What is our primary use case?
Before we had the Radware services, we used another tool. It was not very good at understanding the bot traffic, and it couldn't really stop it. So, we had to utilize Radware for that.
How has it helped my organization?
It can categorize different bots, which is helpful because, initially, we were seeing large spikes in connection attempts. They were bots. Our assumption was they were all bad bots, but that's not necessarily true. There are many different services out there that are like aggregators collecting information on the internet. Understanding that there are good bots versus bad bots out there helps you to make different decisions. You definitely want good bots to come in, but you want them to be able to use only a certain threshold of bandwidth. That's a metric that we can set. So, it has definitely helped us to see what they are and how to limit their utilization against our site and then cut out the bad ones.
Bot Manager has helped reduce the number of false positives that our organization receives in response to attacks. As a part of the project, we did a lot of reporting to show the impact when we turn the services on. It was pretty staggering. There was a pretty big drop in terms of the bandwidth used and attacks on the site. The impact was huge. From about half a million connections a day, we brought them under a hundred thousand at that point. About 80% or higher of that was noise, where bots were just scanning our website and looking for flaws.
Bot Manager didn't reduce downtime associated with attacks, but there was a reduction in CPU, memory, and disk usage. We reduced the CPU, memory, and disk usage for those services because they weren't getting nearly as many connections. So, there are definitely cost savings.
There have definitely been time savings because I now just get alerts instead of having to go in and take action to see what was occurring. I get a few hours back a week instead of actively going out there looking for issues and taking care of them. I don't have to interact with them.
Previously, I was probably spending six to eight hours a week looking at the site trying to determine utilization. I don't have to do that anymore. I'm getting back six to eight hours a week, which comes out to be more than 800 hours for the past two years.
The Bot Manager crypto mitigation algorithm is a capability in there, and we do have it turned on. Essentially, what it does is that if anyone is trying to use bots in a malicious way to attack your service, it will essentially cause them to consume more resources in order to try to attack or do something against your site. It's an interesting concept. It has helped to reduce the number of bot attacks, but I don't have the metrics.
What is most valuable?
It's very good at categorizing the different types of bots, whether they're malicious or good. Bot is a very generic term. It could be good, it could be bad. Quite a lot of legitimate businesses are using bot-type services to just scrape the internet for information.
The Bot Manager portal is very easy to use now. When we first started, it was a separate portal. It was different. It's all wrapped into one now, so it's easy to use.
Bot Manager’s ability to detect and mitigate bots in real-time is very good. It has been very effective. We definitely see different bot types. It has done a good job of stopping them. We do get alerts that something has occurred, so it's an effective service.
What needs improvement?
It would be good to have more integrations. It's very hard to get data in and out of their portal. It doesn't have any integrations with any of our tools, such as our SIEM tool. It only depends on emails. Having that tied into the warehouse, SIEM, and maybe our on-call tools would be very helpful because it would just give us a holistic picture of everything.
The Bot Manager portal sometimes doesn't refresh correctly. There are cosmetic issues that can be improved on.
For how long have I used the solution?
I have been using this solution for two years.
What do I think about the stability of the solution?
It has been great. It has been a very good service. Never had a problem with it. Never had an outage. I'd give it a pretty good rating.
What do I think about the scalability of the solution?
We haven't had any performance issues due to scalability. It has done a good job.
How are customer service and support?
We have contacted them a lot. When tools are new to us, there are a lot of questions. While we were doing the implementation, the bot service used to be a separate portal, and then they rolled it into the Cloud WAF portal. It's a unified portal. During that time frame, there were a lot of issues where something was no longer available in the portal and moved somewhere else. We had questions about their migration to the new portal, and then we just had questions about any issues that we had with the service or the way they implemented the application, such as what was your intent or what am I supposed to do in the situation. They were good. I'd rate them an 8 out of 10.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
Azure has some built-in native capabilities, but they're not great. It's called Azure Application Gateway, but it's not nearly as good as what Radware offers.
The protection capabilities against attacks are very good inside of Radware. It can definitely categorize an attack and determine the attack type. I can easily see what the attack is. It can tell me a little bit about the client that's trying to connect, whereas none of that was available in Azure Application Gateway. It was even hard to get a user agent type. That's a basic capability.
How was the initial setup?
It's straightforward. You have to work with the engineer you are assigned by Radware. It works great when you have someone like a project manager and someone from the technical side to do the implementation.
Bot Manager is integrated with our WAF. They weren't integrated natively. We had to integrate them. When we first initially launched, we turned on the Cloud WAF, which gave us some visibility, and then the next step was to turn on bot protection. We tried to turn them on piecemeal so that we could understand the impact it caused on the application and what kind of traffic it was modifying.
What about the implementation team?
We had two people involved in its implementation.
What was our ROI?
Its time-to-value was immediate. Once we enabled those services, it definitely cleared out all of the attacks where the traffic was not customer based or that didn't need to be there and served us no value.
Which other solutions did I evaluate?
We did demos of the Citrix, Imperva, and Azure products.
What other advice do I have?
To someone who wants an anti-fraud solution but isn’t considering Bot Manager, I'd say that it really comes out of visibility in the tool. In our case, we can show the metrics of our typical website application usage and then the significant decrease in traffic after the Bot Manager service. There was a significant decrease in traffic that was not just bots. That was not related to any type of customer. We filtered it down to just what our core customer usage is, and having that data helps a lot because otherwise, you just make assumptions that all those connections 24/7 are legitimate usage patterns, but they're not. So, it has a lot of value.
I'd definitely recommend utilizing such a service, whether it's Radware or anyone else. It definitely provides a lot of business value. You cut down on all of that noise and traffic and protect your resources from malicious traffic. It's definitely worth the time and effort.
Our experience with Bot Manager has been very good. It served the reason we bought it, and it's doing its job from a scalability, performance, and reliability perspective.
Overall, I'd rate it a 10 out of 10.
*Disclosure: PeerSpot 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.