AI marketing has gone mainstream: almost every marketing team uses AI now, but far fewer can say whether it’s paying off. Alexander Group’s 2026 research of more than 300 B2B companies found that 96% of marketers use AI, but only 41% can prove ROI. A Gartner survey of 2,002 CIOs and technology executives, conducted from May to July 2026, found that only 13% said their enterprise had realized significant or transformational value from AI. demandgenreportgartner
The gap between using AI and knowing what it returned is where money gets wasted. These seven questions will help you close it before you buy another tool, start another pilot, or renew another subscription.
1. What specific task is this supposed to improve?
“Use AI for marketing” isn’t a goal. “Draft the first version of our monthly email” is. Vague goals are the easiest to declare a success and the hardest to prove.
A good answer names one task, one owner, and one place in your marketing where it fits.
2. What does that task cost or produce today?
You can’t tell whether something improved if you don’t know where you started. Before you change anything, write down how long the task takes, what it costs, and how it performs.
A good answer includes at least one number from your own business: hours per task, cost per lead, response time, or open rate.
3. Which outcome will prove my AI marketing ROI?
Hours saved only count if they turn into something. The better measures of AI marketing ROI are things like a lower cost per lead, faster response to inquiries, more output at the same quality, or more revenue. Jasper’s 2026 survey of 1,400 marketers found that marketers are increasingly being asked to connect AI to pipeline, performance, and revenue, not just time saved, and that 60% of those who track AI ROI report at least a 2x return. That figure is self-reported and comes from a company that sells AI software, but the direction is useful: the teams that measure are the ones that know. jasper
A good answer names the one result that would make you keep the tool.
4. Who checks the output before customers see it?
AI is fast, but it can sound generic and state wrong information with confidence. Unedited copy tends to read like everyone else’s, which works against the reason customers choose you.
A good answer names a person, and a step in your process, that reviews anything customer-facing before it goes out.
5. Can I actually see the results?
If you can’t tell which calls, leads, and sales came from which source, you can’t tell whether AI changed anything. If you can’t tell which calls, leads, and sales came from which source, you can’t measure AI marketing ROI, because you can’t tell whether AI changed anything. It’s the same advice we gave in our post on keeping customers coming in when budgets are tight: see what the money did before you decide where to put more. That’s the kind of work our data and analytics team handles.
A good answer is a dashboard, report, or spreadsheet you can open and read without asking anyone.
6. When will I decide, and what result ends it?
This is the question most teams skip. Gartner defines “vanguards” as the CIOs reporting stronger business outcomes and high CEO-perceived AI value. It found that organizations with guidelines for deciding when an AI pilot shouldn’t move to production were 2.3 times more likely to be vanguards, and those with guidelines for abandoning low-value AI work already in production were 2.6 times more likely. Yet only 31% and 17% of CIOs, respectively, had those practices in place. Those are executives’ own assessments, not audited returns, and the survey covers enterprise technology rather than marketing. Even so, the habit transfers to any business: decide in advance what keeps a tool and what cuts it. Only 13% of CIOs See Major AI Value, Gartner Finds — Few Have Exit Rules +2
A good answer is a specific review date you pick before you start, and a result that decides whether you keep, adjust, or stop.
7. What does it really cost?
The subscription price is only part of it. Add setup time, the time spent editing and reviewing output, and any training for your team. A tool that speeds up the first draft but adds time in cleanup may not be saving you much.
A good answer is a total cost, in dollars and hours, that you can compare against the result from question 3.
How to use the checklist to check your AI marketing ROI
Run any AI tool or idea through all seven questions before you spend more. If there’s one you can’t answer, that’s the one to fix first. The tool will still be there next week.
A simple version you can copy:
- I can name the one task this improves.
- I know today’s cost or performance for that task.
- I’ve picked the outcome that proves AI marketing ROI.
- A person checks the output before customers see it.
- I can see the results in a report I trust.
- I have a review date and a stop rule.
- I know the total cost in dollars and hours.
Want a second set of eyes?
If you’re not sure whether your AI marketing ROI is real, we can help you find out. Our AI marketing work includes an AI opportunity audit across marketing, advertising, and sales, plus AI-assisted production and automation that’s checked by people who did the work by hand first. Book a 15-minute call and tell us what you’re trying, or start with the Deep Dive, a paid, hands-on look at your marketing, advertising, and sales that leaves you with a clear plan whether you hire us or not.
Frequently asked questions
How do I measure AI marketing ROI? Measuring AI marketing ROI starts with a baseline: record how the task performs today, then track an outcome that matters to the business, such as cost per lead, response time, or revenue.
How do I know if AI is worth it for my marketing? Pick one task, record how it performs today, and compare the result after you use AI. If cost, speed, or quality improved enough to cover the total cost, it’s worth keeping.
What’s a good first AI project for a small business? Choose something repetitive and easy to check, like first drafts of emails, ad copy variations, or summarizing customer feedback. A person should review anything customers will see.
How do I measure AI ROI in marketing? Track an outcome that matters to the business, such as cost per lead, response time, or revenue, not only hours saved. Compare it to the baseline you recorded before you started.
How long should I test an AI tool before deciding? Set a review date before you begin. The right length depends on the task and how long results take to show, such as a full email cycle or ad flight. Deciding the date in advance means you judge on evidence instead of habit.
Should I stop using AI if I can’t prove it’s working? Not necessarily, but you should stop spending more on it until you can. Set up the tracking, run a clear test, and let the results decide.
Related: AI marketing at Spectruss · Generative Engine Optimization for Small Business · Fractional CMO vs. Full-Time CMO
Sources
- Gartner Survey Finds Only 13% of CIOs Say Their Enterprise Has Realized Significant or Transformational Value from AI, Gartner, October 19, 2026.
- Alexander Group Research Finds Only 41% of Marketers Can Prove AI ROI, Demand Gen Report, October 6, 2026. The research was originally announced via Business Wire on September 16, 2026.
- New Research: The State of AI in Marketing 2026, Jasper, January 28, 2026.
