Marketing ROI = (revenue attributable to marketing − marketing cost) ÷ marketing cost, expressed as a percentage. Spend $20,000, generate $70,000 in attributable revenue, and your ROI is 250%. The formula is trivial. The hard part — and where most agency dashboards quietly fail — is proving which revenue was actually attributable.
By Avi Vatsa — CEO, Exchange Four Agency Avi Vatsa is CEO of Exchange Four Agency, where he leads the team that installs and runs AI-leveraged revenue systems for owner-led companies. His background spans law, technology, and marketing; he also co-founded Dialora, an AI voice-agent platform for automated lead capture and booking. (Marketer of the Day #1411, Jeremy Ryan Slate Show)
Last updated: September 5, 2026
What is the marketing ROI formula?
The standard formula is (Revenue − Marketing Cost) ÷ Marketing Cost × 100. A stricter variant used by owner-operators who care about margin is (Gross Profit from marketing-attributed sales − Marketing Cost) ÷ Marketing Cost × 100. Use the second one. Revenue-based ROI flatters every campaign that sells low-margin work at volume.
Worked example, one quarter:
| Line | Amount |
|---|---|
| Marketing-attributed revenue | $70,000 |
| Cost of delivery (60% margin business) | $28,000 |
| Gross profit from that revenue | $42,000 |
| Ad spend | $12,000 |
| Agency retainer | $6,000 |
| Tooling (CRM, tracking, automation) | $2,000 |
| Total marketing cost | $20,000 |
Revenue ROI: ($70,000 − $20,000) ÷ $20,000 = 250%. Gross-profit ROI: ($42,000 − $20,000) ÷ $20,000 = 110%.
Same quarter. Same spend. One number says you tripled your money; the other says you roughly doubled it. Only the second one survives contact with your P&L. If the distinction between top-line and what you keep is fuzzy, start with net revenue: the number owner-operators should actually run on and revenue vs profit: which number should your marketing be held to.
What counts as "marketing cost"?
Marketing cost is every dollar that had to be spent for those sales to exist — not just the ad account. Owners routinely under-count by 30–50%, which inflates ROI into fiction. Count media spend, agency and freelancer fees, software and tracking tools, creative production, and the loaded cost of internal hours spent managing it all.
The line most often left off the sheet is coordination. Industry vendor-management research puts the hidden labor cost of juggling multiple specialized vendors at 8–15% of annual vendor spend — time that never appears on any invoice, and businesses running a stack of point-solution shops report spending roughly 30% more overall than those working with one integrated partner. If you run an SEO shop, a PPC shop, and a design shop, that coordination tax belongs in your denominator. We break the arithmetic down in what does using multiple marketing vendors actually cost.
How do you attribute revenue to marketing in the first place?
Attribution is the real work. You need a closed loop from first touch to closed-won money, held in one system. That means: every lead source captured at entry, every lead written to the CRM, every deal stamped with its source, and every won deal carrying a dollar value. Without that chain, your ROI number is an estimate wearing a suit.
Three attribution models an owner can actually use:
- First-touch. Credit the channel that created the lead. Best for judging top-of-funnel acquisition — which sources bring people who eventually buy.
- Last-touch. Credit the final interaction before purchase. Best for judging closing mechanics and retargeting.
- Self-reported ("How did you hear about us?"). Undervalued. For owner-led businesses selling through trust and referral, a single required field on the booking form frequently beats platform-reported data, because peer referral and word of mouth leave no click trail.
Pick one, apply it consistently for at least two full sales cycles, and never let a vendor switch models mid-report to make a quarter look better.
Why don't agency dashboards ever tie to a dollar?
Because most dashboards report activity, not money. Impressions, reach, sessions, click-through rate, and "engagement" are inputs. None of them appear on a bank statement. A dashboard that cannot be traced from a chart to a specific closed deal is reporting effort, and effort is not a result.
This is not a niche complaint. An industry survey distributed via Businesswire found that 71% of brands report frustration demonstrating marketing ROI effectiveness — a majority of the market cannot prove what it spent produced. That is a systems failure, not a math failure.
Three structural reasons it happens:
- The tracking was never installed. Platform pixels report platform-favorable conversions. Nobody connected them to the CRM, so "conversions" means form fills, not customers.
- The vendor is scoped to activity. If the contract buys deliverables — posts, ads, pages — the reporting will measure deliverables. You get what you scoped. This is the core argument in what changes when an agency owns the outcome instead of the scope.
- The follow-up leaks. Leads arrive and die in an inbox. Revenue that never happened cannot be attributed. Before spending more on acquisition, run an audit of your follow-up process.
"They know exactly how to connect marketing execution to real business outcomes." — Riggs Eckleberry, Chairman, OriginClear
That sentence is the whole standard. Execution connected to outcomes, or it doesn't count.
What we do when an owner can't calculate ROI
Here is the sequence we run, in order, when a company comes to us with spend and no answer. It is deliberately unglamorous.
1. Stand up one source of truth. Every lead, from every channel, lands in one CRM with a source field that is required, not optional. Until this exists, every other step is guesswork. The first-install decisions are covered in CRM automation for small business: what to install first, and what to skip.
2. Backfill the last 90 days manually. Yes, by hand. Pull closed-won deals, call the customers or check the notes, and assign a source to each. It takes a day or two and it is almost always more revealing than the dashboard was — we typically find that a channel the owner was about to cut is producing, and a channel with beautiful engagement metrics has produced nothing.
3. Add cost to the same sheet. Media, fees, tools, coordination hours. One tab, one period, no separate systems.
4. Compute both ROI numbers. Revenue ROI and gross-profit ROI, per channel. Rank them.
5. Re-run monthly, same method. The number only becomes useful when it's comparable across periods. A one-off calculation tells you where you were; a repeated one tells you which direction you're moving.
The output isn't a dashboard. It's a sentence an owner can say out loud: "We put in $20,000, we got back $42,000 in gross profit, and here are the two channels that made it."
How long before marketing ROI is a fair number to judge?
Judge it after at least one full sales cycle, plus a lag for the pipeline to fill. For a two-week close, 60–90 days is fair. For a six-month enterprise cycle, judging at 90 days will tell you nothing except that money left the account. Content and organic channels compound on a 6–12 month curve, not a 6–12 week one.
This is where owners most often abandon a working system too early — expecting immediate returns, seeing none in month two, and exiting before the machine has produced. Measure leading indicators early (leads created, source coverage, follow-up speed, pipeline value) and lagging indicators (closed revenue, ROI by channel) only from month four onward.
What ROI number should you actually aim for?
There is no universal benchmark worth quoting, and any agency that gives you one without seeing your margins is guessing. Your real threshold is set by your own economics: ROI must exceed the return you'd get from deploying the same dollar elsewhere in the business, after delivery cost. For most owner-led companies that means gross-profit ROI comfortably above 100% — every dollar in buys back more than a dollar.
That's the same test we apply to our own engagements. If it doesn't return more than it costs, it isn't finished.
Where to go next
If your reporting still can't answer the ROI question, the problem is upstream of the spreadsheet. Read why most marketing reporting doesn't prove anything for the reporting side, and the revenue formula, broken down for the mechanics underneath the number. If you're weighing who should own it, in-house marketing vs agency: the third option owner-operators miss and the diligence checklist for evaluating an agency before you sign are the two to read before your next contract.
A marketing ROI calculation is not a reporting exercise. It's the proof that a system exists. If the number can't be produced, there's nothing to measure yet — and that's the thing to fix first.
