How Zero Trust Network Access (ZTNA) Protects Modern Enterprises

How Zero Trust Network Access (ZTNA) Protects Modern Enterprises

Key Takeaways

  • Zero trust network access (ZTNA) connects each user to specific applications after checking identity and device health, instead of placing the user on the corporate network the way a virtual private network (VPN) does.
  • ZTNA limits how far an attacker can move after one login is stolen, and it removes the exposed VPN server as a target.
  • ZTNA decisions depend on cloud identity and access management (IAM), so weak identity data produces weak access policy.
  • Buying ZTNA is the easy part. Designing application policies and tuning device checks, then keeping policy current as apps and users change, is where most rollouts stall.
  • Managed ZTNA hands that ongoing work to a provider, and Red River backs it with Zscaler Delivery Services and Managed Security Services Partner authorizations.

A traditional VPN gives anyone who signs in a path onto the internal network. For a contractor who needs one application, that’s far more access than the job requires. For an attacker holding a stolen password, it’s an open door.

Zero trust network access alleviates risk by connecting people to individual applications instead of to the network. Managed security providers often take on the daily work of running it, from writing policies to keeping them current as applications and users change.

How does ZTNA protect an enterprise, and why do so many rollouts stall after the purchase? This article answers both questions and shows where a managed provider fits.

What Is ZTNA and How Does It Work?

ZTNA is a service that gives each user access to individual applications instead of the whole network. For every request, it checks who is asking and whether their device is healthy before it opens a connection. A finance analyst, for example, might reach the accounting system from a managed laptop but not from a personal tablet.

A request moves through the same steps every time:

  1. The user asks for an application: The request goes to the ZTNA service, not straight to the application.
  2. The service checks identity: It confirms who the user is and which groups they belong to, and it notes how they signed in.
  3. The service checks the device: It confirms the device meets your security standards, such as encryption and current patches.
  4. The service applies the policy for that application: Each application has its own rules about who can reach it and from which devices.
  5. The service connects the user to that application only: Nothing else on the network becomes visible.

Two inputs drive those checks. The identity system supplies the user’s details, and endpoint management tools report the device’s health. ZTNA doesn’t replace either one. It combines them into a single decision for each request.

Because of that last step, the application never faces the internet. A broker sits between the user and the application, and in most designs the application connects out to the broker instead of listening for connections. With no open inbound port to scan, an attacker has far less exposed network to find.

The National Institute of Standards and Technology (NIST) describes the principle behind this in Special Publication 800-207: no user or device earns trust just because of where it sits on the network.

ZTNA vs VPN: How Do They Differ?

A VPN builds an encrypted tunnel from a user’s device to the corporate network. Once the tunnel is up, the device sits on that network and can usually reach whatever the network allows.

VPN ZTNA
Access scope The whole network or large parts of it One application at a time
Attack surface Internet-facing VPN servers that attackers can probe Applications stay hidden behind a broker, with no open inbound ports in most designs
User experience Users connect manually, and traffic often detours through a central data center Users reach applications directly through the nearest cloud service, often without noticing
Scalability Limited by VPN server capacity and licenses Scales as a cloud service when users and applications grow

The user experience gap shows up in daily work. VPN users often reconnect after a network change or a timeout, and remote traffic can detour through a central data center before it reaches a cloud application. ZTNA sends each request toward its destination more directly, which usually feels faster and asks less of the user.

Scalability matters most during traffic surges. A VPN has a fixed number of connections it can handle, and an unplanned spike in remote work can overload it. A cloud-delivered ZTNA service adds capacity without a hardware project.

The National Security Agency (NSA) and the Cybersecurity and Infrastructure Security Agency (CISA) describe the VPN’s weak spot in joint guidance: remote access VPN servers are entry points into protected networks, which makes them targets. The agencies wrote this document with leaders in the defense industrial base in mind, and it notes that multiple nation-state attackers have weaponized known vulnerabilities to reach vulnerable VPN devices.

Where Does ZTNA Fit with SASE and SSE?

Secure Access Service Edge (SASE) combines cloud-delivered networking, such as software-defined wide area network (SD-WAN) services, with security services in one platform. Security Service Edge (SSE) is the security half of that package, without the networking piece. ZTNA is one of the services inside SSE. Vendors package these services differently, so ask what a bundle includes before assuming ZTNA is part of it.

Many enterprises start with ZTNA because the VPN is the one part of their setup that attackers can reach from the internet. From there, they can add the other services in a SASE platform as their needs grow.

How Does ZTNA Protect Against Real Attacks?

Five situations show where ZTNA earns its place in most enterprise organizations:

  1. Limiting lateral movement: When an attacker steals a login, getting in is the first problem and moving around is the second. On a VPN, a stolen login can reach every system the tunnel allows. With ZTNA, the same login reaches only the applications that user can use, so the attacker stays boxed in.
  2. Removing the exposed VPN server: A VPN server listens for connections from the internet, which is why attackers scan for them. ZTNA applications connect outward to a broker, so there’s no VPN server to patch in a hurry and no open port to find.
  3. Securing contractor and third-party access: A contractor might need one application for a few weeks. ZTNA lets you grant exactly that, and the access can end when the contract does. Our look at the 2013 Target breach shows the alternative: stolen vendor credentials combined with overly broad vendor reach.
  4. Ending access when a device falls out of compliance: A VPN checks a device when the tunnel opens and rarely again. ZTNA can check posture continuously, so a laptop that loses its encryption or turns off its security software loses access to the application until someone fixes it.
  5. Handling unmanaged devices: Employees and partners sometimes work from devices you don’t control. Agentless ZTNA, covered next, lets them reach approved web applications through a browser while the rest of your environment stays out of reach.

These protections work together, because every request meets the same wall: the policy for one application.

Agent-Based or Agentless ZTNA: Which Do You Need?

Agent-based ZTNA installs software on individual devices. That software can check device health continuously and supports a wide range of applications, so it suits company laptops.

Agentless ZTNA requires no installation. Users open an approved application through a browser, so contractors and employees on personal devices can get to it without installing anything. An outside auditor, for example, can review one reporting system for a week and then lose access. The tradeoff is that the service sees less about the device and supports fewer kinds of applications.

Most enterprises use both: the agent for managed devices and the browser for everyone else. The agent also becomes one more piece of software to keep updated, which belongs in your device management routine. A short pilot with each group shows will show you where the gaps are. An employee who works in a desktop application or a database tool, for example, usually needs the agent, because a browser can only reach web applications.

Why Does ZTNA Depend on Cloud Identity and Access Management?

Every ZTNA decision starts with identity. Cloud identity and access management is the system that knows who each user is and which groups they belong to, and it records how they signed in. ZTNA reads that information on every request.

That makes identity data the foundation of your policy. Stale accounts and overly broad groups don’t disappear when you add ZTNA, and neither does inconsistent multifactor authentication (MFA). They show up as access you never meant to grant.

Red River’s IAM blog series explains why identity comes first in zero trust and includes a step-by-step framework for implementing IAM. Device signals matter here too. A policy that says only compliant devices can reach payroll needs a reliable record of which devices are compliant, which comes from your endpoint management tools. Policy can only be as accurate as the data it reads.

As a best practice, clean up identity before you write ZTNA policy. Otherwise, the mess in your accounts and groups carries straight into the new system.

Why Do ZTNA Rollouts Stall After the Purchase?

Buying ZTNA is the easy part. The product installs, a pilot group connects and the demo works. The hard work starts when the organization decides, application by application, who should reach what.

Policy design comes first, and it starts with a list most enterprises don’t have. Few organizations know every application their users rely on, so teams often discover the forgotten ones when someone complains. Once the list is complete, each application needs rules for who can reach it and from which devices.

Device checks need tuning. Rules that are too strict lock out healthy laptops and flood the help desk, while rules that are too loose leave the checks feeling little more than decorative.

Then the environment changes. New applications arrive and departments reorganize while contractors come and go, and each change needs a policy update. Under deadline pressure, teams add a broad exception to unblock someone and forget to remove it, and months later the policy resembles the VPN it replaced.

Users complain when ZTNA blocks something that worked on the VPN. A laptop fails a device check, or an application nobody listed isn’t in the policy yet, and the person can’t reach what they need. To them it looks like an outage, so the help desk loosens a rule to fix it, and without a clear owner those quick fixes pile up.

Most internal teams handle the first month well. The strain may show up in month six, when the project team has moved on and nobody owns the policy.

How Does a Managed ZTNA Provider Handle the Ongoing Work?

How Does a Managed ZTNA Provider Handle the Ongoing Work

Managed ZTNA hands that ongoing work to a provider, which builds your policies and keeps them current. A managed provider handles the recurring work:

  • Application inventory: The provider lists every application and maps who needs each one.
  • Access policies: Each application and user group gets its own written rules.
  • Device checks: The provider tunes posture rules so they block risky devices without locking out healthy ones.
  • Onboarding and retirement: The provider adds new applications to policy and retires old ones as the business changes.
  • Access reviews: On a set schedule, the provider removes stale accounts and old exceptions.
  • Monitoring and response: Analysts watch logs and alerts around the clock and respond to suspicious access.

Ask your provider to report on how the program is doing. Two numbers tell you the most: how many applications have a written access policy, and how many exceptions have stayed open past your limit. A good report also shows who had access to which application and which requests the system denied, along with the reasons, and that record helps during audits and incident reviews.

Your team keeps the decisions that depend on business knowledge, such as which applications are critical and who approves exceptions. The provider handles the volume and the vigilance. When you compare providers, ask how each one onboards a new application and how quickly it changes policy on request. Ask, too, who answers when a user can’t connect at night. Ask early whether a provider will run the platform you already own or require a switch to its own.

Where Should Your ZTNA Rollout Start?

Start with the applications, not the product. Build a complete list first, since you can’t write policy for an application nobody has named, then rank the list by risk and move the most exposed application first. A finance system is a common first choice, and it leaves your team with a working policy to copy.

Run the identity cleanup alongside the inventory instead of after it, since the migration moves faster that way. Keep the VPN for exceptions until the migration ends, and decide before launch who will own policy afterward.

Why Choose Red River for Managed ZTNA?

Red River holds Zscaler Delivery Services and Managed Security Services Partner authorizations. Zscaler awards these designations only to select partners that show proven technical expertise and a commitment to customer success, and the program includes a training track so partners are ready to serve customers on Zscaler’s Zero Trust Exchange platform. That preparation shows up most in the policy work, where small mistakes either block people who need access or leave doors open.

Zscaler also named us its Americas Public Sector Partner of the Year, which reflects our public sector work.

We serve commercial and government customers, so the policies we design work in environments with strict compliance requirements as well as in ordinary enterprises.

Our Zero Trust Lab gives customers and partners a cloud-hosted environment to test and validate multi-vendor security technologies, including SASE and identity access management, before they commit to a design.

ZTNA replaces the VPN’s broad access with application-level access, and it works only as well as the identity data and the policy behind it. We help enterprises build that policy and keep it current, so the access layer stays accurate long after the project team has moved on. If your VPN still reaches more of your network than anyone intends, contact us to talk through how a managed ZTNA approach would work in your environment.

Frequently Asked Questions

Does ZTNA fully replace a VPN?

Not always on day one. ZTNA covers most application access, but some environments still need a VPN for legacy protocols or for site-to-site connections, which link two offices. Those connections are the usual holdout. Where a VPN remains, harden it using the NSA and CISA guidance and shrink its reach as more applications move to ZTNA.

How long does a ZTNA migration take?

It depends on how many applications you have and how clean your identity data is. Most organizations start with a pilot group and a small set of applications, which often goes live within weeks. Moving every application, retiring the VPN and tuning policy usually takes months, and larger or older environments take longer. Application discovery tends to take the most time, because every forgotten application surfaces only when someone tries to use it.

Does ZTNA work for on-premises applications?

Yes. A small connector installed in your data center connects outward to the ZTNA service, so users can reach on-premises applications without those applications sitting on the internet. Many older applications can sit behind a connector too, though protocol support varies by product, so test the legacy ones early.

written by

Corrin Jones

Corrin Jones is the Director of Digital Demand Generation. With over ten years of experience, she specializes in creating content and executing campaigns to drive growth and revenue. Connect with Corrin on LinkedIn.

Go to Top