
The Privileged User Audit Most Orgs Aren’t Going to Like
By Justin Seaman, IT Manager, and Jory Cohen, Development Director
Salesforce starts enforcing phish-resistant MFA in sandboxes on June 22 and in production on July 1. Every privileged user — System Administrator profile, or anyone with Modify All Data, View All Data, Customize Application, or Author Apex — has to register a security key, Touch ID, Windows Hello, or Face ID before that date. Authenticator apps and SMS don’t satisfy the requirement. There is no on-platform bypass. If admins don’t take the appropriate steps, locked-out users will have to open a Salesforce Support case and wait.
That’s the visible problem, but not the most impactful one.
The real impact comes from the audit that uncovers which users have access to what permissions. Last week we ran a Health Check for a client. It came back green. Then we reviewed their permission set assignments end-to-end. We found three Modify All Data grants nobody on the admin team could explain. An integration user no one had touched in a year. A consultant who’d been gone eight months, still active and with a permission set granting Modify All Data privilege.
The pattern is consistent across every org we’ve scanned: admins are under-counting their privileged users by three to five times. Most of the gaps are invisible when you only look at the profiles people are assigned. They hide in permission sets and permission set groups, in integration accounts that were stood up during implementation and never cleaned up, in dev and QA logins that get sandbox-only privileges in theory and production privileges in practice.
A common pushback: “We use SSO, so this doesn’t apply to us.” Here’s the catch. Salesforce reads the authentication signals coming from your SSO — Microsoft Entra ID, Okta, Ping, and so on. If your Entra or Okta Conditional Access policy doesn’t enforce a phish-resistant authentication strength for your Salesforce-bound app, those privileged users get challenged again at the Salesforce layer — and blocked if they can’t satisfy it. SSO doesn’t exempt you from the requirement. It just changes where you have to configure it.
The fix itself isn’t complicated. Two checkboxes in Salesforce Settings can prevent privileged users from getting locked out, and for SSO orgs, a Conditional Access policy that asserts phish-resistance will make the login process seamless. What takes time is the audit before that. Building the queries to identify who needs to enroll. Identifying integration users who need a bypass, which requires submitting justification in a Salesforce case. The decision on whether the remaining consultants actually need phish-resistant MFA or should just have their permissions trimmed.
Don’t wait to have these conversations and decisions until it’s too late and you are locked out of the org along with the rest of your admins. Use the deadline as an opportunity to audit and strengthen your security before it forces the issue.
The enforcement is the forcing function, and ForeFront’s MFA Readiness Scan is the value. The teams that come out of this in good shape will be the ones that used the deadline as a reason to read their permission set assignments end-to-end — probably for the first time since their original implementation. Don’t get bogged down in manual SOQL queries or building one-off reports before the deadline. ForeFront’s MFA Readiness Scan skips straight to the actionable insights your admins need to keep your company secure.
And this is just the tip of the spear.
At TDX in April, Salesforce announced Headless 360: the entire platform becomes API-accessible, with 60+ Model Context Protocol (MCP) tools and 30 preconfigured coding skills that expose your data, business logic, and workflows to agents — AI and otherwise — without a browser in the loop. The pitch is developer productivity. The implication is identity governance.
When an agent acts on a user’s behalf, it inherits that user’s permissions. A privileged user with weak MFA is also a privileged agent. Sloppy permission set assignments aren’t just a login risk anymore — they define what an AI agent is allowed to read, write, and delete on the platform.
Headless 360 raises the stakes of every audit you haven’t run. ForeFront is here to help make sure you’re ready for today and tomorrow. The same permission set group that surprised you in the MFA scan is what an agent uses when it answers a Slack message, executes a workflow, or writes to an Opportunity. Identity gates everything. That’s why phish-resistant MFA for privileged users is the first thing Salesforce is enforcing — before the agent platform fully lands.
The orgs that treat MFA enforcement as a compliance checkbox are going to walk into Headless 360 with the same mess, scaled by every agent they deploy. The orgs that use the deadline to clean up identity, permissions, and governance now are the ones who’ll be able to use the agentic platform safely when it lands.
- Assess your current MFA posture and identify gaps against Salesforce’s new requirements
- Design and implement phishing-resistant authentication (FIDO2, passkeys, and more) tailored to your environment
- Minimize user disruption with practical rollout strategies and change management support
- Validate readiness in sandbox environments before enforcement hits production
- Ensure long-term resilience with governance, monitoring, and best practices
Take action now. Avoid business disruption, user friction and security exposure and Partner with ForeFront to secure your Salesforce environment before enforcement begins.
Want to take action? Contact us to learn more!

