How to defend an IT budget in a cut

A cut is rarely a judgment on IT. It is usually arithmetic happening somewhere else in the business. How you respond decides two things: what actually gets cut, and whether the next round starts with your budget again.

Do not trim everything by the same percentage

It is the fastest response and the worst one. It slows every project slightly, protects nothing, and leaves nobody able to say what the organization gave up. It also signals that the original budget had that much slack in it, which invites the next cut.

Separate obligations from choices before the meeting

Three buckets, priced:

• Contractual and regulatory — signed agreements, licence commitments, anything with a compliance obligation attached. Not available, and saying so requires the contract in hand.
• Operationally required — support, security, the refresh that keeps things running. Deferrable, at a cost that should be stated.
• Discretionary — new projects, improvements, anything not yet committed.

Most of a first-pass cut lands in the third bucket. The work is showing that the first two are genuinely the first two.

Give options, not a protest

The useful format is short: three scenarios at different numbers, each with what gets delivered and what does not. A CFO facing a gap needs to choose, and a leader who brings choices is a different kind of participant than one who brings objections.

Each option should name the thing that stops. “We can absorb this by deferring the laptop refresh into next year” is a decision the business can take. “This will impact delivery” is not.

Say when a deferral is borrowing

Pushing a refresh, a renewal or an upgrade into next year moves cost rather than removing it, and often increases it. That is a legitimate choice — organizations make it every year — but it has to be presented as a loan with a repayment date, or the next budget cycle will treat the deferral as the new baseline.

Bring the things you were already stopping

The strongest position in a cut conversation is a list of applications, licences and contracts you had already decided to retire. It demonstrates the budget was being managed before the pressure arrived, and it converts part of the cut into work already in flight.

This is the argument for doing portfolio and licensing work when there is no crisis. It is what you reach for when there is one.

Protect the thing that is hardest to restart

Some costs can be stopped and resumed cheaply. Others cannot: a team that disbands, a migration halted partway, a security program stopped before the enforcement phase. Know which of your lines are in the second category and defend those specifically, with the restart cost attached.