What Is ZTNA? How Zero Trust Network Access Fits Zero Trust (and When You Still Need an SWG)

What Is ZTNA? How Zero Trust Network Access Fits Zero Trust (and When You Still Need an SWG)

Last updated: November 2026

Short answer: ZTNA (Zero Trust Network Access) gives a user access to one specific private application, and only after checking who they are, what device they're on, and whether policy allows it. It never puts that user on the corporate network the way a VPN does.

Definition: Zero Trust Network Access (ZTNA) is a way of reaching private, internal applications where each connection is brokered app by app, based on identity and device context, instead of granting access to the whole network.

ZTNA is not the same thing as "Zero Trust," and it is not a Secure Web Gateway (SWG). Zero Trust is the principle. ZTNA applies that principle to private apps. An SWG applies it to outbound web and SaaS traffic. Most organizations need more than one of these, and this guide explains which does what.

We make dope.security, an on-device Secure Web Gateway, so read this with that in mind. One caveat up front: dope.security does not ship ZTNA today. Where we talk about our product below, it's about the web and SaaS side, not private-app access.

Key takeaways

  • ZTNA is per-app access to private applications. A user is connected to one app at a time, after identity and device checks. They don't get a route to the rest of the network.
  • Zero Trust is a model, not a product. NIST defines it as a set of concepts for making least-privilege, per-request access decisions "in the face of a network viewed as compromised."
  • ZTNA is one piece of Zero Trust. It covers access to internal apps. It does not filter web traffic, inspect SaaS uploads, or stop data from leaving through a browser.
  • A VPN puts the device on the network. ZTNA connects the user to an app. That difference is why ZTNA limits lateral movement. For the full comparison, see ZTNA vs VPN.
  • An SWG and ZTNA solve different problems. If your users reach internal apps, you need ZTNA or a VPN. If they use the web, SaaS, or AI tools, you need web and SaaS controls too. Most teams need both.

What is Zero Trust?

Zero Trust is a security model built on one assumption: being on the network, or owning the device, does not earn trust by default. Every request for a resource gets checked, and access is limited to what the task needs.

The reference definition comes from NIST SP 800-207, Zero Trust Architecture (August 2020):

"Zero trust (ZT) provides a collection of concepts and ideas designed to minimize uncertainty in enforcing accurate, least privilege per-request access decisions in information systems and services in the face of a network viewed as compromised."

NIST also states plainly that zero trust "assumes there is no implicit trust granted to assets or user accounts based solely on their physical or network location … or based on asset ownership." Among its basic tenets: access to individual resources is granted per session, access is decided by dynamic policy (including the identity of the client, the application, and the state of the requesting device), and authenticating to one resource does not automatically grant access to another.

A zero trust architecture (ZTA) is the plan that puts those principles to work. NIST describes it as "an enterprise's cybersecurity plan that utilizes zero trust concepts and encompasses component relationships, workflow planning, and access policies." In NIST's model, a policy decision point decides whether a request is allowed, and a policy enforcement point carries the decision out.

For a practical roadmap, the CISA Zero Trust Maturity Model, Version 2.0 (April 2023) organizes the work into five pillars (Identity, Devices, Networks, Applications and Workloads, and Data) and three cross-cutting capabilities (Visibility and Analytics, Automation and Orchestration, and Governance). It tracks progress through four stages: Traditional, Initial, Advanced, and Optimal. CISA's Zero Trust topic page sums it up as "Never Trust. Always Verify."

The idea isn't new. Forrester analyst John Kindervag popularized the term "Zero Trust model" in 2010 (Wikipedia has the history). What changed is that remote work, personal devices, and cloud apps made the old "inside the firewall means trusted" assumption impossible to defend. NIST names those same trends as the reason zero trust exists.

The takeaway for buyers: zero trust is a model you implement across several controls. No single product "is" Zero Trust, including ZTNA.

What is ZTNA (Zero Trust Network Access)?

ZTNA is the part of Zero Trust that governs how people reach private applications: internal web apps, admin consoles, file servers, and services running in a data center or a private cloud network.

Gartner's glossary defines it this way:

"Zero trust network access (ZTNA) is a product or service that creates an identity- and context-based, logical access boundary around an application or set of applications. The applications are hidden from discovery, and access is restricted via a trust broker to a set of named entities. The broker verifies the identity, context and policy adherence of the specified participants before allowing access and prohibits lateral movement elsewhere in the network."

In plain terms, three things set ZTNA apart:

  1. Access is per app, not per network. A contractor who needs the design review tool gets that tool. They don't get a route to the finance server on the same subnet.
  2. Apps are hidden. Private apps don't sit on the open internet waiting to be scanned. They're reachable only through the broker, and only by users the policy names.
  3. Every connection is checked. Identity (usually through your SSO provider with MFA), device posture (is it managed, encrypted, patched?), and context are evaluated before the session starts.

How does ZTNA work?

The details vary by vendor, but most ZTNA products share the same parts:

  • An identity provider (Okta, Microsoft Entra ID, Google, and others) confirms who the user is.
  • A policy engine and broker decides whether this user, on this device, can reach this app right now. That's NIST's policy decision point, often run as a cloud service.
  • A connector sits next to the private app, in the data center or cloud network. It usually makes an outbound-only connection to the broker, so the app doesn't need an inbound port open to the internet.
  • A client on the device, or for agentless ZTNA, a browser. Agent-based ZTNA can check device posture in more depth. Agentless (browser-based) ZTNA is easier to give to contractors and unmanaged devices, but it's generally limited to web apps.

When a request passes, the broker stitches the user to that one app. When it fails, the user doesn't see the app at all.

ZTNA vs Zero Trust vs SWG vs VPN: what's the difference?

These four terms get used interchangeably, and they shouldn't be. Here's how they compare:

What it is What it protects How access or control works What it doesn't do
Zero Trust A security model (NIST SP 800-207) Everything: users, devices, apps, data Verify every request; least privilege; assume the network is hostile Isn't a product you can buy on its own
ZTNA An access control for private apps Internal and private-cloud applications Brokered per-app connection after identity and device checks Doesn't filter web traffic or inspect SaaS and AI uploads
SWG (Secure Web Gateway) An inspection control for outbound web traffic Users and data on the internet, SaaS, and AI tools URL filtering, SSL inspection, tenant controls, DLP on web traffic Doesn't give access to apps behind your firewall
VPN A network tunnel Remote access to the corporate network Encrypted tunnel that places the device on the network Doesn't limit access app by app; broad trust once connected

The split shows up in vendor product lines, too. Zscaler, for example, sells Zscaler Internet Access (its SWG) and Zscaler Private Access (its ZTNA) as separate products. We walk through what that means in Is Zscaler a VPN?. Some vendors call the web-facing side "zero trust internet access." The label varies, but the job is the same: inspect what users do on the open internet and in SaaS.

Two practical consequences follow:

  • ZTNA decides who can reach an app. It doesn't watch what happens next on the web. A verified employee on a healthy laptop can still paste a customer list into a personal AI account. That traffic never touches a private-app broker. We go deeper on this in Identity Is Not Enough: Why Zero Trust Needs On-Device Context.
  • An SWG doesn't replace ZTNA or a VPN. URL filtering and SSL inspection don't get anyone into an internal app.

If you're weighing ZTNA against your current VPN, our ZTNA vs VPN guide covers when to replace, when to layer, and how to migrate.

What is the purpose of ZTNA?

The purpose of ZTNA is to let the right people reach private applications without exposing the rest of the network. In practice, teams adopt it to:

  • Limit lateral movement. If a laptop or account is compromised, the attacker reaches only the apps that identity was allowed to use, not the whole network.
  • Shrink the attack surface. Private apps sit behind a broker instead of an internet-facing VPN gateway or open inbound port.
  • Scope access for contractors and third parties. Per-app policies are easier to grant, audit, and revoke than network-level VPN access.
  • Tie access to device health. A device that fails posture checks can be blocked from sensitive apps, not just from "the network."

What are the limits of ZTNA and Zero Trust?

Zero Trust is the right direction, but it isn't free, and ZTNA isn't a complete answer. The honest limits:

  • ZTNA covers access, not data. Once a user is in an app, or on the open web, ZTNA's job is mostly done. Uploads, AI prompts, and personal SaaS accounts need web and data controls.
  • Identity is a big dependency. NIST's own threat list for zero trust architecture includes stolen credentials and insider threats, denial of service against the policy components, and reliance on proprietary vendor formats (NIST SP 800-207, Section 5). If the identity provider or broker goes down, access goes with it.
  • Legacy apps are hard. Apps that need broad network access, or that can't sit behind a broker cleanly, often stay on a VPN for a while.
  • It's a migration, not a switch. Inventorying apps, deploying connectors, and writing per-app policy takes time. CISA's maturity model exists because most organizations move in stages.

Do you need ZTNA if you already have an on-device SWG?

Yes, if your users need to reach private apps behind the firewall. An SWG, including ours, doesn't provide that access.

And ZTNA doesn't replace your SWG. ZTNA doesn't do URL filtering, SSL inspection of web traffic, SaaS tenant control, or DLP on uploads to web and AI tools. Those are SWG and data-protection jobs.

Here's where dope.security fits, stated plainly:

  • What we do today: dope.SWG is an on-device Secure Web Gateway for managed Mac and Windows devices. URL filtering, SSL inspection, and Cloud Application Control run on the device, and traffic goes straight to its destination instead of stopping at a vendor data center. That's what we call Fly Direct. Dopamine DLP adds AI-based inspection of uploads for PII, PCI, PHI, and intellectual property. This is Zero Trust for web and SaaS traffic.
  • What we don't do today: dope.security has no built-in ZTNA. dope.SWG doesn't give users access to private apps, and it isn't a substitute for ZPA, Prisma Access, Cloudflare Access, or your VPN.
  • What's coming: dope.vpn is listed on our site as coming soon, for point-to-point connections to behind-the-firewall applications. It isn't generally available, and we're not giving a date here.

So if you need private-app access now, keep or choose a ZTNA product or VPN, and run an SWG alongside it. If your users mostly live in the browser, SaaS, and AI tools, the web and SaaS layer is often where the bigger day-to-day risk sits. That side can run on the device. Our Secure Web Gateway deployment guide covers what an on-device SWG does and doesn't cover, and Beyond ZTNA explains our view on where zero trust enforcement is heading.

If you want to see the web and SaaS side running on your own devices, book a 20-minute demo and bring your current ZTNA or VPN setup. We'll show you how the two run side by side.

Frequently asked questions

What is ZTNA in simple terms?
ZTNA (Zero Trust Network Access) connects a user to one specific private application after checking their identity, device, and policy. It doesn't give them access to the rest of the network.

What does ZTNA stand for?
ZTNA stands for Zero Trust Network Access.

What is the difference between VPN and ZTNA?
A VPN places a remote device on the corporate network, and the device can then reach whatever the network allows. ZTNA connects the user to individual apps, one at a time, after identity and device checks. See our full ZTNA vs VPN comparison.

Is ZTNA replacing VPN?
For access to many private apps, yes, gradually. Most organizations run ZTNA and VPN side by side during the move, and they keep a VPN for legacy apps and some admin access longer.

What is the purpose of ZTNA?
ZTNA lets authorized users reach private applications without exposing the network. That limits lateral movement and keeps internal apps hidden from the internet.

Is ZTNA the same as Zero Trust?
No. Zero Trust is a security model that applies to users, devices, networks, apps, and data. ZTNA is one control that applies Zero Trust to private-app access.

Is an SWG the same as ZTNA?
No. An SWG inspects and filters outbound web and SaaS traffic. ZTNA brokers access to private, internal applications. They solve different problems, and many organizations need both.

What are the disadvantages of zero trust security?
Zero trust takes time to roll out, depends heavily on your identity provider and policy systems, and is hard to apply to legacy apps. NIST lists stolen credentials, denial of service against policy components, and vendor lock-in among its risks.

Is ZTNA part of SASE or SSE?
Yes. ZTNA is usually counted as one component of SSE (Security Service Edge), alongside SWG, CASB, and DLP. SASE adds networking (SD-WAN) on top. See What Is SSE? and SSE vs SASE.

Does dope.security offer ZTNA?
Not today. dope.security provides an on-device Secure Web Gateway for web and SaaS traffic. dope.vpn, for point-to-point access to behind-the-firewall apps, is listed as coming soon.

Related reading

Questions about where an on-device SWG fits next to your ZTNA or VPN? Email sales@dope.security.

Zero Trust
Zero Trust
SSE
SSE
Secure Web Gateway
Secure Web Gateway
Remote Work Security
Remote Work Security
← back to blog Home