Center Of Excellency For Sase And Ot And Io T Security at a outsourcing company with 51-200 employees
Real User
Top 20
Jul 15, 2026
Cloud Security Connector for Zscaler agents have some issues while applying the policies, and the main thing is when we try to apply policies over the ZCC agent, it takes a couple of minutes. I think that is a big gap while we are covering these kinds of metrics. When we are doing this thing, that should be proactive and can be applied as soon as possible. Cloud Security Connector for Zscaler does almost everything, but sometimes it can be a pain if we are using the ZCC agent because the logs that we have to get on the end users are really difficult to gather. Regarding Cloud Security Connector for Zscaler's accuracy and reliability of output, I would scale it on a scale of one to ten for accuracy. For accuracy, I would say it is around eight because it does the job for all the major AI platforms that are out there for browsers or maybe the applications, but there are a couple of applications that need to be added and a couple of more features which need to be added. For example, for ChatGPT itself, we can see the prompts what the users are prompting, but for Gemini, the same kind of tool, we do not have access to the prompts, so that can be improved. Obviously, if you do not have any other further tools for particularly AI security, then Cloud Security Connector for Zscaler is very reliable.
Cloud Security Connector for Zscaler can be improved by offering seamless OS upgrades while moving beyond reprovisioning, along with high-throughput burst performance and horizontal scaling. They can also improve deeper cloud-native infrastructure as code error handling. For Cloud Security Connector for Zscaler, improvements are needed for complex overlapping IP and non-HTTP protocol handling, as well as granular wildcard domain and bypass proxy behavior. The friction exists because admins need it. We have also noticed that when attempting to whitelist or bypass a root domain using the wildcard, the underlying proxy logic can occasionally act inconsistently with deep sub-URLs or complex, long-string hostnames, especially those exceeding 64 characters. If one sub-URL gets blocked, admins are forced to explicitly type out dozens of individual fully qualified domain names. The improvements they can make include a highly resilient, context-aware PAC, proxy auto-configuration engine built directly into the local appliance firmware that can cleanly evaluate complex regular expressions and long hostnames.
Senior Consultant at a outsourcing company with 1,001-5,000 employees
Real User
Top 5
May 15, 2026
One thing we can improve regarding Cloud Security Connector for Zscaler is that every time I log into my laptop, it asks me to sign in to Cloud Security Connector for Zscaler again and again. Perhaps we could change that login feature to reset once a month, but currently it resets every day. That is the one feature I feel can be improved; everything else is good. I do not need any other improvement currently; everything is good aside from the login issue. I am okay with other things.
A challenge faced during the initial implementation involved routing and application dependency. After routing cloud workload traffic through Cloud Security Connector for Zscaler, a few applications started failing because they depended on specific external services that were getting blocked due to strict policies or SSL inspection. To resolve this, traffic was analyzed using Zscaler logs, the exact domains and services being impacted were identified, and a controlled policy exception was created while maintaining overall security. This helped strike the right balance between security and application availability, which is very important in production environments. A key learning from this was that while Cloud Security Connector for Zscaler provides strong security control, proper policy tuning and understanding application behavior is critical for smooth deployment. Cloud Security Connector for Zscaler has had a very positive impact, especially in terms of security and operational efficiency. Before implementation, cloud workloads had direct internet access through the NAT gateway, which limited visibility and control over outbound traffic. After adopting Cloud Security Connector for Zscaler, all traffic was routed through Zscaler, giving centralized visibility and full policy enforcement. One improvement observed was a significant increase in threat detection. The ability to identify and block suspicious outbound connections that were previously not visible improved the overall security posture. From an operational perspective, troubleshooting became much faster. Instead of checking multiple cloud logs, Zscaler logs could be directly used to analyze traffic behavior, which reduced incident resolution time. The dependency on traditional firewall appliances in cloud environments was reduced, which simplified the architecture and lowered operational overhead. Additionally, for compliance-driven clients in banking, it helped meet audit requirements by providing detailed logs and consistent policy enforcement across both users and workloads. Measurable improvements were observed after implementing Cloud Security Connector for Zscaler. In terms of visibility and threat detection, a 25 to 35 percent increase in identifying suspicious outbound connections was seen that were previously not visible when traffic was going through NAT. From an operational standpoint, troubleshooting time reduced by 45 to 55 percent because Zscaler logs could be directly analyzed instead of checking multiple cloud logs. Dependence on additional security appliances in the cloud was also reduced, which helped lower operational overhead and simplified the architecture. In terms of compliance, better audit readiness was achieved with centralized logging and consistent policy enforcement across workloads and users. While Cloud Security Connector for Zscaler is a strong solution, one area that could be improved, especially for teams that are new, is configuring route tables, ensuring proper traffic flow, and avoiding asymmetric routing, which can be challenging, particularly in large or multi-VPC environments. More automated deployment options or guided configurations, especially for AWS or Azure, would simplify the onboarding process. Another area for improvement is better visibility at the cloud-native level, such as tighter integration with cloud logs or more context-aware insight for workload behavior, which would make troubleshooting even faster. Once properly implemented, it works efficiently and provides strong security and visibility. In addition to deployment simplicity, improvements around policy tuning and documentation would add significant value. From a policy perspective, when Zscaler policy is extended to cloud workloads, it sometimes requires careful fine-tuning to avoid impacting application dependency. More predefined templates or workload-specific policy recommendations would help teams implement it faster with fewer disruptions. In terms of documentation, while the existing guides are helpful, more step-by-step real-world deployment examples, especially for multi-VPC or hybrid environments, would make onboarding smoother for teams. Tighter integration guidance with cloud-native tools such as AWS logging or monitoring services would further improve troubleshooting and visibility. From a support standpoint, faster access to best practice recommendations or reference architecture would help teams avoid common misconfigurations during the initial setup.
I feel that they are doing great with Cloud Security Connector for Zscaler. If I need to suggest an improvement, it would be to simplify the steep learning curve, as it can be complex for newcomers without prior experience. Apart from that, I did not face any challenges with them in these two and a half years.
If there are any issues at the network level, such as users working from home, then it is very hard to catch up with them and resolve issues on the end-user machine. If the user doesn't have internet access because of Zscaler... Sometimes, it might be a rare case, but it is very challenging for us to take action because they don't have internet. It can be difficult for us to troubleshoot. The user accessibility; it is better to improve that for control accessing and configuring. It's very complicated to do the configuration. The IT job is a little bit complicated. Those who have the experience can do that. It's not easy or handy to use this tool.
Learn what your peers think about Cloud Security Connector for Zscaler. Get advice and tips from experienced pros sharing their opinions. Updated: August 2026.
Cloud Security Connector for Zscaler is designed to enhance security by blocking malicious content and providing agentless integration with cloud workloads. It simplifies policy management and improves scalability, making it a critical tool for securing cloud environments. Cloud Security Connector for Zscaler automates node detection, route selection, and offers centralized visibility, allowing businesses to focus on scaling without traditional firewall constraints. Its high availability...
Cloud Security Connector for Zscaler agents have some issues while applying the policies, and the main thing is when we try to apply policies over the ZCC agent, it takes a couple of minutes. I think that is a big gap while we are covering these kinds of metrics. When we are doing this thing, that should be proactive and can be applied as soon as possible. Cloud Security Connector for Zscaler does almost everything, but sometimes it can be a pain if we are using the ZCC agent because the logs that we have to get on the end users are really difficult to gather. Regarding Cloud Security Connector for Zscaler's accuracy and reliability of output, I would scale it on a scale of one to ten for accuracy. For accuracy, I would say it is around eight because it does the job for all the major AI platforms that are out there for browsers or maybe the applications, but there are a couple of applications that need to be added and a couple of more features which need to be added. For example, for ChatGPT itself, we can see the prompts what the users are prompting, but for Gemini, the same kind of tool, we do not have access to the prompts, so that can be improved. Obviously, if you do not have any other further tools for particularly AI security, then Cloud Security Connector for Zscaler is very reliable.
Cloud Security Connector for Zscaler can be improved by offering seamless OS upgrades while moving beyond reprovisioning, along with high-throughput burst performance and horizontal scaling. They can also improve deeper cloud-native infrastructure as code error handling. For Cloud Security Connector for Zscaler, improvements are needed for complex overlapping IP and non-HTTP protocol handling, as well as granular wildcard domain and bypass proxy behavior. The friction exists because admins need it. We have also noticed that when attempting to whitelist or bypass a root domain using the wildcard, the underlying proxy logic can occasionally act inconsistently with deep sub-URLs or complex, long-string hostnames, especially those exceeding 64 characters. If one sub-URL gets blocked, admins are forced to explicitly type out dozens of individual fully qualified domain names. The improvements they can make include a highly resilient, context-aware PAC, proxy auto-configuration engine built directly into the local appliance firmware that can cleanly evaluate complex regular expressions and long hostnames.
One thing we can improve regarding Cloud Security Connector for Zscaler is that every time I log into my laptop, it asks me to sign in to Cloud Security Connector for Zscaler again and again. Perhaps we could change that login feature to reset once a month, but currently it resets every day. That is the one feature I feel can be improved; everything else is good. I do not need any other improvement currently; everything is good aside from the login issue. I am okay with other things.
A challenge faced during the initial implementation involved routing and application dependency. After routing cloud workload traffic through Cloud Security Connector for Zscaler, a few applications started failing because they depended on specific external services that were getting blocked due to strict policies or SSL inspection. To resolve this, traffic was analyzed using Zscaler logs, the exact domains and services being impacted were identified, and a controlled policy exception was created while maintaining overall security. This helped strike the right balance between security and application availability, which is very important in production environments. A key learning from this was that while Cloud Security Connector for Zscaler provides strong security control, proper policy tuning and understanding application behavior is critical for smooth deployment. Cloud Security Connector for Zscaler has had a very positive impact, especially in terms of security and operational efficiency. Before implementation, cloud workloads had direct internet access through the NAT gateway, which limited visibility and control over outbound traffic. After adopting Cloud Security Connector for Zscaler, all traffic was routed through Zscaler, giving centralized visibility and full policy enforcement. One improvement observed was a significant increase in threat detection. The ability to identify and block suspicious outbound connections that were previously not visible improved the overall security posture. From an operational perspective, troubleshooting became much faster. Instead of checking multiple cloud logs, Zscaler logs could be directly used to analyze traffic behavior, which reduced incident resolution time. The dependency on traditional firewall appliances in cloud environments was reduced, which simplified the architecture and lowered operational overhead. Additionally, for compliance-driven clients in banking, it helped meet audit requirements by providing detailed logs and consistent policy enforcement across both users and workloads. Measurable improvements were observed after implementing Cloud Security Connector for Zscaler. In terms of visibility and threat detection, a 25 to 35 percent increase in identifying suspicious outbound connections was seen that were previously not visible when traffic was going through NAT. From an operational standpoint, troubleshooting time reduced by 45 to 55 percent because Zscaler logs could be directly analyzed instead of checking multiple cloud logs. Dependence on additional security appliances in the cloud was also reduced, which helped lower operational overhead and simplified the architecture. In terms of compliance, better audit readiness was achieved with centralized logging and consistent policy enforcement across workloads and users. While Cloud Security Connector for Zscaler is a strong solution, one area that could be improved, especially for teams that are new, is configuring route tables, ensuring proper traffic flow, and avoiding asymmetric routing, which can be challenging, particularly in large or multi-VPC environments. More automated deployment options or guided configurations, especially for AWS or Azure, would simplify the onboarding process. Another area for improvement is better visibility at the cloud-native level, such as tighter integration with cloud logs or more context-aware insight for workload behavior, which would make troubleshooting even faster. Once properly implemented, it works efficiently and provides strong security and visibility. In addition to deployment simplicity, improvements around policy tuning and documentation would add significant value. From a policy perspective, when Zscaler policy is extended to cloud workloads, it sometimes requires careful fine-tuning to avoid impacting application dependency. More predefined templates or workload-specific policy recommendations would help teams implement it faster with fewer disruptions. In terms of documentation, while the existing guides are helpful, more step-by-step real-world deployment examples, especially for multi-VPC or hybrid environments, would make onboarding smoother for teams. Tighter integration guidance with cloud-native tools such as AWS logging or monitoring services would further improve troubleshooting and visibility. From a support standpoint, faster access to best practice recommendations or reference architecture would help teams avoid common misconfigurations during the initial setup.
I feel that they are doing great with Cloud Security Connector for Zscaler. If I need to suggest an improvement, it would be to simplify the steep learning curve, as it can be complex for newcomers without prior experience. Apart from that, I did not face any challenges with them in these two and a half years.
If there are any issues at the network level, such as users working from home, then it is very hard to catch up with them and resolve issues on the end-user machine. If the user doesn't have internet access because of Zscaler... Sometimes, it might be a rare case, but it is very challenging for us to take action because they don't have internet. It can be difficult for us to troubleshoot. The user accessibility; it is better to improve that for control accessing and configuring. It's very complicated to do the configuration. The IT job is a little bit complicated. Those who have the experience can do that. It's not easy or handy to use this tool.
The connectivity could be faster. Sometimes, the issue is internal.