AI Is Making Marketing More Productive. CFOs Are Going to Ask Where the Savings Went

Marketing teams have rarely had access to so much productive capacity so quickly.

AI can research markets, summarize customer information, generate creative variations, analyze campaigns, produce first drafts, personalize content and help teams complete work that previously consumed hours.

For CMOs, that is exciting.

For CFOs, it leads to another question:

Where is the economic benefit?

If AI allows a marketing team to accomplish substantially more with the same people, software and budget, eventually that leverage should appear somewhere.

It might show up as higher revenue.

It could appear as lower agency spending, fewer planned hires, faster campaign launches or more experiments completed by the same team.

But “we are using AI” is not a financial result.

That distinction is becoming one of the most important questions in the relationship between the CMO and CFO.

Research from Open Future Forum CMO FORUM, a Silicon Valley-founded executive community that publishes research across marketing, finance, security and other executive functions, has been tracking how AI is changing both sides of this conversation.

Its CMO research looks at adoption and leverage inside marketing. Its CFO research looks at investment, productivity and financial return.

Together, they point toward a question companies will increasingly need to answer:

If AI is making marketing more productive, how is the company capturing that productivity?

Marketing AI Has Moved Beyond Content Generation

The first wave of generative AI in marketing was easy to see.

People generated copy.

They created social posts, email subject lines, advertising variations and first drafts of articles.

Those applications remain useful, but they represent only one part of the opportunity.

Marketing teams are increasingly applying AI to research, analytics, segmentation, campaign operations, personalization, competitive intelligence, sales enablement and customer insights.

The important change is not that AI can produce content.

It is that AI can reduce the amount of human time required across multiple workflows.

That creates capacity.

What the organization does with that capacity determines whether AI becomes economically meaningful.

Saving Time Is Not the Same as Saving Money

This is where marketing teams can overstate AI ROI.

Imagine a process that previously required ten hours of work.

With AI, it takes five.

It is reasonable to say productivity improved.

It is not necessarily reasonable to say the company saved five hours of salary.

The employee is still employed.

The company still pays the salary.

The economic question is what happened to those five hours.

If nothing changed, the company has created available capacity but may not have captured much financial value.

If those hours allow the employee to run another campaign, perform more customer research, replace outside spending or support additional revenue, the value becomes more tangible.

That gives CMOs a useful framework:

Time saved → capacity created → capacity redeployed → business result.

The first step is easy to measure.

The last two determine whether the CFO will consider the project successful.

The CMO and CFO May See AI Differently

The CMO and CFO naturally approach AI from different positions.

Marketing leaders can see changes happening inside workflows.

They know a research process is faster.

They can see that creative teams produce more variations.

They know employees can analyze information that previously required outside support.

The CFO sees the company’s financial statements.

If headcount, software costs, agency spending and revenue all remain essentially unchanged, the CFO can reasonably ask where the value went.

Neither perspective is wrong.

They are measuring different stages of the same process.

The opportunity is to connect them.

Headcount Avoided May Matter More Than Headcount Cut

Discussions about AI productivity often jump immediately to job reductions.

That misses an important part of the economics.

Consider a growing marketing department that expects workload to increase substantially next year.

Without AI, the CMO might expect to add six people.

After redesigning several workflows, the organization discovers it can achieve its objectives with three additional hires.

Nobody was replaced.

But three future hires were avoided.

That is economically meaningful.

For growth companies in particular, headcount avoided may become one of the most important measures of AI leverage.

The same principle applies to contractors.

If a company can handle increasing workloads without expanding external resources at the same rate, AI can affect costs even while the organization continues growing.

Agency Spending Is Another Place to Look

AI could also change the relationship between companies and marketing agencies.

Organizations traditionally outsource work because they lack expertise, capacity or specialist resources.

AI changes the capacity side of that equation.

An internal marketing team that can research faster, produce more variations and analyze campaigns more efficiently may no longer need the same level of outside production support.

That does not mean agencies disappear.

It may change what clients value.

Original creative thinking, specialist expertise, strategy, relationships and complex execution can become more important as routine production gets cheaper.

For the CMO and CFO, the question is measurable:

Has AI actually changed external spending?

If a company says AI replaced $300,000 of agency work but still spends exactly the same amount with agencies, that $300,000 is theoretical.

If the expenditure actually disappears or is avoided as the company grows, it becomes a much stronger ROI claim.

More Content Is Not Necessarily More Value

Marketing provides another measurement trap.

AI can dramatically increase content production.

A team might produce five times as many articles, social posts, advertisements or creative variations.

That sounds impressive.

But content volume is an output metric.

It is not a business outcome.

If additional content does not increase qualified traffic, conversion, pipeline, customer engagement or another meaningful objective, producing more of it may simply create more noise.

In fact, AI may make generic content less valuable precisely because everyone can produce more of it.

That could increase the value of things that remain scarce:

original research

customer insight

expertise

brand

distribution

community

distinctive creative ideas

trusted third-party validation

The CMO’s job is not to maximize AI-generated output.

It is to determine where cheaper production creates actual competitive advantage.

Revenue Attribution Needs Discipline

Revenue is another area where AI ROI can become exaggerated.

Suppose AI helps produce creative for a campaign that generates $2 million in sales.

Did AI generate $2 million?

Probably not.

The campaign also depended on the product, offer, brand, audience, media spend and other factors.

A more defensible approach is incremental measurement.

Did AI-assisted creative outperform a comparable control?

Did personalization improve conversion?

Did AI allow the team to test enough variations to identify a winning campaign sooner?

Did better lead qualification improve sales conversion?

Did faster customer research improve a launch?

The objective should be to isolate the change created by AI rather than claim credit for everything AI touched.

That produces numbers a CFO is more likely to trust.

AI May Change the Shape of the Marketing Department

The economic effect of AI may ultimately go beyond individual productivity.

It can change organizational design.

Marketing jobs consist of many tasks.

AI may automate some, accelerate others and have little impact on the rest.

If those changes accumulate across a department, the composition of the team can change.

Some production-heavy roles may require fewer resources.

Marketing operations may become more important.

Employees with strong customer understanding, judgment, data literacy and the ability to design experiments may become more valuable.

Teams may become smaller relative to the amount of work they produce.

The key question is not:

How many marketing jobs can AI replace?

It is:

What should the marketing organization look like when routine knowledge work becomes substantially cheaper?

That is a much more useful management question.

AI Is Also Changing What CMOs Buy

As AI becomes part of marketing operations, CMOs face another problem: software proliferation.

A marketing organization can accumulate AI tools very quickly.

One for writing.

Another for research.

Another for video.

Another for analytics.

Another for personalization.

Another for sales enablement.

The individual costs may look small.

Collectively, they can become substantial.

CMOs therefore need to distinguish between tools employees enjoy using and systems that materially improve an important workflow.

The test should not be whether the software produces an impressive demonstration.

It should be whether the workflow improves enough to justify the cost.

The Best AI Metric May Be a Workflow Metric

This suggests a practical way for CMOs to measure AI.

Instead of starting with an abstract company-wide “AI ROI” number, measure individual workflows.

For example:

How long does customer research take?

What does a campaign cost to produce?

How quickly can creative be tested?

How much external spending does the process require?

How many campaigns can the team run each quarter?

How many people are required?

What conversion does the workflow produce?

Measure the baseline.

Introduce AI.

Measure again.

This is less exciting than announcing an enormous theoretical AI ROI figure.

It is also much more useful.

CMOs Need to Speak the CFO’s Language

The difference often comes down to communication.

A CMO might say:

Our marketing team is producing 40% more content using AI.

The CFO may hear:

Marketing is producing more things.

A stronger statement is:

We increased campaign capacity without making two planned hires and reduced external production expenditure.

Likewise:

AI saves our team hundreds of hours.

is less useful than:

Customer research now takes half as long, which allows the same team to support twice as many product launches.

The second version connects the capability to the business.

That is the language required to move AI from experimentation into operating budgets.

Cross-Functional Research Is Becoming More Important

One reason the CMO-CFO relationship matters is that no executive sees the entire AI transformation from one seat.

Open Future Forum has built its research program around that idea, examining AI from the perspective of CEOs, CFOs, CMOs, CISOs, General Counsel and AI leaders.

That cross-functional approach has attracted support from executives participating in the community.

Silvio Sangineto, a Microsoft executive and author of The AI Leader Mindset, has described Open Future Forum as helping define the enterprise AI agenda in Silicon Valley by bringing executives, investors and AI leaders into the same conversation.

The value of that approach becomes particularly clear with AI economics.

A marketing deployment can look successful to the CMO because output increased.

The CFO may question whether the additional output changed the economics.

The CISO may see new data exposure.

General Counsel may see contractual or governance issues.

The CEO has to decide whether the overall result creates enterprise value.

Those perspectives need to connect.

What Should CMOs Measure?

A practical marketing AI scorecard does not need dozens of metrics.

It should answer a few important questions.

Investment: What are we spending?

Workflow: What important process changed?

Productivity: How much time or capacity was created?

Cost: What spending was removed or avoided?

Growth: What incremental commercial result can we demonstrate?

Organization: Did the workflow change hiring or agency requirements?

Risk: Did the deployment create material customer, data, brand or governance issues?

That provides a much more complete view than counting AI licenses.

What Should CFOs Ask CMOs About AI?

CFOs do not need to become marketing technologists.

A handful of questions reveal a great deal.

Which marketing workflows have changed most because of AI?

What measurable productivity has been created?

Where has that productivity changed spending, hiring or revenue?

Which AI investments are not producing measurable results?

What would happen if we doubled the investment?

What would happen if we stopped it?

Those questions force the organization to distinguish enthusiasm from economics.

AI ROI Is Ultimately a Management Question

AI can make marketing dramatically more productive.

But productivity does not automatically appear on an income statement.

Management has to decide how to capture it.

Sometimes that will mean reducing costs.

Sometimes it will mean avoiding future hiring.

Sometimes it will mean producing more with the same team.

Sometimes it will mean shifting people toward higher-value work.

And sometimes the correct conclusion will be that an AI tool is impressive but does not create enough value to justify its cost.

That is healthy.

The objective is not to prove that every AI investment works.

It is to identify the ones that do.

For CMOs, the next stage of AI adoption will therefore require more than knowing what the technology can produce.

They will need to show what changed in the business.

And increasingly, the CFO will be asking the same question:

Where did the leverage go?

---

About the Research

This article draws on Open Future Forum’s executive AI research, including its CMO AI Leverage Report and CFO AI Leverage Report.

Open Future Forum is a Silicon Valley-founded executive community bringing together CEOs, CFOs, CMOs, CISOs, General Counsel, investors, founders and AI leaders. Its research examines enterprise AI adoption, investment, productivity and organizational change from different executive perspectives.

Research findings should be interpreted within the methodology and sample described in the relevant underlying reports.