Budget season at most small and mid-sized businesses follows the same pattern. Someone opens this year's actuals, adds a few percent to each cost line and puts the result in front of management. The debate goes to headcount and marketing, and everything else gets waved through. Then in mid-January the software line is already over budget, because a licence agreement renewed on 1 January at a new rate nobody had factored in.
That is not an arithmetic error. It is a sequencing error. For a sizeable part of your costs, the budget is not written in November. It was written when someone signed the contract. Map that part first and the budget takes less time to build and holds up better once the year starts.
Why "last year plus three percent" breaks down
The method is not foolish. For plenty of cost lines it works fine. But for contract costs it misses four things that this year's actuals cannot show you:
- Contracts that end partway through next year, so a cost stops mid-year or gets renewed at a different rate.
- Contracts that only started partway through this year, so the actuals show six months of something that will cost a full twelve next year.
- Indexation and tiered pricing that kick in on a fixed date, regardless of what average inflation does.
- Minimum volumes you are committed to, even if you need less next year.
The catch is that hardly anyone has this overview. Weshare found that 95% of organisations lack full visibility of their contractual obligations (Weshare, 2025). Without it, last year's actuals are the only starting point available, even when everyone knows they are incomplete.
Three kinds of cost, not two
Budgets usually split costs into fixed and variable. For contracts, a different split works better, because it answers the question that actually matters in the budget meeting: what can we still change?
Committed. Costs you are locked into next year no matter what. A multi-year contract that runs until 2028, a lease, a vehicle fleet agreement, a purchase obligation where you pay for a minimum volume whether you use it or not. There is no choice to make here. There is only arithmetic to get right.
Decidable. Contracts that expire or become cancellable at some point next year. This is where the real room sits, but only until a given date. Once the notice period has passed, a contract drops back into the first column, and with automatic renewal it usually stays there for another full year.
Open. Spend with no contract behind it: one-off orders, ad hoc contractors, anything decided case by case. This is the column the budget debate tends to focus on, even though it is often the smallest.
Run the sort for a 25-person services firm. Office rent, the lease fleet and two software contracts running to 2028 add up to 310,000 euros: committed. Telephony, the accounting package, cleaning and the managed IT provider all expire or become cancellable in 2027, together 95,000 euros: decidable. Spend with no contract comes to 40,000 euros: open. Of 445,000 euros in external costs, roughly 70% is settled before the budget meeting starts. The discussion belongs with the 95,000 in the middle.
Five facts per contract
To sort your contracts this way you need five facts for each one. No more.
- The annual contract value, not the total over the full term. The annual figure is what lands in the budget.
- The end date, and whether the contract rolls over automatically after it.
- The last date on which you can give notice. With a three-month notice period and an end date of 1 April, that is 1 January, before the first monthly report of the new year is even out.
- Whether there is a price indexation clause, and if so from which date and against which index.
- Whether a minimum volume or tiered pricing applies.
If you keep a contract register, pulling this out takes an hour. If you do not, the budget is a good reason to start one: you have to open the contracts anyway to get the costs right, and what you write down now you will need again next year.
Give software extra attention. Gartner estimates that around 40% of SaaS spend goes unmonitored (Gartner, 2024), partly because departments sign up for subscriptions on a company card. Those costs show up in your actuals but not in any contract overview, and they renew per user or per month. This is also where ghost licences pile up: seats for people who have left, or a tool that was replaced in March and never cancelled.
Notice dates belong in the budget, not just the calendar
This is the part most businesses skip. Every contract in the decidable column should come with a decision in the budget, or at the very least the date by which that decision has to be made.
Take the managed IT provider at 28,000 euros a year. The contract runs until 1 April 2027 and then renews for a year. Notice is three months. If you want to change anything at renewal, notice must be given by 1 January, and to be in a position to do that you need a quote from an alternative in November or December. In practice, the November budget meeting is the last calm moment to make that call.
So next to every decidable contract, record two things in the budget: the amount you assume if you do nothing, and the date by which you have to decide. The budget line becomes a conditional figure instead of an assumption, and everyone at the table can see there is still something to act on. Our guide to cancelling a business contract covers how to give notice without missing a deadline, and the termination clause template shows the wording to aim for in the contract itself.
Budget indexation contract by contract
A flat uplift of three or four percent across all contract costs is too blunt for most businesses. One contract indexes on 1 January against October's CPI, another on 1 July against a wage index, and a third is capped at 2.5%. The gap between those three can run to several percentage points in a single year.
Go through the indexation clauses of your ten largest contracts and work out the increase and its effective date for each. For the smaller contracts, an average uplift is fine. The combination is more accurate than a blanket figure and takes barely any extra time. We worked through what a few percent of indexation does to a low-margin business in Price indexation in wholesale. The same mechanism applies in any sector where fixed contract costs are large relative to profit.
What it gets you
Bain & Company puts external procurement at 43% of total business costs on average, and much of that runs through contracts. A budget that does not start with the contracts is therefore leaning on estimates for close to half its costs, while the real figures sit in a drawer.
The biggest gain is not the accuracy of the budget itself, though. It is that the decidable column gets discussed while there is still time to act on it. A contract you review in November can still be cancelled or renegotiated in December. A contract you discover in February has already renewed itself for a year.
If you ran a summer contract cleanup, most of the work is already done: you have the list, the notice dates are in the calendar and the unused subscriptions are gone. The budget is simply the moment to turn that list into numbers. If not, start small. Take your twenty largest contracts, note the five facts above and sort them into three columns. It is an afternoon's work, and it makes the budget meeting a good deal shorter.