The Headline Numbers
The most-cited IT spending forecast for 2026 puts worldwide spend at $6.31 trillion, a 13.5% increase on 2025 and the first time the figure has cleared $6 trillion comfortably. The growth is concentrated, not broad: data centre systems are forecast to grow 55.8% to surpass $788 billion, driven by AI infrastructure build-out, while generative AI model spending is projected to rise 80.8%. AI infrastructure software alone is set to jump from around $60 billion in 2025 to nearly $230 billion in 2026.
For a buyer, the important point is that headline growth is being driven by a handful of AI-related categories — not by your existing estate becoming more valuable. When a vendor cites “double-digit market growth” to justify your renewal increase, the growth they are quoting is largely happening in data centres and AI infrastructure you may not be buying. We unpack this distinction across the market intelligence pillar.
It is also worth noting how volatile these forecasts are, because the volatility is itself a buyer’s argument. The 2026 worldwide figure was revised repeatedly inside a single year — from 9.8% growth in an October projection, to 10.8% in February, to 13.5% by April — as analysts chased the AI build-out. A vendor that anchors your renewal to the highest available forecast is cherry-picking; the same source published materially lower numbers months earlier. When the underlying estimate can move three points in two quarters, it is not a stable basis for a multi-year price commitment.
One distinction anchors the whole picture: of the 15.1% software-spend growth, only a fraction reflects organisations choosing to buy more. The rest is the market absorbing higher prices for the same capability. Holding that difference in mind through every vendor conversation is the single most useful habit a budget-holder can build in 2026.
Where the money is not going
The flip side of concentrated growth is that large parts of the traditional estate are growing slowly or flat. Mature categories — on-premises databases, legacy middleware, established productivity suites — are not riding the AI wave, yet vendors routinely apply blanket increases across the whole portfolio as if every line were growing 15%. Identifying which of your spend categories sit in the slow-growth bucket is direct leverage: a vendor cannot credibly justify a double-digit uplift on a product line the market is barely expanding.
Software Is the Pressure Point
Software is forecast to reach $1.44 trillion in 2026 at 15.1% growth — the largest and fastest-growing enterprise IT segment. Critically, close to 9 percentage points of that increase is pure price inflation rather than new consumption: vendors raising prices and loading AI premiums into existing SKUs. That is the number that should frame every renewal conversation, and it connects directly to the enterprise software market trends reshaping contracts this year.
| 2026 segment forecast | Growth | Buyer implication |
|---|---|---|
| Worldwide IT spend ($6.31T) | +13.5% | Vendors will cite this to anchor increases |
| Software ($1.44T) | +15.1% | ~9pts is price inflation, not new value |
| Data centre systems ($788B+) | +55.8% | AI build-out, not your renewal’s justification |
| GenAI model spend | +80.8% | Negotiate AI as a separate, optional line |
| AI infrastructure software | ~$60B → ~$230B | Demand usage caps before committing |
“The market is growing 15%” is not a reason to accept a 15% increase. Most of that growth is AI infrastructure and price inflation — neither of which makes your existing licences more valuable.
What Forecasts Mean for Your Budget Cycle
Multi-year forecasts matter because enterprise contracts are multi-year commitments. If software inflation runs near 9% annually through 2026–2028, a flat three-year renewal with no uplift cap quietly compounds into a double-digit overpayment by the final year. The forecast is therefore an argument for tighter caps, not looser budgets. Pair it with the vendor’s own trajectory — our work on vendor revenue reports shows which suppliers are most reliant on price-led growth and therefore most likely to push uplifts.
The forecast also reframes infrastructure decisions. With data-centre and AI spend rising fastest, the cost gap between public cloud and owned infrastructure for steady-state workloads is widening — one driver behind the cloud repatriation trend. And because cloud providers are absorbing much of the AI build-out cost, the competitive dynamics captured in cloud market share directly affect the discounts on offer.
There is a budgeting trap hidden in the AI line specifically. AI infrastructure software is forecast to nearly quadruple, from around $60 billion to $230 billion, and much of that spend is consumption-based — it scales with usage you do not fully control. A budget built on a fixed software-inflation assumption will be blown apart by an uncapped AI consumption line, which is why the forecast argues for ring-fencing AI spend with hard ceilings rather than folding it into a general software-growth allowance. Treat the AI portion of your budget as variable and capped, and the rest as the inflation-managed base.
Turning Forecast into Leverage
The disciplined response is to use the forecast against the vendor. When an account team anchors on market growth, separate the components: confirm how much of your proposed increase reflects genuine new consumption versus price inflation, and cap the latter. Demand written uplift ceilings for the full contract term — ideally below the ~9% software inflation rate — and isolate AI line items so they remain optional rather than baked into the base. The Microsoft repricing wave detailed in our Microsoft 2026 strategy analysis is the clearest near-term example to plan around.
Benchmarking is what converts a forecast into a number you can defend. A market growing 13.5% overall tells you little about your specific category; transaction data showing what comparable enterprises actually pay tells you everything. The two together are powerful — the macro forecast frames the conversation, and the transaction benchmark sets your target. Walking into a renewal with both is the difference between debating the vendor’s list price and anchoring on the market’s real price.
The multi-year view also shapes timing. If software inflation is expected to ease after the current AI-driven peak, a shorter renewal preserves the option to renegotiate into a softer market; if it is expected to accelerate, locking a longer term with a hard cap makes sense. Either way, the forecast should drive the term length deliberately, rather than defaulting to the vendor’s preferred three years. Reading the direction of travel — not just the headline number — is where the forecast earns its place in your strategy.
Procurement teams should also translate the forecast into a defensible internal budget narrative. When finance asks why the software line is rising, “the market is up 15%” invites a blanket increase; “roughly nine points of that is price inflation we intend to cap, and the rest is optional AI consumption we will ring-fence” invites a strategy. Framing the forecast this way inside your own organisation is what gives the negotiating team a mandate to hold the line rather than absorb whatever the vendor proposes.
Forecasts are most useful as benchmarking context. Knowing the market is growing 13.5% while your specific category is mature tells you the vendor’s growth has to come from price, not volume — which is leverage. To pressure-test your exposure against current transaction data, start with our price benchmarking report or request a confidential briefing on your forward contract commitments.