Transcript

Whitney:

Hello everyone. Let's do a quick sound check before we begin. If you can hear my voice, can you type 'yes' into the questions of the chat box? Awesome. There's a bunch of yeses.

 

Okay. Good morning, good afternoon, or good evening depending on where you're joining us from today. It's great to see so many faces in this webinar today. I know the Thanksgiving holidays start tomorrow. Thank you for taking some time out of your day to join us. We've got a pretty exciting webinar, Synthetic Monitoring Best Practices, 99% uptime 24/7. With Zishan Malik, our technical account manager.

 

He's probably one of, if not, the most knowledgeable guy when it comes to synthetic monitoring I hear. He's been with the company the longest. You guys are all in for a treat. Feel free to ask your questions throughout the presentation. Once he's just finished with his presentation we'll make sure we get through that.

 

For those of you who are new today feel free to follow us on social. We've got a Facebook account. You can find us @ApicaSystems as well as Twitter @ApicaSystems. If you like this webinar feel free to find more webinars as well as other educational resources on our website under the resources tab. With that being said, let's get right to it and welcome Zishan. 

Zishan Malik:

How's everybody doing today? Hope everybody's all ready and excited for Thanksgiving. Today we're going to cover a few topics. Environmental testing and monitoring, top-down approach to monitoring, mobile synthetic monitoring, monitoring success story, and remedies and recipes that we've applied throughout the years. At the end of it all, I'll answer all of the questions that come through to Whitney as we go along.

 

Now I've been doing this for about three and a half years to four years. Doing synthetic monitoring and before that I was doing operational monitoring, infrastructure monitoring. What I have realized over my time is that monitoring is an after thought. Monitoring is something that we discuss or talk about once everything has gone into production. Now we are running like headless chickens or turkeys, in this case, trying to look for ways to monitor the application that we've just launched. We want to make sure that it's up and running and that we haven't caused any havoc on holiday season.

 

What most people end up doing is as they're implementing it they implement monitoring along with it. In my experience, and as I've talked to more and more Fortune 500 companies out there, I've made the suggestion to implement monitoring in the design phase. As you're developing your application or as you're planning and analyzing your application you should also think about how you're going to monitor that application. What points are very critical to monitor.

 

How many of you have actually experienced that in your day-to-day environment where you've actually had a monitoring application, or an application go out into production and then decided to monitor it? Instead of actually monitoring it from near your deadline, in your staging environment, in your UAT environment. That's actually the case for everybody. Sometimes it has to do with resources. It has to do with finances. I completely get all of that. It's so much easier when you just utilize monitoring and make it a conversation from the get go in the design phase, rather than trying to look for ways to monitor your own application after the fact.

 

Apica, that's the tool that we utilize a lot. We've taken scenarios and user journeys that we thought were going to be most relevant. At first it's hypothesis and as you collect more and more data you make it more of a concrete decision on what to monitor and adjust your user journeys based on that monitoring. Your hypothesis as a user journey, I log in. I go to a website, a homepage. I log in. I search for something. I add it to the cart. Then i go ahead and checkout and you go to a payment gateway. From there on you make your payments and the transaction. That's a whole full user journey.

 

You can see that on your picture to the right. You got amazon.com and on average that whole user journey takes about eight seconds. That's your response time. Out of that just the homepage only takes about two to three seconds. That's what we should be thriving towards. Everybody should be thriving towards that. We need that faster response because nowadays retention and attention have the same span, life span. It's very minimal. It's actually one second less than that of a goldfish. Just imagine that and take that into consideration when you're actually going for developing your application.

 

This just isn't happening in applications. Even your current applications, how do you monitor that? Obviously, every application will have a chance for a systematic approach to upgrading that application or releasing patches or releasing bug fixes. You want to make sure that you monitor that in the dead environment first then the staging, the UAT environment, alpha beta. How many ever environments you have. Just difference in them is the same thing. How many ever environments you have in your company policies that's what you want to be monitoring. Every single one of those environments before it hits the production.

 

That takes us to our next point, which is what kind of approaches are you using to monitor this? Are you using the bottom-up approach or the top-down approach? The difference is top-down approach is more of a user journey. It's not just your infrastructure monitoring or it's your web server doing what it's supposed to be doing. Is your database making all those queries and all those calls and what the rendition time for those calls are. That's the bottom-up approach.

 

It's great to have the bottom-up approach. What we suggest is taking that bottom-up approach and mixing it with your top-down approach. Making sure that you have your user journey in place. That you can see what your users are experiencing. You're able to catch those third-party URLs at that break. How many of you are actually using that right now in your approach? Are you guys using a more bottom-up approach right now? Are you guys using a more top-down approach? Why is it that you guys are not using both of them? You should be using both of them.

 

Apica can integrate with a lot of APM tools out there. We have a lot of integration possibilities with a lot of vendors out there as well. Taking both of those and putting it together you get the best of both worlds. You get the user experience and you get your infrastructure monitoring in the back-end as well, all your application performance monitoring in the back-end.

 

Your QA team is now happy. Your marketing team is now happy because they're not having to put out fires at the end of the day. Your operational deadlocks team is happy because they can see what the performance on the application there is, right? On the infrastructure there as well. 

 

What I have on the right hand side is the 10 slowest URLs from the Amazon site that I showed you earlier. What you can see is that Facebook is among one of those. There's a few third-party URLs. That is actually [inaudible 00:08:12] their site down. It is very interesting to sometimes see that your site, your applications working completely fine, but you're now starting to get errors because a third-party vendor is having issues on their end. To catch that in the death side or in the UAT side, it's much easier to resolve and bring it up with your third-party vendor than if it's happening in your production site. That could cause a bad experience for your users. There could be some DOM completes that never do get completed or your DOMs never get loaded, so they never get complete. That could cause some values to be misplaced. I've seen that happen actually many times in my life. It has happened more and more with mobile applications rather than your web applications, but it has happened with web applications as well.

 

That brings us to our next point, mobile-tablet synthetic monitoring. Right now, on average, there's about three hours that we in the US spend on mobile devices or tablet devices. The conversion rate's actually gone up than that of traditional. There's about 5% conversion rate on mobile and tablet than there is that of traditional. This is for eCommerce geeks out there. If you're interested in that. You should be because it's all about getting more money for the company. That's what we're all here for, right?

 

Mobile synthetic and to have a synthetic monitoring is a new concept that's coming out. More and more we see that the trend is that more people are using applications. More people are using applications on their tablets, on their phones. They're moving more and more away from the traditional laptop and computer. Obviously, there's never going to be a point where we're going to get rid of completely that, but that is happening gradually. We need a way to monitor that. How do you monitor that?

 

We here at Apica use something called Zebratester which is a proprietary tool to Apica. What it does is it sits on that mobile device and it basically sniffs all the traffic that comes in. That's how we're able to record the whole user journey for you on the application level whether it's mobile or tablet. Anything that is over HTTP or HTTPS or any web traffic we can actually capture that through using Zebratester. It gives you that certainty. You just make sure that your functionalities are working. More than that you actually start getting response times from that. You see how responsive your applications are. If your application is not responsive enough then there are multiple alternatives out there that people will just abandon your application.

 

Again, going back to the retention time that I talked about earlier and the attention time, people don't have that kind of patience anymore. If you don't provide the utmost service like the best service there is, excellent service, and on top of that it has to be in a timely and fast manner. Then that's the way to retain a customer. That's the way to grow your business. We help you do that. We help give you the numbers, the knowledge, the ability to monitor your applications, so you can improve on that. You can also use Zebratester to do mini-load tests to see how much of an impact if you have simultaneous users that come in, how much of an impact your application will have. How slow will it perform? How fast will it perform? This is also true for recreational as well traditional web applications as well.

 

All right. I want to talk about one monitoring success story that actually has relations to everything I just talked about. I want to kind of sum up everything together.

 

A recent customer that actually released an application into the iOS market and the Android market. The application, great application by the way, came into the market. People downloaded it and started using it. It was a subscription based application, so when they logged in and they had already purchased the subscription they would be redirected to repurchase a subscription even though they had one existing

 

For any new customers once they purchased the subscription it was going back in a loop and they were again redirected to purchasing a new subscription. That caused a lot of bad reviews in the iOS market and the Android market. They got a really bad rating, so they had to actually abandon the whole project and build a new application. Bad ratings, nobody else is going to be starting to use the application itself. When they did that they actually contacted us, me personally. We sat down with them to understand what their whole infrastructure or whole application stack looks like, all the stacks. We analyzed everything.

 

In the meantime, the short-term solution, what we did was implement URL version two checks just to make sure that all their links were up and all their sites were up. In the longterm we started building up scenarios to make sure that every component of that application was being monitored. We did this not just for the mobile application, but also for their web application as well. The traditional user laptop, the desktop, and all that. We made sure that every signup, all the utilities, and the services that were behind the signup page were being monitored such as Mule Services, load balancers.

 

Also, if there was any firewalls that were there, we made sure to monitor all that and get you back a response time and functionality. Make sure to get all that information back to see exactly how that application was doing. Overall, that remedy basically, understanding the stack, spoke to the developers to understand the main components of the application stack. We did an assessment, implemented uptime checks on the servers, built specific user journeys.

 

All that encompassed together gave them what they were able to comfortably re-release the application into the market. Right now they're doing about, I believe, four and a half stars which is a great rating. Which is absolutely a great rating for an application. I just kind of wanted to cover that with everybody. Get that all to sum up everything I was talking about until now. They were able to retain higher, I believe, they even did about 47% business by doing that.

 

It kind of really gives you an idea of taking all those concepts and putting it together how they were able to turn something that was really bad and not working for them into something successful and increase their sales by doing so. That basically sums up my webinar for today. If anybody has any questions please at this time go ahead and let Whitney know and she'll ask them out loud to me. 

Whitney:

Awesome. Thanks Zishan. I've got a couple questions that have come in here. Right before I get them I just want everyone to know that we are recording this webinar. It will be available on the Apica resources page within the next week. You guys will be sent the recording as well.

 

The first question here is from John. John asks, "That theory that you mentioned in this presentation, is it applicable to all industries or just a specific industry?"

Zishan Malik:

It's actually applicable to any industry. We've done this with a finance industry. We've done this with an educational industry, eCommerce. We've also done this with relation industry as well, believe it or not. It's actually applicable to all industries. We've implemented this in various different industries out there, but most heavily in eCommerce and financial.

Whitney:

Then I got David who wants to know how much Zebratester costs.

Zishan Malik:

If you guys want to know about the costs, please do reach out to our sales at Apica. They will be more than happy to negotiate the price with you guys. I don't know exactly how much it is. I'm more a technical resource rather than a sales resource.

Whitney:

Zebratester's actually a free tool that you guys can download. Please check out our Zebratester community. I can type out the URL to you guys as well after this along with the recording. Then I think, sorry, a couple of the other questions are actually very situational questions, so I think we're going to hold off on the questions that are coming out. I'm actually going to email you guys and have a couple of our technical guys reach out to the rest of the questions. Other than that Zishan thanks so much for your time today. 

Zishan Malik:

Welcome, anytime. It's always great doing this. 

Whitney:

Yeah. Thank you everyone for taking time out of your day. We will see you guys next time.