What is our primary use case?
We use it for our development teams to make sure we have development feedback loops. We run about 50 different development teams and development streams. It is very important for us to keep on top of our deployments. We ran about 950 deployments and the tool gives us the flexibility to ensure that we are staying as close to those deployments moving forward.
How has it helped my organization?
Before we had the tool we had no visibility into the user experience and capturing what was going on inside the browser. We utilized tags so we knew how many times people were doing certain things, but we did not know how the performance was, if users were satisfied with what they were doing, and if we were serving up errors. We had no ability to correlate anything that was going on in our back-end systems and what users were doing. Being able to have that viability and that insight into what is going on for our customers was immensely valuable across the board from the development perspective all the way through to higher level business people.
One of the reasons that we are going into the new Dynatrace platform. We have a lot of data. With that amount of data and my team being very small, we are not specifically developers, we do incident management and problem management. The administer of the Dynatrace tool makes sure the monitoring is out and available for everybody where it needs to be. We do not have time to look at all the data, and with all the AI and automatic stuff being able to do management zones when that coming to us soon, a lot of the feature sets which are moving forward will make my job so much easier. I will not have to work 60 to 65 hours a week to ensure I am getting stuff to the developers so they can do what they need to do.
Dynatrace is staying up with IoT and a lot of the cloud solutions, which is really going to be helpful for us in the future.
What is most valuable?
The most valuable things that we have seen are the user experience and capturing what the users are doing inside the browser, and being able to equate that back to the business and telling them how much of their company is doing what. Also, what the performance time is, so we can give that back to our developers and make sure that the developers are spending time on what they need to spend time on to make sure that they are noting performance of the website.
What needs improvement?
Stability and scalability have been issues right now. My understanding going forward, and I am cautiously optimistic, is that we will not have these problems anymore. I would really like for that to be the case. We are a large company. We do a lot of microservices. We are going into the cloud. We are doing a lot of different things. We use PCF and Docker. We do a lot of the different technologies. The ability for us to scale the solution is going to be very important, especially going forward, because we are exploding in size. We are supposed to grow at least two times in the next year.
Session replay availability is going to be the most amazing game changer for our company. We are very heavy into user analytics. There is a completely separate segment inside of our company that looks into things like user tagging and making sure that we are gathering who is doing what inside the site. The session replay ability and the ability to send that over to the call center to say, "Hey, we know, say this," or an automated response to our users to say, "Hey, we have a problem on our website clipping coupons", or pulling in some kind of eCommerce would be absolutely pivotal. It would absolutely change the game inside the company.
What do I think about the scalability of the solution?
We have had some struggles with scaling. We are on our way to the new platform, the new Dynatrace platform, which will alleviate some of these pains. We were not expecting the level of adoption that we got with the product. We brought it in thinking a few of the teams in a segment of our company would want to use it. Everybody jumped in on it and jumped in on it really quickly. Therefore, we quickly ran into scalability issues, but we are working on alleviating that going forward.
We were on AppMon, and AppMon has its own monolithic drawbacks. On the new platform, we will not have any these problems. We can scale in the cluster horizontally.
The role of AI when it comes to IT's ability to scale into the Cloud and manage performance problems is very important. We were hitting a problem with our scalability issues but now we are going larger. Part of the problem is we have so much data and we have no idea how to use it all. We would find blind spots and be able to help and do what we can when the issues came to us. If we had the issues telling us when there were problems before development and before call centers got the problems, we could retroactively go out and get the problems before customer call centers had problems. We have a problem inside the company that we have so much data and we have no idea what to do with it all. AI will help solve that issue and help move us forward. It could pull the stuff that is problematic, the most performing or non-performing issues, in areas where we want to see certain things.
How is customer service and technical support?
We have used support quite extensively. We have had many very imperative tickets with them, and they are very supportive. They are very good with communication. They are very transparent in issues, what they need to do, and how they need to fix certain issues and problems. We have worked very closely with a lot of the support. We get a lot of offshore support from the Austrian development teams and a lot from the Polish development teams. Anybody they could pull in to make things happen and to make the pain go away for us, they do it, and they do it very well.
Which solutions did we use previously?
I have used Wiley, though not at this company.
We used to use AppDynamics. We did a PoC for CA Wiley. Before we brought in Dynatrace, we did a PoC with all of them. We just got rid of AppDynamics (out of our environment). They did not allow for the deep dive visibility.
A lot of the problems that we had with products like AppDynamics was it got us to a certain point, then we were not able to see any deeper. It would dump us in something they called a metric browser, then we just got metrics, but we did not see what was going on in the underlying code.
In Dynatrace, we could decompile the source code. We could see the things on the fly. We could see what is actually going on inside the tool. It has been very helpful. We did this thing with the Wiley tool, but it was just way too immature and they were not even close to even having any of the conversations that we wanted to have.
How was the initial setup?
I was involved in the initial setup, but I have done it before. I have done it on multiple different sites so probably done three stand ups so far: two at the prior company and one at this one. It is not bad. It was easiest with this company, because this company had a level of technical maturity that was not available in the other ones. If you have companies that are doing things like continuous delivery, having built pipelines and having the ability to do these things, it is a little bit easier. I have a feeling that the companies that are more technically challenged, the initial setup is going to be a little bit harder.
Our main problems were around security and network, but those are hurdles that almost everybody has got to get over and build. It was not bad.
What about the implementation team?
The vendor team tried to help during implementation. A lot of the struggles were not with the product. It was more with the way that we do things inside the company, so it was internal struggles from the company side. The product was always there and I could always reach out for support, if I needed additional help.
What other advice do I have?
Look at your audience. Who is the audience that is going to be consuming your data? If it is going to be primarily developers then you want to be able to push this out through the business, there is no other solution. If you want your developers and people to actually see what is going on inside the code and be able to fix and proactively fix stuff before it happens, this is the solution that you want.
Most important criteria when selecting a vendor: The vendor needs to be very future facing. We jump on technologies. We are a huge company with over 400,000 employees. We jump on new technologies within days or weeks of it coming out. This is not the best strategy in most cases and most larger companies tend to stay away from change. I have been in production environments where we have upgraded and changed the version of one of our most pivotal production servers within two days of it being released from the company. This is not usually the best thing.
Dynatrace does a good job to make sure they stay out in front of the new technologies. We have had to ratchet back our development teams and tell them, "You need to wait at least a two or three weeks before you jump on the newest version, especially going into production." We know that most technologies inside of Dynatrace that they will move with us to make sure that we are keeping up with our development teams.
Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.
Feb 20 2018