"You're a cost center."
Every IT leader has heard this. Most have internalized it. Some have structured entire departments around it — optimizing for cost efficiency, measuring success by headcount ratios and ticket closure rates, reporting to CFOs who only engage with IT when the budget conversation comes around.
In 2026, that framing didn't go away. It just got harder to hide from.
The question changed
Two years ago, the question CFOs asked IT was: "Can you cut 10%?" Today the question is: "Show me the ROI on our AI investment." Same underlying framing — IT as a spend problem — but now with a deadline, a board slide attached to it, and a CEO who read a McKinsey report on the flight over.
The IT leaders who can answer that question with confidence are the ones who made a shift years ago. They stopped reporting inputs. They started reporting outcomes.
The IT leaders getting tapped for VP and CIO roles right now are not the ones with the cleanest uptime record. They're the ones who walk into a board meeting and speak the language of the business — with data that connects IT decisions to business outcomes.
Inputs vs. outcomes: what the difference looks like
This isn't abstract. Here's what the shift looks like in practice — using real examples from work I've led in Global IT and Enterprise Engineering:
Input framing: "Uptime was 99.98% last quarter."
That's an engineering metric. It tells your team something useful. It tells the CFO nothing actionable.
Outcome framing: "We deployed full remote work capability for 4,000+ employees across 12 countries in 10 days with no SLA breach during peak demand. When COVID hit, the business kept running. That was an IT decision made two years before the crisis."
Input framing: "We reduced Tier 1 ticket volume by 28%."
That's an operational number. It sounds like IT talking to itself.
Outcome framing: "Our AI assistant deflected 28% of support tickets in Year 1, our CSAT went up 12 points simultaneously, and the team shifted capacity to higher-complexity, higher-value work. That operating model is now running more efficiently than it did with a larger headcount."
Input framing: "We executed the divestiture on time."
So did every other IT team that didn't blow the deadline.
Outcome framing: "We separated a business unit's entire IT infrastructure — identity, network, endpoints, SaaS, compliance — in under 60 days across 30+ countries. That protected deal timelines, de-risked the transaction, and preserved business valuation during a critical window."
Why Enterprise Engineering belongs at the strategy table
The function called "IT" in most organizations is doing work that directly determines competitive capability — endpoint architecture shapes how fast the company can onboard new employees or integrate an acquisition. Network architecture determines what AI workloads are even possible. Identity infrastructure determines what data is accessible to whom and under what conditions.
These are not operational decisions. They are strategic ones. The Enterprise Engineering function — when led well — is the team that makes AI possible, makes M&A survivable, makes remote-first work sustainable at scale. That's not a cost center. That's a capability center.
The problem is that most IT organizations report the wrong things. They report effort, not impact. They report what they did, not what it enabled. And so, to the business, they look like a cost — because that's the only signal they're sending.
The AI ROI question is a reframe opportunity
Every board in the world is now asking their CIO some version of: "What is the ROI on our AI investment?" Most CIOs are scrambling to answer with a number. The ones who will differentiate themselves are the ones who answer with a framework.
What does AI ROI actually mean in enterprise IT? It means: ticket deflection that freed team capacity for higher-value work. It means network and identity architecture that made agentic AI workflows deployable without creating new security exposure. It means the monitoring and telemetry infrastructure that lets you measure AI impact in the first place.
You can't answer the AI ROI question with a deployment timeline. You answer it by showing that IT was building toward this capability before the board asked the question. That's what separates a technology leader from a technology manager.
"The work hasn't changed. The framing everything runs on has."
How to make the shift
This isn't a communications exercise. Changing your reporting language without changing your operating model is just better-worded input metrics. The real shift requires three things:
First, know what the business is actually trying to do. Not the IT roadmap — the business goals for the year. Revenue targets. Market expansion. M&A pipeline. Product launch timelines. Every IT initiative should map to one of those. If it doesn't, it probably shouldn't be on the roadmap.
Second, pre-agree on outcome metrics before starting significant initiatives. Before you deploy AI, before you execute a network modernization, before you run a Zero Trust rollout — sit down with a finance or business partner and agree on what success looks like in business terms. That conversation is harder upfront. It makes the reporting trivial afterwards.
Third, report outcomes at the cadence the business uses, not the cadence IT prefers. Quarterly business reviews, board presentations, executive all-hands — these are where the framing gets set. If IT isn't in the room, or is in the room but speaking operational language, the "cost center" label fills the vacuum.
The IT leaders getting called for VP and CIO roles right now made this shift already. They walk in with a perspective on the business — not a list of completed tickets. In 2026, that's the baseline expectation. Not the differentiator.