Does Zscaler Spy on Employees? What Your Employer Sees, What Zscaler Keeps, and Where Decryption Happens
.jpeg)
Short answer: No, Zscaler doesn't spy on employees for its own purposes. It's a data processor working for your employer, and its own privacy FAQ says transaction content is inspected in memory and not stored for most services. Our take: the real privacy question isn't Zscaler's intent, it's architecture. Zscaler Internet Access decrypts your traffic inside Zscaler's data centers and logs every transaction with your identity, while an on-device secure web gateway like dope.security does the same inspection without the decrypted content ever leaving the laptop.
If you've typed this question into a search bar from a work laptop, you're not paranoid. You noticed a Zscaler icon in your menu bar, a certificate you didn't install, or a block page with your company's logo on it, and you want to know who can see what. That's a fair question, and the honest answer has three parts: what Zscaler the company processes, what your employer can see in the console, and where the decryption physically happens. If you need the basics first, our explainer on what Zscaler is and how its cloud proxy works covers the product itself.
This guide is written for both sides of the desk: the employee who wants a straight answer, and the IT or security leader who has to explain the deployment to a works council, a privacy team or a skeptical engineering org. We'll lead with Zscaler's own documentation, then show how the picture changes when inspection runs on the device. For the deeper technical mechanics, see our breakdown of how Zscaler ZIA SSL inspection works and what it breaks.
Does Zscaler spy on employees?
Not in the sense most people mean. Zscaler says it is a data processor that handles personal data on behalf of its customers under a Data Processing Agreement. Your employer is the controller. Zscaler doesn't decide what gets logged, who reviews it, or why. Your company does.
What Zscaler does is sit in the path of your web traffic. With Zscaler Internet Access (ZIA), the Client Connector on your laptop forwards your browsing to the nearest Zscaler data center, where the request is inspected against your company's policy and then sent on to the destination. Zscaler's privacy FAQ lists the personal data it processes as limited: IP addresses, URLs, user IDs, and user groups and departments pulled from the corporate directory.
So the accurate framing is this: Zscaler isn't watching you for itself, but it is the machine your employer uses to see your web activity, and that machine lives in Zscaler's cloud, not on your desk.
What your employer can actually see in Zscaler
This is the part that matters day to day. A Zscaler admin with reporting access can typically see a transaction-level log of web activity tied to a user identity. Based on the data types Zscaler lists, that means who (user ID and department), from where (IP address), and what (the URL requested), plus the policy verdict and category.
Because ZIA performs SSL inspection, it can see more than the domain. Full URL paths are visible, which is how a policy can tell the difference between browsing a site and uploading a file to it. If your company has licensed Zscaler's data protection features, file uploads and certain content can be evaluated against DLP rules, and violations show up in the console.
What employers generally do not get from a secure web gateway is a screen recording or a keystroke log. A SWG sees web transactions. It doesn't see your local documents, your desktop apps that don't touch the web, or your personal phone on cellular data. If you're worried about broader monitoring, that's a question about other tools, and our guide to employee internet monitoring in 2026 walks through the categories.
Zscaler can hide your name from admins
One detail employees rarely hear about: Zscaler supports obfuscating user names for admins. According to Zscaler's help documentation, obfuscation is set on the admin role, so every admin with that role sees pseudonymized user names in dashboards, reports and insights, and it can extend to log streaming feeds. A parallel setting covers device information. Zscaler notes that some regions legally require this, and an auditor role can grant permission to reveal the real names when an investigation calls for it.
If you're an IT leader rolling out a SWG in Europe or anywhere with a works council, this is the setting to have ready before anyone asks.
What Zscaler the company keeps
Zscaler's privacy FAQ is clear on a few points, and it's worth quoting accurately rather than guessing:
- Content is not stored for most services. Zscaler says that for the majority of its services, transaction content (messages, files and other substantive parts of a request) is never stored or written to disk, and inspection takes place in memory.
- Logs are kept. Once inspection is complete, Zscaler says no record of the source data is preserved beyond the log of the transaction itself. The log is the record your employer reviews.
- SSL inspection doesn't change storage. Zscaler states that enabling TLS/SSL inspection does not change the limited amount of data it processes or stores.
- Location follows the user. Zscaler processes traffic in whichever of its 150+ data centers is closest to the user, so a European employee traveling to the US is processed in a US data center. Depending on the product, customers can choose the log storage location.
- Support is global. Zscaler says it provides support from the EU, the US, India and Costa Rica.
None of that is sinister. It's a well-documented cloud service doing what cloud services do. But it does answer the question underneath the question: your decrypted traffic exists, briefly, in a third party's data center, and a log of it exists for as long as your retention policy says.
The real privacy question: where does decryption happen?
Every serious secure web gateway decrypts HTTPS. That's not optional anymore, because almost all web traffic is encrypted and a gateway that can't read it can't stop a malicious download or a sensitive upload. The difference between architectures is where the plaintext lives while it's being inspected.
With a cloud proxy like ZIA, your laptop encrypts the request to Zscaler, Zscaler decrypts it in its data center, inspects it, re-encrypts it and forwards it. For that moment, your banking session, your health portal and your HR benefits page are plaintext on Zscaler infrastructure. That's why many organizations bypass inspection for whole categories like health and finance: they'd rather not have that content decrypted off the device at all. It's also why cert-pinned apps and developer tools often end up on bypass lists, which Zscaler's own documentation acknowledges for apps like Microsoft 365, WebEx and Dropbox.
With an on-device SWG, the decryption happens on the endpoint. dope.security's Fly Direct architecture runs SSL inspection, URL filtering and Cloud Application Control in a lightweight agent on the laptop, so the decrypted content never leaves the device, and traffic goes straight to the internet. Our piece on on-device vs cloud SSL inspection and what it means for privacy goes deeper on why that matters for regulated data.
Zscaler vs dope.security on employee privacy
Here's how the two models compare on the questions employees and privacy teams actually ask. Both are legitimate security tools; they just put the sensitive part in different places.
- Where decryption happens: Zscaler decrypts in its cloud data centers; dope.security decrypts on the employee's own device.
- Where decrypted content travels: with Zscaler, plaintext exists in memory on Zscaler infrastructure during inspection; with dope.security, decrypted content stays on the endpoint.
- What the employer sees: both produce web activity logs your IT team can review in a console. On-device inspection doesn't make an employer blind; it removes the third-party data center from the content path.
- Sensitive categories: cloud proxies often bypass health and banking sites to avoid decrypting them off-device; dope.security can inspect consistently because the plaintext never leaves the laptop.
- Travel and data residency: Zscaler processes traffic in the data center nearest the user, wherever they are; dope.security processes traffic where the device is, with no detour.
- DLP content: Zscaler's data protection evaluates content in its cloud; dope.security's Dopamine DLP keeps the original file on the device and sends only the extracted text needed for classification, to an LLM with zero data retention.
The point isn't that Zscaler is careless. It's that a cloud proxy can only be as private as its data center path, and an on-device gateway changes the path itself.
What employees should assume on any work laptop
Whatever gateway your employer runs, the safe assumption is simple: web activity on a managed device is visible to your employer at the URL level, and sometimes at the content level for uploads. That's true with Zscaler, Netskope, Cisco, Palo Alto or dope.security.
Practical habits follow from that. Do personal banking and medical browsing on a personal device. Don't sign into personal email or personal AI accounts on a work laptop if you'd mind your employer seeing that you did. And if you want to know exactly what's logged and for how long, ask IT or your privacy office for the acceptable use and retention policy. A well-run security team will have a straight answer, and in many jurisdictions they're required to give you one.
For IT leaders, the mirror image applies. Publish what you log, how long you keep it, who can see it, and when names are revealed. Turn on name obfuscation where your rules call for it. Employees accept security tools far more readily when the rules are written down and the tool isn't a black box.
For IT leaders: privacy is now a selection criterion
Privacy used to be a footnote in SWG evaluations. In 2026 it's a deal shaper, for three reasons. Data residency rules keep tightening, and a proxy that processes a traveling employee's traffic in another country creates paperwork. Works councils and unions increasingly ask where decryption happens, not just whether it happens. And AI tools have turned the browser into the main path for sensitive data, so the content a gateway decrypts is more sensitive than ever.
That's the case for flying direct. dope.security inspects on the device, applies policy instantly from a single console, runs in under 100 MB of RAM, and doesn't backhaul traffic through a data center. Our analysis of data sovereignty and SSL inspection covers the residency side in detail. Teams that made the switch say the deployment was the easy part: Outreach Health secured 99% of devices within a week, as described in the Outreach Health customer story.
If you want to see on-device inspection for yourself, explore the dope.security Fly Direct secure web gateway or book a 20-minute demo.
The bottom line
Zscaler isn't spying on you; your employer is using Zscaler to see web activity, and Zscaler's documentation is upfront about what it processes and keeps. The sharper question is where your decrypted traffic goes while it's checked. A cloud proxy reads it in someone else's data center. An on-device gateway like dope.security reads it on the laptop and sends nothing but the verdict and the log, which is a better deal for employees and a simpler story for the privacy team.
Frequently Asked Questions
Can my employer see my browsing history through Zscaler?
Yes, at the URL level. Zscaler processes URLs, user IDs, IP addresses and directory groups, so an admin with reporting access can generally see which sites a user visited and what policy verdict applied. Some organizations enable Zscaler's user name obfuscation so admins see pseudonyms unless an auditor reveals the real name.
Does Zscaler store the content of my traffic?
Zscaler says that for the majority of its services, transaction content such as messages and files is never stored or written to disk, and inspection happens in memory. It keeps the log of each transaction. The content is still decrypted on Zscaler infrastructure during inspection, which is the key difference from on-device inspection.
Can Zscaler see my passwords or banking details?
When SSL inspection is enabled for a site, the gateway can technically read the decrypted request during inspection. Many organizations bypass inspection for banking and health categories to avoid that. dope.security inspects on the device, so decrypted content never leaves the laptop, which removes the third-party data center from the equation.
Does Zscaler work when I'm on my home Wi-Fi?
Usually, yes. The Zscaler Client Connector forwards traffic to Zscaler's cloud from any network, so policy and logging follow the device home. The same is true of dope.security, which enforces policy on the device on any network without routing traffic through a data center.
Is on-device SSL inspection more private than a cloud proxy?
It changes who handles the plaintext. With on-device inspection, decrypted content stays on the endpoint and only logs and verdicts reach the console. With a cloud proxy, decrypted content exists briefly in the vendor's data center. Your employer still sees web activity either way, so on-device inspection is about third-party exposure and data residency, not hiding activity from IT.
How do I find out what my company logs?
Ask IT or your privacy office for the acceptable use policy and log retention policy. Zscaler retains logs according to its published log retention terms and lets customers choose log storage location for some products, so your company can tell you exactly where logs live and for how long.


.jpeg)
.jpeg)

