The real benefits of SSO are fewer passwords, one place to enforce MFA, and one switch that blocks new logins everywhere at once. The risks are that the identity provider becomes a single point of failure, that disabling someone does not end the sessions they already have, and that a misconfigured setup can open more doors than it closes.
Those benefits are worth having. Whether they are worth paying for depends on a cost most benefit lists leave out: the plan upgrade every tool charges before it will accept SSO at all. Here is the case for and against, without the sales deck.
The benefits that hold up
Single sign-on (SSO) means your people log in once, to one account run by an identity provider (the service that holds your staff list and checks passwords, such as Google Workspace, Microsoft Entra ID or Okta), and every connected tool accepts that login. What SSO is covers the basics. Six gains survive contact with a real company.
Fewer passwords, so fewer bad ones. One strong password and a second factor replace dozens that people reuse, write down or share. This is the biggest security benefit, and it is real.
One place to set login rules. MFA (multi-factor authentication: a code, an app prompt or a security key on top of the password), password length, and sometimes which devices or countries can sign in. Set once, applied to every connected tool, instead of hoping each tool's own setting is switched on. SSO vs MFA explains how the two fit together.
One switch for future logins. Disable someone in the identity provider and they cannot start a new session in any tool connected through SSO. On a bad day, that matters.
One log of who signed in where. When something looks wrong, you can see logins across tools in one place rather than in fifteen admin panels.
Faster first days. With SSO in place, a new hire signs in to each connected tool with the account they already have. Some tools even create the account on first login, which is called just-in-time provisioning.
Customers stop asking. A security questionnaire that asks "Do you use SSO?" gets a yes.
The benefits that are oversold
Some claims on the slide do not hold up, or only hold up for much bigger companies.
"SSO handles offboarding." It does not. It blocks new logins. The leaver's account stays in each tool, holding a paid seat, with their files, API tokens and integrations still attached. Removing accounts is a separate job called deprovisioning, usually done through SCIM, a provisioning standard that tools often keep on an even pricier plan (what SCIM is).
"Everything is covered." Only tools that support SSO, on a plan you pay for, and that you have actually connected. Every other tool keeps its own passwords. In a small company that is often most of them, which is why a password manager stays on the shopping list (SSO vs a password manager).
"It saves IT hours." It saves password resets. It does not save the time spent inviting people, setting their roles, or removing them, which is where most small-company admin time goes.
"It makes you secure." It moves the risk to one place and lets you protect that place well. If that place is protected badly, it makes things worse.
Vendors also like productivity figures for time saved on logins. We have not seen one we would repeat. The honest version is that people type fewer passwords, and that is pleasant.
The risks: single point of failure, session persistence, misconfiguration
Ask "is SSO secure?" and the answer is: more secure than reused passwords, provided you handle three risks.
Single point of failure. The identity provider is now the key to every connected tool. If someone takes over an admin's identity provider account, they can get into everything that account can reach. That is why SSO without strong MFA on the identity provider is a bad trade, and why admin accounts deserve phishing-resistant factors such as security keys. The same applies to availability: if the identity provider has an outage, nobody can start a new session in any SSO tool until it is back. People already signed in usually carry on; new logins wait.
Session persistence. Disabling a person in the identity provider stops the next login, not the current one. A tool that was opened yesterday may keep that browser or phone app signed in until its own session expires, which each tool sets for itself. For a leaver who left on bad terms, that gap matters. Ending sessions means going to each tool, or having a tool that does it for you.
Misconfiguration. Most SSO failures are setup mistakes, not broken standards. A tool that matches people by email can link the wrong accounts if emails are reused or changed. Signing certificates (the keys that prove a login message came from your identity provider) expire, and when one does, nobody can log in to that tool until someone updates it; Google's guide to maintaining SAML certificates, read on 27 October 2026, gives its certificates a five-year lifetime, long enough for everyone to forget. Enforcing SSO before every admin has tested it can lock you out of your own tool. And leaving password login switched on "just in case" means SSO's rules do not actually apply to everyone.
None of these are reasons to avoid SSO. They are reasons to set it up deliberately, one tool at a time, which how to implement SSO walks through.
The cost
SSO has two bills, and the second is usually the bigger one.
The first is the identity provider. If you are on Google Workspace or Microsoft 365, you already have one. A dedicated one costs per user: Okta's Starter suite is $6 per user a month, billed annually, with a $1,500 yearly minimum, on Okta's pricing page as we read it on 27 October 2026.
The second is the tool upgrades. Most tools only accept SSO on their Business or Enterprise plans, and you upgrade every seat to get it. GitHub, for example, lists Team at $4 per user a month and keeps SAML SSO for Enterprise at $21 (GitHub pricing, same day). For 20 people that one tool adds $4,080 a year; GitHub Enterprise pricing looks at whether that jump is worth it. This is the SSO tax, and across a normal stack it runs into five figures.
So is SSO worth it for 10 people? Usually not across the board. The benefits are real but they are the same size for 10 people as for 1,000, while the price was set for the 1,000. The SSO tax calculator will give you your own number.
Who gets the most out of SSO
The benefits do not land evenly. They grow with three things: the number of tools that hold sensitive data, the number of people coming and going, and how often someone outside the company asks about it.
A ten-person agency on Google Workspace, with clients who never send security questionnaires, gets little from enforced SSO that Google sign-in and MFA do not already give it. A 50-person software company selling to banks is in a different place. Its customers are likely to ask for SAML by name, its staff turnover means more leavers to lock out, and its code and customer data sit in a few tools where a lingering login would hurt. That company should buy SSO for those few tools, and still not for everything.
Somewhere in between is most small businesses. For them the question is not "SSO or not" but "which three tools, if any". SSO for small business works through that decision.
Getting the benefits without SSO
Most of the gains above can be had more cheaply, if you are clear which one you want.
For fewer passwords, turn on "Sign in with Google" or Microsoft wherever a tool offers it on your current plan. Many do, for free. Put everything else in a password manager.
For central MFA, enforce 2-step verification in Google Workspace or Microsoft 365, and switch on MFA in each important tool. It takes an afternoon.
For one switch for leavers, accept that SSO does not really give you this anyway. What you need is a list of who has access to what, and a fast way to remove them from all of it; an employee offboarding checklist is the manual version.
For answering questionnaires, be honest about what you do. "Google sign-in where available, MFA enforced, documented offboarding" is a credible answer for many buyers. When a buyer insists on SSO for systems holding their data, buy it for those systems.
Where SmashSSO fits
SmashSSO is not SSO. It does not sit in front of anyone's login, so it gives you none of the login benefits above. What it covers is the part SSO is oversold for: adding and removing people.
It reads your people from Google Workspace with read-only access, and shows an access grid of who has an account in each tool and with what role, flagging accounts that match nobody in your directory. When someone is suspended in Google Workspace, it alerts you and shows every tool they are still in. Removals, invites and role changes run through a tool's member API where you have added a token, or through an agent in your own Chrome, in tabs where you are already signed in as an admin. That agent needs the browser open, is slower than an API, and can fail when a tool changes its screens; a failed run says so and shows its last screenshot.
It costs a flat fee from $199 a month billed annually, for the whole company (pricing). If a customer or auditor requires enforced login through an identity provider, you need real SSO for those systems; SmashSSO will not replace it.