Your AI Saved 400 Hours This Year. Where Did They Go?

Why "What's the ROI on AI?" is the wrong first question for marketing leaders, and what to ask instead

Let's try a little thought experiment.

Say your team's AI tools saved 400 hours this year. (I made that number up, but stay with me. For a lot of teams it's conservative.) First drafts in minutes instead of days. Research summaries that used to eat a Friday afternoon. Twelve versions of an email for twelve segments, done before lunch.

Now the awkward part: where did those 400 hours go?

Not "what did AI do." Where did the time go? For a lot of marketing teams, the honest answer is somewhere between "I'm not totally sure" and "into more of the same." More content. More campaigns. More meetings about the content and the campaigns.

And that might be why the AI return conversation feels so slippery right now.

Everyone's asking about AI's ROI

The pressure is real. Gartner's 2026 CMO Spend Survey found CMOs are putting an average of 15.3% of their marketing budgets into AI, yet only 30% say their organizations are ready to scale it. Meanwhile, 56% say they don't have the budget to deliver this year's strategy. So every AI dollar is getting a hard look, and rightly so.

The usual response is to hunt for the perfect AI metric. Cost per asset. Time to launch. Content volume. All useful. But after more than 20 years in B2B marketing, here's my slightly contrarian take:

ROI is the wrong first question. The better question is: What did we do with what AI gave back?

Because AI's first and most reliable return isn't revenue. It's time. And time is the one return you can waste without anyone noticing.

Where saved time goes to hide

There's an old idea called Parkinson's Law: work expands to fill the time available. AI didn't repeal it. If anything, AI put it on caffeine.

Here's how saved hours usually disappear:

• More volume. We can make ten assets now instead of three, so we do. Nobody asked whether buyers wanted seven more.

• More polish on things that didn't need it. The internal update gets its fourth revision because, hey, it only takes a minute.

• More meetings. Time freed up on the calendar rarely stays free. It gets claimed.

• Nowhere in particular. It just quietly dissolves into the week.

None of this is anyone's fault. It's what happens by default when nobody decides where the time should go. And it's exactly why so many teams feel busier than ever while their AI dashboards say they're saving hours.

Try a reinvestment ledger

So here's a simple idea. Instead of only tracking what AI saves, track where the savings get reinvested. Think of it as a ledger with three columns. Here is how you could approach it:

Hours saved	Where they went	What it produced
First drafts for the nurture program: ~40 hrs/quarter	Customer interviews	Two new proof stories with real numbers
Research summaries: ~25 hrs/quarter	Coaching sellers on the new narrative	Fewer "what do I say?" requests from sales
Versioning for segments: ~30 hrs/quarter	Executive point-of-view pieces	A monthly LinkedIn article from the CEO

That's it. No new platform required. A shared spreadsheet works fine.

The magic is in the middle column. It forces a decision that most teams never make out loud: what's worth more of our time now that the routine work takes less of it?

And the third column? That's your AI return story. Not "we produced 40% more content," which, let's be honest, makes a CFO nervous rather than excited. Instead: "AI took over the routine work, and we put that time into the things that move deals."

Where the time should go

If you're not sure what goes in the middle column, here's where I'd look first. These are the things AI can't do for you, and the things buyers value most:

1. Talking to customers. Real conversations, not surveys. This is where your best messaging, proof and product ideas come from, and it's usually the first thing cut when teams get busy.

2. Building proof. Customer outcomes with real numbers. AI can help you write the case study, but it can't land the interview or earn the customer's trust to share results.

3. Developing a point of view. When everyone has the same tools, the content that stands out is the content with a perspective someone could disagree with. That takes thinking time.

4. Developing your people. Coaching, feedback, and teaching the craft. If AI is doing the junior work, your team needs another way to learn the judgment that used to come with it.

5. Working with sales. Riding along on calls, sitting in on deal reviews, hearing objections first-hand. The fastest way to make marketing more relevant.

Notice what's not on the list: "make more stuff."

Enterprise vs. mid-market: same idea, different scale

For enterprise teams, the challenge is visibility. Hours are saved in dozens of places across regions and business units, and nobody sees the total. Start by asking each team lead to report two numbers each quarter: hours saved and where they went. You'll learn a lot just from the teams that can't answer.

For mid-market teams, the opportunity is focus. On a team of five, AI might give back the equivalent of a part-time person. That's a big deal. Decide together, as a team, where that "extra person" works, and protect it like a real headcount.

How to start on Monday

You don't need a transformation program for this. Three small moves:

1. Make your best guess at the hours. Ask each person on the team: "Roughly how many hours a week does AI save you?" Rough is fine. You're looking for direction, not precision.

2. Pick one or two reinvestment priorities. From the list above, choose where you want the time to go this quarter. Say it out loud to the team, so the time doesn't quietly drift back into volume.

3. Check the ledger once a quarter. What did the reinvested time produce? That's the story you bring to your CEO and CFO, and it's a far more convincing one than a content count.

   
   

The real return

Here's what I keep coming back to. The teams that win with AI won't be the ones that produce the most. They'll be the ones that get the most human work done with the time AI gives back: more customer insight, more proof, sharper points of view, stronger people.

So the next time someone asks, "What's the ROI on our AI investment?", try answering with a better question:

"Do you want to know what AI saved us, or what we built with it?"

The second answer is the one worth having.

Your turn: if your team got back an extra day a week, where would you want it to go?