Skip to content
Back to guides
Guide

Allow a phishing simulation in Google Workspace: 2026 guide

Get a phishing simulation (GoPhish or a platform) into Gmail without spam or warnings: spam bypass by organizational unit, content compliance, Chrome Safe Browsing, checks. Updated November 18, 2026.

Thomas Ferreira 10 min read

To allow a phishing simulation in Google Workspace, add a Spam rule in the Admin console (Apps, Google Workspace, Gmail, Spam, Phishing and Malware) that bypasses spam filters for an address list containing the tool’s sending domain, keep sender authentication on, and apply it to the organizational unit you test. Google has no setting or help page dedicated to simulations: this method, or a content compliance rule on the sending IP, is the one that follows its recommendations most closely.

Updated November 18, 2026. Every Google setting quoted here was checked that day in the Google Workspace Admin Help, which has moved from support.google.com to knowledge.workspace.google.com (pages dated November 10, 2026). Where Google says nothing, we quote a vendor’s guide and say so. The method works for GoPhish as well as for a hosted platform.

nophi.sh values for Google Workspace

  • Sending IPs: 85.208.144.52 and 2a09:8740:0:3::52
  • Sending domain: the one your campaigns use, shown in the console, Settings › Sending simulations. Simulations leave from it with SPF, DKIM and DMARC aligned, so they pass the authentication Google asks you to keep.
  • Domain of simulation links and pages: app.nophi.sh

The console recognises Google Workspace from your MX records, shows these values, then sends you a real test email through the same path as the simulations. You confirm delivery with one click on “I received this email”.

What Google provides for a phishing simulation

Nothing specific. No page of the Google Workspace Admin Help covers phishing simulations (search of the help site map, November 18, 2026). You work with three general Gmail settings instead:

SettingScopeWhat it doesOur view for a simulation
Spam rule with approved senders listOrganizational unitBypasses spam filters for senders or domains on the listRecommended, with authentication on
Content compliance ruleOrganizational unitAction “Bypass spam filter for this message” based on source IP, a header or authenticationGood alternative, more precise on the IP
Email allowlistEntire domainMail from these IPs isn’t marked as spamAvoid; Google advises against it

None of these turns off advanced phishing and malware protection. Google says so: “even if you’ve listed a domain as a safe sender in spam settings, the enhanced security features are still applied” (advanced protection). That is a good thing: an email spoofing your sending domain would still be checked.

Method 1: bypass spam filters for approved senders

This is the method in Google’s help page Add custom spam filters to Gmail. It needs the Gmail Settings administrator privilege.

  1. In the Admin console (admin.google.com), open Menu, Apps, Google Workspace, Gmail, then Spam, Phishing and Malware.
  2. On the left, select the organizational unit to test. Custom spam filters are set per organizational unit, and child units inherit them; they can’t be applied to groups.
  3. In the Spam section, click Configure, or Add another rule if one exists. Give the rule a clear name, for example “Phishing simulation, tool X”.
  4. Check Bypass spam filters for messages from senders or domains in selected lists, then pick an address list or create one.
  5. In the address list, add the tool’s sending domain and leave Require sender authentication on. Google always recommends it: “We recommend you always leave the sender authentication option on to help protect your domain from spoofing” (address lists).
  6. Save, then check that the rule appears in the Spam settings table. Google warns that “Changes can take up to 24 hours but typically happen more quickly.”

Two notes. First, the Gmail settings health check advises against putting whole domains in an approved senders list, because of spoofing risk (Gmail settings health). With authentication on, a message matches the list only if it passes DMARC, which rules out a sender spoofing the domain. Keep the rule to the tested unit and to the tool’s dedicated sending domain, never a consumer domain.

Second, the basic option leaves a banner: “This message wasn’t sent to Spam because of your organization’s settings.” With Bypass spam filters and hide warnings for messages from senders or domains in selected lists, Google says warnings about possible suspicious messages aren’t displayed. Test both on a mailbox you control before you choose.

Method 2: a content compliance rule on the sending IP

Several simulation vendors, KnowBe4 among them in its help center (header match, edited March 11, 2025), use a content compliance rule. Google documents each option used (content compliance rules), but not their use for a simulation.

  1. Open Menu, Apps, Google Workspace, Gmail, then Compliance. Select the organizational unit on the left.
  2. Hover over Content compliance and click Configure (or Add another). Enter a description.
  3. Check Inbound messages.
  4. Add an expression of type Metadata match: attribute Source IP, condition Is within the following range, then the tool’s sending IP. Google says Source IP “represents the IP address of the sending mail server”. To combine, add a second expression Message authentication, Message is authenticated, and require all conditions to match.
  5. Under the action, check Bypass spam filter for this message, then save.

The rule only covers messages from the tool’s IPs, in the unit you chose. Google’s help on address lists even suggests this kind of rule for a small volume of legitimate unauthenticated mail. With GoPhish, a common variant matches a custom header added by the sending profile; a fixed IP is harder to imitate than a header.

The email allowlist and the inbound gateway

The email allowlist, in Spam, Phishing and Malware, takes public IPs, single or in CIDR notation: “Messages from these addresses won’t be marked as spam by Gmail” (email allowlists). Three limits: “Email allowlists are always applied to your entire domain”, the page does not mention IPv6, and Google’s health check recommends not configuring them to reduce spoofing risk. The Alert Center even has a dedicated alert, “Phishing in inboxes due to bad allowlist”.

Some vendors also tell you to declare their IPs as an inbound gateway to remove Gmail banners. KnowBe4 itself notes that these steps “aren’t documented as a whitelisting recommendation by Google”. Google describes an inbound gateway as a server that processes your mail before Gmail, with your MX records pointed at it, and says that Gmail does not enforce the sender’s DMARC policy for messages that arrive from it (inbound gateway). A simulation tool is not that server.

What stays active: advanced protection and pre-delivery scanning

Advanced phishing and malware protection is set in Apps, Google Workspace, Gmail, Safety, per organizational unit. It still applies to approved senders. Its banners only go away with the two “hide warnings” options of the Spam section, and only show in Gmail on the web. If you turned on the option to apply future recommended settings automatically, a new check can be added without you choosing it: test again before each campaign.

Enhanced pre-delivery message scanning can also delay a message it finds suspicious (pre-delivery scanning). For a simulation, a few minutes’ delay does not skew the results.

Chrome: allow the landing pages (SafeBrowsingAllowlistDomains)

A simulation page imitates a sign-in page, so Chrome Safe Browsing may end up flagging it as dangerous. KnowBe4 and Hoxhunt document this in their help centers. The Chrome policy SafeBrowsingAllowlistDomains is meant for it: “Safe Browsing will trust the domains you designate. It won’t check them for dangerous resources such as phishing, malware, or unwanted software”, subdomains included (Chrome Enterprise policy list).

In the Admin console, open Menu, Devices, Chrome, Settings. On the User & browser settings page, Safe Browsing section, fill in Safe Browsing allowed domains with the tool’s link domain (for nophi.sh, app.nophi.sh). Google lists this setting for ChromeOS and Chrome browser on Windows, Mac and Linux. On Windows and macOS, the policy only applies to managed browsers: machines joined to an Active Directory or Entra ID domain, managed by MDM, or enrolled in Chrome Enterprise Core.

Check: Email Log Search and the Alert Center

Send a simulation to two or three mailboxes in the tested unit. Then, in the Admin console:

  • Reporting, Email Log Search: find the message by sender or message ID. The result shows whether it was delivered normally, marked as spam or quarantined, and the Matched rules field gives the name of the compliance rule that applied (Email Log Search results);
  • Alert Center: a campaign can trigger “User-reported phishing” or “Suspicious message reported” when employees report, “Phishing in inboxes due to bad allowlist” if you use the email allowlist, and “Phishing message detected post-delivery”, which automatically reclassifies and removes unopened messages (alert details). Google gives no guidance on simulations: warn the team that reads these alerts before the campaign.

Gmail alert notifications may arrive up to 4 hours after the alert rule is triggered. Keep the settings of a successful send as a reference in case Google changes its options.

With nophi.sh

The console does the checking for you: it recognises Google Workspace from your MX records, shows the IPs and sending domain to allow, and sends you a real test email through the same path as the simulations. If you find it in your inbox and confirm, delivery is proven. For reporting, the nophi.sh Gmail add-on puts a report button in Gmail, and each employee gets an AI verdict with its reasons, usually in one to two minutes.

The rest follows: 135 ready-made scenarios (69 in French, 66 in English), a short course assigned after a click, a PDF audit report and data hosted in France, for €1 per person per month excl. VAT, minimum 10 people, no commitment; details on the pricing page. If some of your mailboxes are on Microsoft 365, follow our Microsoft 365 Advanced Delivery guide. Moving off GoPhish? See migrate from GoPhish, and our SPF, DKIM and DMARC guide for the authentication Gmail checks.

Frequently asked questions

How do I allow a third-party phishing simulation in Google Workspace?

Google has no setting or help page dedicated to phishing simulations (checked on November 18, 2026). The method closest to its recommendations is a Spam rule, in Apps, Google Workspace, Gmail, Spam, Phishing and Malware, that bypasses spam filters for an address list containing the tool’s sending domain, with sender authentication on, applied to the organizational unit you test.

Is the Gmail email allowlist enough?

It stops mail from the listed IPs being marked as spam, but Google says email allowlists always apply to the entire domain, never to an organizational unit, and its Gmail settings health check advises not configuring them. They can also trigger the “Phishing in inboxes due to bad allowlist” alert. A Spam rule or a content compliance rule scoped to the tested unit is more precise.

Do I have to turn off sender authentication for the simulation to get through?

No, if the tool sends from an authenticated domain. Google recommends always leaving that option on: a message then matches the address list only if it passes DMARC (aligned SPF or DKIM). Some vendors tell you to uncheck it; Google advises against it, because it lets through unauthenticated messages spoofing the same domain.

Do Gmail warnings go away with the spam bypass?

Not with the basic option: the message lands in the inbox with the banner “This message wasn’t sent to Spam because of your organization’s settings.” With “Bypass spam filters and hide warnings”, warnings about possible suspicious messages aren’t displayed. Advanced phishing protections still apply in every case.

How do I stop Chrome blocking the simulation landing page?

With the Chrome policy SafeBrowsingAllowlistDomains, called Safe Browsing allowed domains in the Admin console (Devices, Chrome, Settings). Safe Browsing no longer checks the listed domains or their subdomains. It applies to ChromeOS and to Chrome on Windows, Mac and Linux managed by the organisation.

How do I check that a simulation email was delivered in Gmail?

In the Admin console, Reporting, Email Log Search: search for the message by sender or message ID. The result shows whether it was delivered normally, marked as spam or quarantined, and the Matched rules field gives the name of the compliance rule that applied.