Most marketing reports answer the question "were we busy?" Very few answer "did this make money?"
The second question is harder, but it's the only one that justifies a budget. Here's how to answer it.
The Basic Calculation
ROI=Marketing costRevenue from marketing−Marketing cost×100
Spend $10,000, generate $40,000 in attributable revenue, and your ROI is 300%.
Simple enough. The difficulty is in three places: defining "revenue from marketing," capturing the full cost, and choosing a time window that's honest.
Get the Cost Right
Marketing cost isn't just ad spend. Include:
- Media spend across all channels
- Agency or freelancer fees
- Salaries of internal marketing staff, or the fraction of time spent
- Software: analytics, automation, CRM, design tools, SEO platforms
- Content production: writing, design, photography, video
- Any one-off costs amortized over the period they benefit
Reporting a 400% return while excluding two salaries and a $2,000 monthly software stack is a fiction that will eventually be discovered.
Get the Revenue Right
Use gross profit, not top-line revenue.
If you sell a $10,000 project at 40% margin, the marketing contributed $4,000, not $10,000. Businesses with low margins routinely calculate impressive-looking ROI on campaigns that actually lost money.
For subscription or repeat-purchase businesses, decide whether to count first-purchase value or lifetime value — and then stay consistent. Mixing the two between reports makes trends meaningless.
The Metrics That Actually Matter
Customer Acquisition Cost (CAC)
Total sales and marketing spend divided by new customers acquired in the same period.
Segment it by channel. A blended CAC of $400 might conceal $150 from organic search and $1,100 from a paid channel you should probably pause.
Customer Lifetime Value (LTV)
Average order value × purchase frequency × retention period × gross margin.
LTV:CAC Ratio
The single most useful number in marketing.
- Below 1:1 — you lose money on every customer
- 1:1 to 3:1 — thin, possibly unsustainable
- 3:1 — generally healthy
- Above 5:1 — you're likely underspending and leaving growth on the table
That last point surprises people. A very high ratio often means you're being too conservative, not that you're brilliant.
CAC Payback Period
How many months of gross profit it takes to recover the acquisition cost. Under 12 months is comfortable for most businesses; over 18 creates cash-flow pressure regardless of how good the LTV looks.
Marketing Qualified Lead to Customer Rate
If this is very low, you have a lead quality problem, not a lead volume problem — and buying more of the same leads will make it worse.
Metrics to Deprioritize
Not useless, but frequently mistaken for results:
Impressions and reach. Measures how much you paid, not what you got.
Social followers. Correlates poorly with revenue for almost every business.
Sessions. Useful as a diagnostic, misleading as a goal. See the article on traffic rising while leads fall.
Email open rates. Increasingly unreliable since privacy features began pre-fetching images. Track clicks and replies instead.
Bounce rate in isolation. A high bounce rate on a page that answers a question completely is a success, not a failure.
Rankings without traffic. Position three for a term nobody searches is not an achievement.
Report these as context if you like. Don't report them as outcomes.
Attribution, Honestly
Attribution is imperfect. Anyone claiming otherwise is selling something.
Last-click gives all credit to the final touchpoint. Simple, and it systematically undervalues everything that created awareness. It'll make your brand-term ads look like heroes.
First-click does the reverse — overvalues discovery, ignores what closed the deal.
Linear and time-decay models spread credit across touchpoints. More realistic, harder to explain to a finance director.
Data-driven attribution uses your own conversion patterns to assign weights. Best available option if you have the volume to support it.
The practical approach for most businesses: use a multi-touch model as your default, and supplement it with two things that cost almost nothing.
Ask people. A single optional "How did you hear about us?" field on your form captures offline and word-of-mouth influence that no analytics platform will ever see.
Run holdout tests. Turn a channel off in one region or for one segment for a month. The change in total conversions tells you its real contribution far more honestly than any attribution model.
Building a Report Worth Reading
A useful monthly report has four sections:
1. Headline. Spend, revenue attributed, ROI, CAC, and change versus last month. Five numbers.
2. By channel. Same metrics broken down. This is where decisions get made.
3. What we learned. Tests run, what worked, what didn't, what surprised us. This is the most valuable section and the one most often missing.
4. What we're doing next. Specific actions, with expected effect.
If a report can't be understood in five minutes by someone outside marketing, it's a dashboard, not a report.
Set the Timeframe Honestly
Different channels pay back on different schedules. Paid search returns within days. SEO and content return over quarters. Brand work returns over years and resists measurement entirely.
Judging a content program on 30-day ROI will always produce a decision to cancel it — right before it would have compounded. Set the evaluation window at the start of the investment, and hold to it.
The Test
Bring your reporting to someone who doesn't work in marketing and ask them one question: based on this, should we spend more or less next month?
If they can answer, your measurement works. If they can't, no amount of additional charts will fix it.
Not confident your numbers are real? We'll audit your tracking and attribution setup and tell you exactly what's being miscounted.