How to prove marketing ROI to a CFO for a multi-location business

To prove marketing ROI to a CFO, show how many extra customers your ads brought into your locations compared with similar locations where spend didn’t change, turn that into extra gross profit, and present it as a range with a payback period. Then put it on one page, including the markets that didn’t work. That is the whole job. Doing it well takes some care.

The stakes are real. Marketing budgets averaged 7.7% of company revenue in 2025, flat on the year before (Source: Gartner CMO Spend Survey, 2025). For a multi-location business that is often the largest discretionary line on the P&L, and a CFO looking for savings will look there first.

What a CFO means by marketing ROI

When marketing says ROI, it often means return on ad spend as reported by Meta or Google: attributed revenue divided by spend. When a CFO says ROI, they mean something narrower: the extra profit the business made because it spent the money, divided by what it spent. Those can differ by a factor of five, and the CFO knows it.

The gap comes from two places. First, platform-reported conversions include people who would have come anyway. Second, revenue isn’t profit. A gym member paying $45 a month doesn’t put $45 a month on the bottom line once you account for the variable cost of serving them.

Start with extra customers, not attributed conversions

Ad platforms count conversions they can link to their own ads. That includes the regular who clicked a brand-search ad on the way in and the member who was going to join regardless. Academic work comparing observational ad measurement against randomized experiments found the observational methods often got the answer wrong, usually overstating the effect (Source: Gordon, Zettelmeyer, Bhargava and Chapsky, Marketing Science, 2019; 15 experiments at Facebook). Your CFO doesn’t need to read the paper. They already assume it.

The number that survives scrutiny is incremental customers: customers above what would have come without the ads. For a multi-location business, the most practical way to estimate it is to compare markets where spend changed with markets where it held steady, weighted so the comparison tracked each market closely beforehand. Economists call this a synthetic control. It needs only weekly customer counts by location and spend by market, no pixels or customer IDs.

Show ranges, and decide on the lower bound

Advertising effects are hard to measure precisely. In one well-known study of 25 large field experiments, the median 95% confidence interval on return on investment was more than 100 percentage points wide (Source: Lewis and Rao, Quarterly Journal of Economics, 2015). That isn’t a reason to hide uncertainty. It is a reason to show it before the CFO asks.

A range does two things for you. It signals that you know the difference between an estimate and a fact, which most finance people will respect immediately. And it gives you a clean decision rule: if the low end of the range still pays back inside your target, the spend is defensible even on a bad day. If only the midpoint does, say so and propose a test to narrow it.

A worked example: one quarter of paid media at a 30-club gym brand

An illustrative 30-club fitness brand spends $180,000 on paid social and search in one quarter, across three tiers of markets. Platforms report 3,100 new-member sign-ups, implying $58 per member. The marketing lead instead compares each market with a weighted blend of the brand’s steady-spend markets.

Now turn members into money. Dues are $45 a month. After the variable cost of serving a member (cleaning, towels, payment fees, wear), about 70% is contribution: $31.50 a month. The brand’s own retention data says an average new member stays about 9 of their first 12 months. So each extra member is worth about $283.50 of first-year contribution.

Notice how the story changes. The platform number ($58 per member) would have suggested a payback under two months. The honest number says 4.8 months, and even the low end of the range pays back inside the first year. That is a better story, not a worse one, because it holds up when the CFO pokes it.

And Tier C? The read is inconclusive and the midpoint is expensive. The recommendation writes itself: move most of that $30,000 to Tier A and Tier B next quarter, and keep a small amount in two Tier C markets as a deliberate test.

Why payback lands better than ROI with finance

ROI depends on assumptions about how long customers stay, and CFOs know those assumptions can be stretched. Payback is harder to fudge. It asks one question: how many months of contribution does it take to earn back what you paid for the customer? Most finance teams already use payback for capital projects, so it is a language they speak.

Use your industry’s own noun and unit. For a restaurant it is extra guests and the gross profit per visit, with repeat visits as a bonus. For a dental group it is new patients and the first year of treatment value. For home services it is booked jobs and gross margin per job. The arithmetic is the same.

The one-page marketing ROI report for a CFO

1. One headline sentence. “Q3 paid media added about 1,200 new members (800 to 1,600) at $150 each, paying back in under five months.”
2. Four numbers, each with a range. Spend, extra customers, cost per extra customer, payback.
3. A breakdown by market or region. Including the ones that didn’t work. This is where you earn trust.
4. What you can’t tell yet. Inconclusive markets, channels without enough variation to read, anything else that is still open.
5. What you’ll do next. The specific budget moves, with dollar amounts, and when you’ll report back.
6. Method in two sentences. How the comparison works and how the range is built. A CFO who wants more will ask.

What CFOs push back on, and how to answer

• “Wouldn’t those customers have come anyway?” That is precisely what the comparison removes. Show the chart of the market against its comparison before and after the spend change. If they tracked closely before and separated after, the gap is the effect.
• “Meta says $58. Why are you saying $150?” Because Meta counts everyone who converted after seeing an ad, including people who were already coming. You are reporting the customers the ads actually added. CFOs tend to like the person who brings the less flattering number.
• “Your range is huge.” Agree, then point at the low end: “even at 800 members we pay back in about seven months.” If the low end doesn’t clear the bar, propose a longer or bigger test to narrow it, and size it in advance.
• “What happens if we cut 20%?” The best answer is one you have measured: a market where spend dropped. If you don’t have one, offer to cut 20% in a few matched markets for eight weeks and measure what you lose. That is a test a CFO will happily fund.
• “What about brand? You can’t measure everything.” True, and say so. Then be clear that the number you’re showing is a floor for short-run effects, not a claim about everything marketing does.
• “Who checked this?” Show the pre-period fit and the placebo results: how big a “lift” the method finds in markets where nothing changed. If those are small, your result is unlikely to be noise.

Common mistakes when presenting marketing ROI to finance

• Leading with ROAS. The first number sets the tone. If it’s a platform number, the rest of the meeting is spent discounting it.
• Reporting revenue as return. Revenue isn’t profit. Do the margin math before the CFO does.
• Hiding the losers. A report where every market wins reads as cherry-picked. Show the markets that didn’t work and what you’re doing about them.
• Comparing to last year. Year-on-year growth includes price rises, new clubs, the economy and the weather. It isn’t the effect of the ads.
• Claiming certainty. One overconfident number that later moves can cost you more credibility than a year of honest ranges.
• Inventing a lifetime value. Use first-year contribution unless you have real retention data. CFOs discount long tails anyway.

Countlift’s Proof Report PDF is built for exactly this meeting: extra customers, cost per extra customer and 95% ranges, on pages a finance team can read. And when the CFO asks a follow-up you didn’t prepare for, Ask Countlift answers from your own results rather than from a platform dashboard.

Questions people ask about proving marketing ROI to a CFO

Written by The Countlift team at First Byte in The Woodlands, Texas. Questions or corrections? Email hello@countlift.com.