Cyberpediaby Brain

SASE and secure web gateways for remote work

Away from the office, your web traffic may pass through a company service that filters dangerous sites. What you see and what to do.

On this page
  1. 1In 30 seconds
  2. 2What it is
  3. 3Straight answers
  4. 4How it works
  5. 5What you do
  6. 6Pitfalls
  7. 7Sources
  8. 8Cite this article

In 30 seconds

Where it happens
When working from home, travelling or at the office, whenever the company has set up an SWG or a SASE service. A cloud-hosted SWG can secure users' internet access wherever they are[1].
What the attacker wants
To get you to open a phishing site or a site that distributes malware. Address filtering can, for example, prevent visits to fake sites that imitate legitimate companies[1].
The reflex
If a site is blocked, read the message shown, do not try to get around the system, and report to IT support a block that looks unjustified.
If it is already done
If you changed your device settings to bypass a block, put them back to their original state and tell support.

What it is

When you work from home, your computer is no longer protected by the office network. A secure web gateway (SWG) restores a control point: according to Microsoft, it is a solution that protects the organization from web-borne threats and acts as the passage point for all internet traffic, which it analyzes and filters[1].

The SASE model (secure access service edge) goes further. Microsoft describes it as a way to deliver networking and security through a single cloud-hosted service that combines, among other things, secure web gateways and zero trust network access[2]. For an employee, the consequence boils down to this: a site can be blocked or allowed according to company rules, wherever you are.

Straight answers

What is a secure web gateway?

It is an intermediary between employees and the internet. It intercepts traffic headed for the web and checks that it complies with the organization's security rules, then blocks harmful content and unauthorized access[1]. It handles outbound traffic, the traffic you send to sites, whereas a firewall protects the network perimeter[1].

What is SASE?

SASE brings network and security functions together in a single cloud service. Microsoft lists among its components the secure web gateway, the cloud access security broker (CASB), firewall as a service and zero trust network access (ZTNA)[2]. SSE (security service edge) is the subset of SASE that contains only the security functions[2].

Does my traffic really go through the company?

For its own solution, Microsoft states that web filtering protects users and their devices, whether they are remote or inside the company network. For an unencrypted HTTP site, the service relies on the full address; for an HTTPS site, it relies on the server name (SNI)[3]. The UK's NCSC considers that sending all internet traffic through a central service may bring only a limited additional security benefit, since most modern services already use TLS, and that it should not be the default choice[4].

Does SASE replace the VPN?

Not automatically. The NCSC warns against replacing a VPN with a SASE or SSE service that would leave the access model unchanged: the user authenticates at connection and then gets broad access to the network. If an account is compromised, the attacker can inherit that access, and the situation differs little from a traditional network[4]. CISA, together with the FBI and New Zealand and Canadian agencies, nonetheless recommends moving toward solutions such as zero trust, SSE and SASE, which give more visibility into network activity[5]. ANSSI specifies that the zero trust model must on no account be seen as a replacement for perimeter defense[6].

What do you see when a site is blocked?

It varies. The NCSC describes three cases for protective DNS filtering: a block page explaining that the domain is blocked (the browser may then display a certificate error), a redirection to a server that provides a custom response, or an NXDOMAIN response that shows an error page and does not say the block comes from the service[7].

How it works

The gateway applies three mechanisms described by Microsoft. It intercepts traffic and compares it with the organization's rules. It inspects content, for example with anti-malware analysis and a sandbox where suspicious files are isolated. Finally, it enforces company rules, for example a ban on accessing certain sites[1]. It can also limit the web applications that can be used to those that are approved[1].

Address filtering relies on lists of domains to block. With protective DNS, these lists come from threat intelligence sources, and the company can keep an allow list that overrides the block list so that useful services do not become unreachable[7]. At Microsoft, filtering rules are grouped into security profiles, which are themselves linked to conditional access rules specific to each user. Rules are processed in priority order, and a baseline profile applies to all traffic as a last-resort rule[3].

For devices away from the office, a protective DNS service can offer a small client program to install on the device, so that it always queries the service's resolver[7]. NIST cites SASE as an example of an evolving wide area network (WAN) infrastructure that provides a complete set of security services[8].

What you do

  • Read the block page to the end: it may contain an explanation and information about the service that blocked the site[7].
  • Do not change your device's DNS resolver and do not modify your browser settings to avoid the filter. A company can block DNS requests that do not go through its resolver, and disable encrypted DNS (DoH) or route it through that resolver[7].
  • If a site you need is blocked, write to IT support and name the site concerned. The NCSC recommends that companies provide clear procedures to resolve such access problems quickly and give support the visibility needed to diagnose a block caused by the rules[4].
  • Know that the IT team can adjust the rules. Microsoft cites regular policy tuning and allowing trusted sites among the ways to reduce false positives[1]. The NCSC states that administrators should be able to add domains to the allow list through a user interface[7].

Pitfalls

  • An error page with no explanation. With an NXDOMAIN response, you will not know that the filtering service blocked the site[7]. If a site you know does not open at all, report it to support.
  • A wrongful block. ANSSI notes that zero trust access rules are complex to define and can wrongly block legitimate access[6]. Microsoft, for its part, writes that wrongly flagged traffic can disrupt work[1].
  • A false sense of security. According to the NCSC, a product alone does not deliver zero trust, which is an architecture approach[4]. A filtering service does not excuse you from staying alert to a suspicious message or link.
  • Inspection of encrypted traffic. Some gateways inspect encrypted traffic[1]. The NCSC specifies that where such inspection is deployed, organizations place a significant amount of trust in the inspecting component and in the organization that operates it, and that where inspection is necessary it should be done deliberately and transparently[4].
  • Requests visible to the service. A protective DNS service sees all DNS requests made from the network[7]. You can ask your IT team what data is retained.

Sources

  1. Microsoft, “What is a secure web gateway (SWG)?”. https://www.microsoft.com/en-us/security/business/security-101/what-is-secure-web-gateway-swg (accessed 14 November 2026)
  2. Microsoft, “What is secure access service edge (SASE)?”. https://www.microsoft.com/en-us/security/business/security-101/what-is-sase (accessed 14 November 2026)
  3. Microsoft, “Learn about Microsoft Entra Internet Access for all apps”. https://learn.microsoft.com/en-us/entra/global-secure-access/concept-internet-access (accessed 14 November 2026)
  4. NCSC (UK), “Zero Trust Network Access (ZTNA)”. https://www.ncsc.gov.uk/sites/default/files/2026-05/NCSC-Zero-Trust-Network-Access-(ZTNA)_0.pdf (accessed 14 November 2026)
  5. CISA, “Modern Approaches to Network Access Security”, 18 June 2024. https://www.cisa.gov/resources-tools/resources/modern-approaches-network-access-security (accessed 14 November 2026)
  6. ANSSI, “Modèle Zero Trust”, 20 June 2025. https://messervices.cyber.gouv.fr/documents-guides/anssi-fondamentaux-zero-trust-v1.0.pdf (accessed 14 November 2026)
  7. NCSC (UK), “Protective DNS for the private sector”, 9 June 2021. https://www.ncsc.gov.uk/guidance/protective-dns-for-private-sector (accessed 14 November 2026)
  8. NIST, “Guide to a Secure Enterprise Network Landscape”, 10 November 2022. https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-215.pdf (accessed 14 November 2026)