What IBM's Earnings Miss Means for AI Marketing Budgets

The practical answer for the next budget meeting is this: IBM's earnings miss is not a clean reason to cancel AI marketing plans. It is a reason to revise the calendar, tighten the finance story, and stop treating every AI dollar as if it lands in the same line item.

That distinction matters because IBM's earnings miss is already being flattened into a broader anxiety story. IBM reported $17.2 billion in revenue, up 1% year over year, and earnings per share of $2.93 versus the $3.01 expected; the stock reaction was severe, with shares falling roughly 24% after the miss became public.[1] For a marketing leader who has already built internal support for an AI content, analytics, personalization, or workflow tool, the tempting question from finance is obvious: did IBM just prove enterprise AI demand is slowing?

The better question is narrower: which AI budget got pulled forward, and which purchase orders got pushed back?

The miss was about sequencing, not a vanished AI budget

The most useful line in IBM's July 14 investor letter was not the revenue figure. It was the explanation of the late-quarter budget motion: clients shifted capital spending from software to AI infrastructure in the final three weeks of June.[1] That is a very different signal from customers deciding AI is overfunded or underperforming.

A late-quarter shift tells you something specific about internal approvals. By that point in the quarter, many buyers are not rediscovering their entire strategy. They are deciding which already-approved commitments can move now, which ones can slip, and which ones are easier to defend to a CFO or CIO who is staring at capacity constraints. If servers, storage, GPUs, or data center commitments suddenly become the gating item for multiple AI initiatives, software can lose the timing fight without losing the strategic argument.

That is where marketing AI tools sit awkwardly. Most marketing teams buy into the software and services layer: campaign copilots, content systems, predictive scoring, experimentation platforms, customer intelligence, analytics, and workflow automation. Those tools may be central to the business case for AI-enabled growth, but they often depend on infrastructure decisions that marketing does not control. If IT is still standardizing data access, compute capacity, governance, model hosting, or security controls, the marketing tool request becomes easier to defer than the foundation beneath it.

This does not make the delay painless. It means the explanation to finance should change. The argument is not, "IBM missed, but our tool is different." The argument is, "IBM's miss points to an infrastructure-first buying sequence. Our marketing AI request remains valid, but the approval path needs to match the infrastructure calendar."

Infrastructure is absorbing the urgency

The broader spending context makes a simple cooling narrative hard to defend. Gartner forecasts worldwide AI spending will reach $2.59 trillion in 2026, up 47% year over year, and expects spending on AI-optimized servers to triple over five years.[2] That is not a marketing-spend forecast, and it should not be used as if it were. It is an economy-wide AI spending forecast, with infrastructure as a major beneficiary.

For the marketing budget conversation, the useful point is not that every AI vendor will share in that growth at the same time. The useful point is that enterprises are still committing capital to AI, but the first dollars may be going into the parts of the stack that make application-layer tools viable later.

The marketing-specific numbers still point upward. Marketing AI investment is estimated to grow from $35.7 billion in 2025 to $47.3 billion in 2026, according to IDC figures compiled by Amra & Elma.[3] That is a much smaller pool than Gartner's economy-wide AI figure, and it belongs in a different conversation. But it does matter for a marketing manager defending a tool request: the category is not being written off; it is being forced to wait its turn behind infrastructure in some enterprises.

The readiness gap explains why that happens. Only 30% of CMOs say they have the infrastructure needed to achieve their AI goals, according to Gartner data cited in secondary reporting.[3] That single figure makes the frustration legible. Marketing teams are being asked to produce AI-enabled growth while the company is still building the conditions that make those tools deployable, governable, and scalable.

A stock drop is a warning, not a budget model

The negative reaction to IBM should not be ignored. A roughly 24% drop is not a footnote, and some market coverage framed the earnings miss as evidence of pressure in IBM's AI-related growth story.[7] Marketing leaders do not need to argue with that reaction. They need to avoid importing it wholesale into a software-buying decision that has a different operating timeline.

Analyst disagreement is reasonable here because IBM's result can carry more than one message. It can show that investors punished a miss. It can show that software revenue timing matters. It can also show that customers are still spending on AI, just not evenly across the stack. The budget mistake would be to treat the market reaction as if it answered all three questions at once.

The New Stack's analysis of IBM's results also emphasized the infrastructure-first dynamic, including IBM leadership's acknowledgement that clients were prioritizing the hardware and platform capacity needed for AI workloads.[4] FullStack Labs and Hudson Labs reached a similar interpretation: enterprise AI spending is being reprioritized toward infrastructure rather than abandoned.[5][6] Those are still interpretations of a moving market, not guarantees about when any individual company's marketing tools will be approved.

That distinction is important. A delayed purchase order can look like lost demand from the vendor side. Inside the buyer's company, it may look more mundane: procurement is waiting for IT signoff, IT is waiting for platform capacity, finance is waiting for a clearer deployment sequence, and marketing is waiting for everyone else while still being measured against growth targets.

How to adjust the AI marketing budget ask

The operating posture should be disciplined, not defensive. Keep the AI marketing request active, but stop presenting it as if approval timing depends only on marketing urgency. In the current sequence, the finance-ready version of the plan should name the dependency: the tool budget produces value when infrastructure, data access, governance, and integration support are available.

A reasonable planning assumption is a 1-2 quarter delay for some marketing AI software decisions when infrastructure work has become the enterprise priority. That timing is an editorial synthesis from the IBM sequencing pattern and the supporting analyses, not a sourced industry forecast. It should be used as a scenario for planning, not as a promise to stakeholders.

• Keep the business case alive, but update the approval path to show which infrastructure or data dependencies must be cleared first.
• Separate budget language for infrastructure, AI software, services, and marketing applications so finance does not treat one IBM signal as a category-wide verdict.
• Ask IT for a readiness milestone, not a vague endorsement: data availability, security review, integration support, model governance, or platform capacity.
• Prioritize use cases with shorter proof paths while waiting, such as workflow automation, content operations, analytics cleanup, or campaign testing where ownership is clear.
• Prepare a delayed-start version of the budget so the request can survive a quarterly timing shift without being rewritten from scratch.

This is also the moment to narrow the ask. A broad "AI transformation" request is easy to postpone when infrastructure is taking the budget oxygen. A specific request tied to measurable campaign cycle time, lead quality, content throughput, customer segmentation, or analytics labor has a better chance of staying on the calendar. If a team needs a broader budget framework, the allocation logic in How to Allocate Your AI Marketing Budget in 2026 is the more relevant conversation than debating whether one enterprise software miss invalidates the category.

For teams already feeling pressure from AI budget anxiety, it is worth separating two finance conversations that often get collapsed. One is whether the company believes AI will matter to growth. The other is whether a specific marketing AI tool should be bought before the organization can support it properly. The first question may remain favorable while the second produces a delay. That is not hypocrisy; it is sequencing.

What to say when finance asks about IBM

The cleanest answer is short: IBM's miss suggests some enterprises are pulling AI infrastructure spending forward and deferring software purchases, not abandoning AI investment. For marketing, that means we should keep the tool plan active, tie it to infrastructure readiness, and build a 1-2 quarter timing cushion into the approval narrative.

Then bring the conversation back to the use case. If the request depends on data pipelines, identity resolution, warehouse access, model governance, or IT integration, acknowledge that dependency directly. If the request can operate inside existing systems with limited implementation burden, say so and show the shorter path to value. The same IBM headline should not produce the same budget response for every marketing AI tool.

The strongest near-term plans will probably be the least theatrical ones: pilots that use existing data, workflows that reduce manual work, analytics projects with clear ownership, and campaign applications where performance can be measured without waiting for a full enterprise AI platform. For prioritization, the more useful filter is not which vendor has the best AI story; it is where AI marketing ROI actually pays off under current infrastructure limits. That is the budget lens behind Where AI Marketing ROI Actually Pays Off.

So the answer is not to cancel the plan. It is to revise the calendar, update the finance narrative, and make the dependency chain visible. IBM's miss is a timing warning for AI marketing investments, not a verdict that the demand case has disappeared.

References

1. Arvind Krishna's Letter to IBM Investors, IBM Newsroom, July 14, 2026.
2. Gartner Forecasts Worldwide AI Spending to Grow 47 Percent in 2026, Gartner, May 19, 2026.
3. Marketing AI Spending Statistics, Amra & Elma.
4. IBM Earnings: AI Infrastructure, The New Stack.
5. What IBM's AI Results Say About Enterprise Spending, FullStack Labs.
6. AI Spending Shifts 2026, Hudson Labs.
7. IBM Stock Q2 2026 Results, Investor's Business Daily.