant data so if we take point number one the data silos and the missing coordination between the departments leading probably also to those data silos. It's no wonder that we have redundant data because different teams are having the same need for information. And redundant data leads them to poor data quality as well. So they are really strong interrelated.

And poor data quality leads to poor business decisions. Yeah, exactly. That's the worst case. You're absolutely right. Like, oh my God, our mobile app is not good, so let's rewrite it. No, it's actually good. People just shop somewhere else

or just shop through a different channel. And even going back to knowing how your code is used, you always bring that up, Andy, the idea of what's the adoption of code or functions that get put in. Who cares if the mobile app is mainly used for finding your store? That's part of the big picture of driving that revenue. So great, embrace it and don't really mess with it it but maybe just refocus on on what's how it's being used you know it's one step on the conversion funnel yeah that's what it is yeah you brought up something really important now revenue this is uh the whole monitoring starts with hey am i making money? Do I make revenue? Do people transact in a positive way?

This is key. We sometimes tend to start with performance first. I mean, we're in a performance company, but actually it is, are we making money first? And then it's about the availability of those touch points. And then it's about the errors. those touch points and then it's about the errors and i have to say performance is unfortunately just uh the fourth uh place in the traditional sense in the traditional sense yeah but performance impacts i was just gonna say but performance can very much have an impact on that revenue. And that's what we see over and over again.

And tying those performance metrics together with the business metrics, having the ability to see what the impact of a slowdown or some errors directly on either your bounces, conversions, or abandonments is quite crucial. Klaus, if people want to watch that webinar, what was the title again? There are two webinars. There's one, the Digital Customer, the Center of Your Universe, and the other one is Digital Mastery and Joy at Panera Bread. Cool.

So, folks, if you listen and if you want to check it out, I guess go to dynatrace.com or go to resources.dynatrace.com and then search for these webinars. Really worth watching. I mean, I didn't see the first one that he mentioned, but I think we may want to link those. Yeah, we'll link those as well. Yeah, that's good. Cool. Cool.

Well, look at this. Two episodes today. Yeah. Two of them in record time. Yes. There at this. Two episodes today. Yeah. Two of them in record time. Yes. There you go. There you go. So performance matters in this case.

Yeah. So I think every time we have a long show, we're going to have to have Andy, I mean, Klaus on for the next time to have a short show. To get the average now. Yeah. All right. Okay. Klaus, thanks again for being a guest

and welcome yes we welcome you back anytime so we will put up a link to your you don't have to try spelling out your well it's not even that hard I can do it right K-E-N-Z-E-N-H-O-F-E-R

right yeah that's your twitter handle and yeah follow Klaus he's always putting up a lot of great stuff and great ideas about P-N-H-O-F-E-R, right? Yeah. That's your Twitter handle. And yeah, follow Klaus. He's always putting up a lot of great stuff and great ideas about that digital experience and just wonderful, wonderful stuff. You're a great asset to our team, Klaus. Thank you.

You're welcome. And you can follow me at Emperor Wilson or Andy at GrabnerAndy. And you can also follow Pure Underscore DT. Send any questions, comments to Pure Underscore DT or pureperformance at Dynatrace.com. And so from me here in Denver, where we're going to have, it was 80 degrees yesterday. We're going to have snow on tomorrow. So, yay. I'll see you all soon uh i got nothing else so anybody else

no we're good greetings from boston where we hit 90 today so that's actually a nice change yeah and uh yeah thanks for listening thanks class thanks for having me. Talk to you soon. Bye. Bye.