Total Cost of Ownership for Business Software: A Practical Guide With a Worked Example

Most software budgets are built on the price on the vendor's pricing page. That number is real, but it is rarely the number you end up paying. Total cost of ownership (TCO) is the discipline of counting everything: the subscription, the add ons, the hours your team spends making the tool work, the price rises at renewal and the cost of leaving one day. This guide gives you a cost model you can reuse, a worked example and the questions that expose hidden costs before you sign.

What total cost of ownership means

Total cost of ownership is the full cost of acquiring, operating and exiting a piece of software over a defined period. In procurement, the idea is older than software: a cheaper machine that breaks often or uses more energy can cost more than a dearer one over its life. Software works the same way. The purchase price is visible; the operating and exit costs are not, until you are already committed.

TCO is different from two related ideas:

• Price is what the vendor charges per user, per account or per unit of usage.
• Return on investment (ROI) compares the benefit a tool produces with its cost. You need an honest TCO before an ROI calculation means anything.

A good TCO estimate does not need to be precise to the dollar. Its job is to stop you comparing a tool's entry price with another tool's full cost, and to make the trade offs visible.

Why the subscription price is not the real cost

Three forces push the real cost of business software above the advertised price.

Prices rise at renewal. Vertice, which analyses software spend processed through its platform, reported SaaS price inflation of 13.6% in October 2025, 14.7% in November and 12.2% in December, averaging 13.5% for the quarter. Its H2 2026 report says June 2026 recorded the highest SaaS inflation rate on record. Whatever the exact rate when you read this, the lesson holds: a three year budget that assumes flat prices is optimistic.

Licences go unused. Zylo's 2026 SaaS Management Index found the average organization uses only 54% of the licences it provisions. Every unused seat is paid for and delivers nothing, which raises the effective cost per active user.

Costs arrive unplanned. Zylo also reports that for 61% of IT leaders, unplanned SaaS costs affected planned projects. Surprise costs usually come from usage based charges, forced upgrades when a plan limit is reached and renewals nobody reviewed.

The seven cost buckets

Use these seven buckets for every tool you evaluate. The table shows what belongs in each and where to find the number.

Two buckets are routinely left out: people time and exit. People time is easy to ignore because no invoice arrives, yet an administrator spending two hours a week on a tool is spending over a hundred hours a year. Exit is ignored because nobody buys software planning to leave, yet the cost of leaving is precisely what gives a vendor pricing power at renewal.

How to build a three year TCO model

Step 1: Fix the period and the scope

Choose three years unless the tool is a system of record that is painful to replace, such as accounting, payroll or an ERP, where five years is more realistic. Write down which teams and which processes the tool will cover, so every vendor is priced for the same scope.

Step 2: Price the plan you will really need

Read each plan's limits against your must have list. If you need an automation, a report or a user role that only exists on a higher plan, price that plan. Note whether the price is billed annually or monthly. HubSpot's CRM pricing page, for example, showed a standard Starter price of $20 per seat per month alongside lower promotional prices for new customers when we checked in November 2026, which is exactly the kind of gap that distorts comparisons if year two returns to the standard price.

Step 3: Model growth and renewal increases

Multiply seats by your realistic headcount for each year, not today's headcount. Then apply a renewal increase. If your contract caps increases, use the cap. If it does not, a prudent planning assumption is to model a double digit annual increase, in line with the SaaS inflation data above, and to treat anything lower as upside.

Step 4: Add one off costs

Estimate implementation and migration once, in year one. Include internal hours for cleaning and importing data, which is usually the slowest part of any rollout.

Step 5: Add recurring people time

Estimate hours per month for administration, user support and keeping integrations working. Multiply by a loaded hourly cost, which includes salary plus employer taxes and benefits.

Step 6: Add a realistic exit estimate

Assume you will leave at the end of the period and estimate the hours to export, migrate and retrain. You may never spend it, but including it changes which tool looks cheapest: tools with good export options and no lock in score better.

Step 7: Compare totals and cost per active user

Divide each tool's three year total by the number of people who will actually use it. Cost per active user is the fairest comparison between tools with different pricing models.

Worked example: a 15 person sales team choosing a CRM

The numbers below are an illustrative model built by the Businesslang team, not vendor quotes. They show how two tools with different entry prices can finish close together, or reverse, once every bucket is counted. Replace every input with your own.

Assumptions: 15 users in year one, growing to 18 in year two and 20 in year three. Loaded internal cost of $50 per hour. Renewal increase of 10% per year where no cap is agreed. Tool A has a lower list price but needs a paid automation add on and more admin time. Tool B costs more per seat but includes automation and has a 5% contractual cap on increases.

Method. Subscription for Tool A: 15 users × $25 × 12 in year one ($4,500), 18 users × $27.50 × 12 in year two ($5,940), 20 users × $30.25 × 12 in year three ($7,260), plus the same pattern for Tool B at $35, $36.75 and $38.59 per seat with the 5% cap ($6,300, $7,938 and $9,261, rounded to the dollar). Internal hours are multiplied by $50; recurring hours run for 36 months. The average user count is (15 + 18 + 20) ÷ 3.

The point is not that Tool B always wins. If Tool A did not need the add on and took the same admin time, it would be cheaper. The point is that the entry price, $25 versus $35, predicted the wrong answer in this scenario. A TCO model forces the right questions.

Hidden costs to ask about before you sign

• Plan limits. What happens when you exceed contacts, records, storage, automations or API calls? Is the next step an add on or a whole plan upgrade?
• Usage based charges. AI features increasingly use credits or tokens. Ask how usage is metered, where you can see it and whether you can set a hard cap.
• Minimum seats and minimum terms. Some plans require a minimum number of seats or an annual commitment even when monthly billing is advertised.
• Support tiers. Is phone or priority support included, or sold separately?
• Onboarding fees. Some vendors charge mandatory onboarding on higher tiers.
• Integration costs. Does the connector you need cost extra, or require a third party automation platform with its own subscription?
• Data export. Can you export everything, in a standard format, at no cost, at any time?

How to reduce TCO without buying worse software

1. Right size seats at every renewal. With licence utilization averaging 54% in Zylo's data, reclaiming idle seats is the fastest saving available.
2. Negotiate an increase cap. A written cap on renewal increases turns an unknown into a known.
3. Choose the plan, not the brand. Many buyers overpay by choosing the top tier "to be safe". Map must haves to the lowest plan that covers them.
4. Consolidate overlapping tools. Two tools doing the same job double the subscription, the admin and the training.
5. Invest in setup once. Clean data and good configuration in month one reduce admin time for years.
6. Assign an owner. A tool with a named owner gets reviewed before renewal; a tool without one renews by default.

A simple TCO template you can copy

Create one row per cost line and one column per year, then a total column. Keep assumptions in a separate box at the top: users per year, loaded hourly cost, renewal increase, period. When a vendor changes a price or you change an assumption, the totals update and the comparison stays honest. Pair this model with our seven step buying process and check the contract against our SaaS contract checklist before you commit.

TCO differences by software category

The weight of each cost bucket changes by category. Use this table to decide where to spend your estimating effort.

For systems of record such as accounting, payroll and HR, give the exit bucket more weight: they are the hardest to leave, which is exactly when vendors have the most pricing power.

Common mistakes in TCO estimates

• Using today's headcount for all three years.
• Comparing an annual price with a monthly price, or a promotional price with a standard one.
• Leaving out internal time because "we would be paying those people anyway". You would, but not for this.
• Treating the exit cost as zero.
• Forgetting that unused seats raise the cost of every active user.

A total cost of ownership estimate takes an hour or two to build. On a purchase that will run for three years, that hour is one of the best returns available to any buyer.