Skip to content

Blog

Why SaaS vendors put SSO on the Enterprise plan

Why SaaS vendors keep SSO on the Enterprise plan: how pricing fences sort big customers from small, what gets bundled to justify it, and how to push back.

· 9 min read

SaaS vendors put SSO on the Enterprise plan because it is the cleanest way to tell a large customer from a small one: big companies cannot buy without it, small ones mostly can, so it pushes exactly the customers with the biggest budgets onto the highest price. It is not priced on what it costs to run. It is priced on who needs it.

The SSO tax covers what this costs you. This post is about the mechanics: how SaaS pricing sorts customers, why SSO does that job better than almost any other feature, what gets added to make the price look reasonable, and what buyers can actually do about it.

Price segmentation in SaaS

Every SaaS company with more than one plan is doing price segmentation: charging different customers different prices for largely the same software. The aim is to charge each group close to what it is willing to pay, without losing the groups that would pay less.

The hard part is making customers sort themselves. A vendor cannot ask "how big is your budget?" and price accordingly. So it builds fences: conditions that make one plan suit one kind of buyer and not another. Airlines are the classic case. Business travelers and vacationers sit in the same seats, but for years a cheap fare required staying over a Saturday night, which vacationers did happily and business travelers would not. The rule had nothing to do with the cost of flying; it existed to separate two groups.

A good fence has three properties. First, the high-paying group cannot live without what is behind it. Second, the low-paying group does not much care about it, so putting it out of reach loses few small sales. Third, it is hard to get around: nobody can quietly enjoy the feature on the cheap plan.

SaaS plans use several fences at once: seat limits, storage, usage caps, admin controls, support levels. Most are leaky. Seat limits push growing teams up, but also push them to share logins. Storage caps irritate everyone. One fence beats all of them.

Why SSO is the perfect fence

Run SSO against the three tests.

Large customers cannot live without it. Once a company has a security team, a written policy usually says staff log in to work systems through the company's identity provider. Security questionnaires ask about it. Auditors look for it. IT departments will not manage hundreds of separate passwords per tool. For these buyers SSO is not a feature to weigh; it is a condition of purchase.

Small customers can live without it. A ten-person team can use "Sign in with Google", a password manager and MFA, and many do. Some want SSO, but they will not walk away over it. So a vendor can fence it off and lose very few small sales.

It is impossible to get around. SSO is on or off for the whole account. You cannot use it for half your people, and you cannot fake it in a security review. Compare a storage cap, which you can dodge by deleting old files.

There are three more reasons it suits vendors so well.

It tracks budget, not usage. The companies that need SSO are, almost by definition, the ones with procurement departments and larger budgets. The fence sorts by wallet without asking.

It upgrades every seat. Because SSO applies to the whole account, the whole account has to move to the higher plan. On per-seat pricing that multiplies the price gap by headcount (per-seat pricing shows how that compounds with hiring).

The buyer is often not the user. In a large company, the person who insists on SSO is in security, and the budget it comes from is not theirs. Price sensitivity is lower than it would be for a founder paying from the same account as payroll.

And what it costs the vendor

Vendors do not publish what SSO costs them, so nobody outside can give a number. What can be said: building SAML or OpenID Connect support is mostly a one-off engineering job, helped by mature open-source libraries. The ongoing cost is support. Every customer's identity provider is configured a little differently, certificates expire, email attributes do not match, and each of those becomes a ticket. When the US Cybersecurity and Infrastructure Security Agency (CISA) ran focus groups with small and medium businesses for a June 2024 piece on SSO adoption, even technically confident customers reported filing numerous support tickets to fill gaps in vendors' SSO instructions. The support cost is real, and partly self-inflicted.

None of those costs grow with the number of seats. A SAML connection for 12 people takes the same support effort as one for 1,200. Charging for it per seat is a pricing decision, not a cost recovery.

What else is bundled to justify it

A bare "SSO: +$17 per user" line would look exactly like what it is. So the SSO plan arrives as a bundle, and the bundle tends to contain three kinds of thing.

The first is things only large companies need: data residency, SIEM connections (feeding logs into a security monitoring system), granular admin roles across many workspaces, compliance reports. GitHub's Enterprise plan, at $21 per user a month against $4 for Team, lists data residency and SOC 1 and SOC 2 Type 2 reports alongside SAML (GitHub pricing). Whether GitHub Enterprise is worth the jump from Team weighs the rest of that list.

The second is things anyone would use: SCIM provisioning (automatic account creation and removal), audit logs, full data exports. Slack puts SAML, SCIM and exports of all messages together on Business+ (Slack pricing).

The third is service: faster support targets and named contacts. Slack includes its Premier support plan on Business+ and above, per the same page.

All as we read the pages on 27 October 2026.

The bundle does two jobs. It makes the price easier to sign off ("we also get audit logs and better support"), and it hides how much of the jump is for the login button. It also creates a second fence inside the first. Notion has SAML SSO on Business but keeps SCIM and the audit log for Enterprise (Notion pricing), as Notion's Business plan against Enterprise sets out. Buy the login and you still remove leavers by hand; automate removals and you are on the next plan up. What "enterprise SSO" means lays out what each plan includes.

The vendors that do not do this

Not every vendor fences SSO, and the exceptions show why the rest do.

Vendors whose product is itself about access tend to include it. Tailscale's whole model is that everyone signs in through an identity provider, and it lists SCIM on all paid plans (Tailscale pricing). Cloudflare offers dashboard SSO free on every plan, Free included (Cloudflare docs). PagerDuty includes SSO on Professional, its cheapest paid plan (PagerDuty pricing). The calculator keeps a longer list, and the SSO wall of shame ranks the vendors at the other end.

These vendors either sell to security-minded buyers who would judge them for charging extra, or price on something other than seats, so the fence would not earn them much. When SSO is not useful as a fence, it stops being treated as a premium.

Who gets caught by the fence

Fences are drawn around the customer the vendor wants to charge more. Everyone else who happens to need what is behind it pays the same price anyway.

For SSO, that is the small company with a reason. A 20-person agency whose biggest client sends a security questionnaire. A startup whose first enterprise deal has an identity clause. A team that just wants leavers to lose access on the day they go. None of them has a procurement department, and all of them pay the full enterprise surcharge on every seat. Across Slack, GitHub, Figma, Notion and Zoom, that is about $935 per person per year at list prices.

The vendor did not set out to charge them. It just has no way to let them through without letting the big customers through too. Whether a small business needs SSO at all is the question worth answering before you pay for someone else's fence.

How buyers push back

The SSO tax will not go away on its own, because for vendors it works. But buyers have more leverage than the price page suggests.

Ask for it, in writing, at the right moment. Renewals and annual commitments are when vendors bend. Ask for SAML on your current plan, or as a flat add-on, not as an all-seat upgrade. Some say yes; a polite no costs nothing.

Cite the public position. In that same June 2024 piece, CISA repeated its Secure by Design position: SSO should be available by default in the base offering, and customers should not have to pay an "onerous SSO tax" for it. It is guidance, not a rule, but it is a useful sentence to quote to a sales rep.

Make it a buying criterion. When you choose between two tools, check which one includes SSO, and tell the one that lost why. Pricing changes when enough lost deals give the same reason.

Buy SSO only where it is required. If a customer contract names three systems, pay for those three. Do not upgrade the rest out of tidiness.

Separate the login from the job. If what you need is to add and remove people across tools, you do not need the vendor's top plan to do it.

Will the SSO tax go away? Not soon. Government guidance and public lists of offenders add pressure, and the exceptions above show it can be done. For most vendors with per-seat pricing, SSO is still the best fence they have.

Where SmashSSO fits

SmashSSO is not SSO and does not get you past the fence for login. If a contract or audit requires people to log in through an identity provider, you need the vendor's SSO plan for those systems.

What it does is the job most small companies were trying to buy the top plan for: adding and removing people across every tool, on the plan you already have. It reads your people from Google Workspace, shows who has an account in each tool and with what role, and runs invites, role changes and removals by role, through a tool's member API where you have added a token or through an agent in your own Chrome where you are signed in as an admin. That agent needs the browser open and can fail when a tool changes its screens, and some tools' terms restrict automated use.

It is priced the opposite way to the fence: a flat fee from $199 a month billed annually, however many people and tools (pricing).

FAQ

Why is SSO so expensive?
Because it is the best way for vendors to sort big customers from small ones. Large companies cannot buy a tool without SSO, small ones mostly can, and it applies to the whole account, so putting it on the top plan moves every seat of every big customer to the top price.
Does SSO cost vendors much to run?
Vendors do not publish the figure. Building it is mostly a one-off job and the ongoing cost is support for customers' setups, but none of it grows with the number of seats, so per-seat pricing for SSO is a pricing choice rather than cost recovery.
Will the SSO tax go away?
Not soon. CISA has said SSO should be part of the base offering, and some vendors already include it, but the guidance is voluntary and SSO remains the most effective pricing fence most per-seat vendors have.
Can I negotiate SSO pricing?
Often, yes. Ask at renewal or when committing to an annual contract, and ask for SAML on your current plan or as a flat add-on rather than an upgrade of every seat. Make SSO a buying criterion and tell vendors when it cost them the deal.

Keep reading

Stop paying for a login button.

Sign in with Google Workspace, connect your tools, and see who has access to what in minutes.

Sign up with Google