The Marketing ROI Equation Most Teams Get Wrong

The marketing ROI equation is straightforward on paper: subtract your marketing costs from the revenue those efforts generated, divide by the cost, and express the result as a percentage. What makes it complicated in practice is almost everything else, including which costs you include, which revenue you attribute, over what time period you measure, and what you do when the numbers don’t tell a clean story.

Most teams either oversimplify the calculation to make it look better than it is, or overcomplicate it to the point where it becomes unusable. Neither approach helps you make better decisions. This article is about finding the middle ground.

What Is the Marketing ROI Equation?

At its core, the marketing ROI equation looks like this:

Marketing ROI = (Revenue Attributed to Marketing, minus Marketing Costs) / Marketing Costs x 100

So if you spent £50,000 on a campaign and it generated £200,000 in revenue, your ROI is 300%. For every pound invested, you got four back, three of which is profit on the marketing investment.

That part is not the problem. The problem is that both inputs, revenue and cost, are far more contested than they appear. Teams routinely inflate the revenue side and deflate the cost side, producing ROI figures that look impressive but don’t survive commercial scrutiny.

I’ve sat in enough board presentations to know that the ROI number on slide three is rarely the number the finance director would recognise. Not because marketers are dishonest, but because they’re often measuring what’s easy to measure rather than what’s accurate. If you want a broader view of how measurement fits into marketing decision-making, the Marketing Analytics hub covers the full landscape.

Why the Revenue Side Is Harder Than It Looks

Revenue attribution is where most ROI calculations quietly fall apart. The question isn’t just “how much revenue did we make?” It’s “how much of that revenue can we reasonably credit to marketing, and to which specific marketing activity?”

Consider a customer who saw a display ad in January, clicked a paid search ad in March, opened an email in April, and then converted after a colleague mentioned the brand in a meeting. Which touchpoint gets the credit? Last-click attribution gives it all to the email. First-click gives it to the display ad. Linear attribution splits it across all three tracked touchpoints and ignores the word-of-mouth entirely.

None of these is right. They’re all approximations with different biases. The display ad might have planted the seed. The colleague’s recommendation might have been the deciding factor. Your analytics platform will never capture that conversation, but it almost certainly influenced the sale.

When I was running paid search at scale, managing hundreds of millions in ad spend across multiple markets, we spent considerable time on attribution modelling. The honest conclusion was that every model had a blind spot. Last-click made paid search look like a genius. Data-driven models were better but still systematically undervalued upper-funnel activity. The right response wasn’t to find the perfect model. It was to understand the limitations of whichever model you were using and factor that into how you interpreted the numbers. Semrush’s overview of data-driven marketing captures some of the nuance around this well.

There’s also a timing problem. Most marketing activity has a lag between investment and return. Content written today might drive conversions six months from now. A brand campaign running in Q1 might lift conversion rates across all channels in Q2 and Q3 without showing up as a direct attribution anywhere. If you’re measuring ROI on a 30-day window, you’re systematically undervaluing anything that builds slowly.

Why the Cost Side Gets Understated

Most marketing ROI calculations use media spend as the cost input. That’s a mistake. Media spend is only part of what you’re actually investing.

A more complete cost picture includes: media spend, agency or freelancer fees, internal team time, creative production, technology and tool costs, and any overhead allocated to the marketing function. When you add all of that up, the denominator in your ROI equation often doubles or triples compared to media spend alone. Which means your ROI figure drops significantly.

This isn’t an argument for pessimism. It’s an argument for accuracy. If your fully-loaded marketing ROI is still positive and still beats your cost of capital, that’s a genuinely good result. If it only looks positive because you’ve excluded half the costs, you’re making budget decisions on false information.

Early in my agency career, I worked with a client who was reporting exceptional ROI on their email programme. When we looked more carefully, they were only counting the email platform cost, not the two-person team managing it, not the content production time, and not the data management overhead. When we included everything, the programme was still profitable, but the ROI was roughly a third of what had been reported. That changed how they thought about scaling it.

Mailchimp’s breakdown of marketing metrics is useful here for thinking through what belongs in a cost calculation versus what tends to get left out.

Channel-Level ROI vs. Portfolio ROI

One of the most common mistakes I see is calculating a single blended ROI across all marketing activity and treating it as a meaningful number. It rarely is.

A blended ROI hides the performance variation between channels. Paid search might be delivering a 400% ROI. Display might be delivering 80%. Organic content might be delivering 600% on a fully-loaded basis once you account for its longevity. If you average those together, you get a number that doesn’t tell you where to invest more and where to pull back.

Channel-level ROI calculations are more useful for resource allocation decisions. They let you compare the marginal return of an additional pound invested in paid search versus an additional pound invested in content versus an additional pound invested in paid social. That comparison is what actually drives smarter budget decisions.

The caveat is that channels don’t operate in isolation. Paid search often benefits from brand awareness built by other channels. Content supports conversion rates across the board. Treating each channel as a standalone P&L can lead you to defund upper-funnel activity because it doesn’t show a direct return, then wonder why your lower-funnel performance deteriorates six months later. I’ve seen this happen more than once. A client cuts brand spend to improve short-term ROI metrics, and twelve months later they’re paying more per acquisition because the pipeline has thinned.

Forrester has written thoughtfully about the right questions to ask when improving marketing measurement, and the tension between channel-level and portfolio-level thinking comes up consistently in that kind of analysis.

How to Build a More Honest ROI Calculation

The goal isn’t a perfect number. It’s a defensible number that you and your finance team can both stand behind. Here’s how I approach it.

Step 1: Define what counts as marketing-attributed revenue. Be explicit about your attribution model and its limitations. If you’re using last-click, say so and acknowledge that it will undervalue upper-funnel activity. If you’re using a more sophisticated model, document the assumptions. The model you choose matters less than being consistent and transparent about it.

Step 2: Set a measurement window that matches your sales cycle. A business with a 12-month B2B sales cycle cannot meaningfully measure campaign ROI on a 90-day window. Match the measurement period to the reality of how long it takes for marketing investment to convert to revenue in your business.

Step 3: Use fully-loaded costs. Include media spend, agency fees, production costs, tool costs, and a reasonable allocation of team time. This will make your ROI figures look worse in the short term. It will make your budget decisions better in the long term.

Step 4: Calculate at channel level, not just in aggregate. You want to know which channels are earning their investment and which aren’t. A blended number can’t tell you that.

Step 5: Build in a margin of error. Any ROI calculation built on attribution data has uncertainty baked into it. Acknowledge that uncertainty rather than presenting a precise figure as if it were fact. A range is often more honest than a single number.

The broader context for this kind of thinking sits within a well-structured analytics practice. If you’re building or refining your measurement approach, the resources in the Marketing Analytics hub cover everything from GA4 setup to attribution frameworks and dashboard design.

Where ROI Calculations Break Down Entirely

There are categories of marketing activity where a direct ROI calculation is either impossible or actively misleading. Brand-building is the most obvious example.

Brand investment affects pricing power, conversion rates, customer retention, and the efficiency of every other marketing channel you run. None of those effects show up cleanly in an ROI calculation tied to a specific campaign. You can measure brand health metrics, track share of voice, monitor NPS over time, and observe how conversion rates shift as brand awareness grows. But you cannot draw a straight line from brand spend to revenue in the way you can with a paid search campaign.

This creates a structural problem. If you apply a consistent ROI framework across all marketing activity, brand investment will almost always look worse than performance marketing. That doesn’t mean it delivers less value. It means the value it delivers doesn’t show up in the places you’re measuring. Marketers who understand this protect brand budgets even when they can’t prove the ROI in a quarterly review. Marketers who don’t end up cutting brand spend to hit short-term efficiency targets, then spending two or three times as much on performance marketing to compensate for the demand they stopped creating.

I judged the Effie Awards for several years, and one of the things that struck me reviewing the entries was how often the most commercially effective campaigns were the ones that had invested in brand over time. The ROI on those campaigns, measured properly over a multi-year window, was extraordinary. Measured over 90 days, most of them would have looked average at best.

HubSpot makes a related point about the difference between marketing analytics and web analytics, and why conflating the two leads to narrow thinking about what marketing is actually achieving.

ROI vs. ROAS: Understanding the Difference

Return on ad spend (ROAS) is not the same as ROI, though they’re often used interchangeably. ROAS measures revenue generated per pound of media spend. ROI measures profit generated relative to total investment.

A campaign with a 5x ROAS sounds impressive. But if the product being sold has a 15% margin and you’re accounting for fulfilment, returns, and customer service costs, that 5x ROAS might represent a negative ROI. The revenue is there. The profit isn’t.

This distinction matters most in e-commerce, where high ROAS figures can mask thin margins. I’ve worked with retail clients who were scaling paid spend aggressively on the basis of strong ROAS, only to find when we modelled the full P&L that they were growing revenue while shrinking profit. The fix wasn’t to stop advertising. It was to set ROAS targets that reflected the actual economics of the business rather than the media platform’s reporting.

If you’re setting ROAS targets, work backwards from your margin. If your gross margin is 40% and you need to cover fulfilment, returns, and a reasonable marketing cost ratio, your breakeven ROAS might be 3x or 4x. Anything below that and you’re paying to acquire customers at a loss. Knowing that number changes how you bid, which products you promote, and which audience segments you prioritise.

The Benchmarking Question

A question I get asked regularly is what a good marketing ROI looks like. The honest answer is that it depends on too many variables to give a universal number.

It depends on your industry and its typical margins. It depends on whether you’re measuring ROI on a campaign, a channel, or the entire marketing function. It depends on your growth stage, because a business investing heavily in customer acquisition should expect lower short-term ROI than a mature business optimising for efficiency. It depends on your attribution model and measurement window. And it depends on what you’re including in your cost calculation.

What I’d suggest instead of chasing a benchmark is to track your own ROI consistently over time and ask whether it’s improving. Are you generating more revenue per pound of marketing investment this year than last year? Are you getting better at identifying which channels and campaigns deliver the strongest returns? Are you making budget allocation decisions based on actual performance data rather than habit or internal politics?

Those questions are more useful than any industry benchmark. The benchmark tells you how you compare to an average that may not reflect your business model, your market, or your competitive position. Your own trend line tells you whether your marketing is getting more effective over time.

Forrester’s thinking on automating marketing dashboards is worth reading if you’re building systems to track ROI consistently rather than calculating it manually each quarter.

Presenting ROI to Senior Stakeholders

One of the most underrated skills in marketing is being able to present ROI data in a way that builds credibility rather than eroding it. Most senior stakeholders have seen enough inflated marketing numbers to be sceptical by default. The way to earn their trust is to present numbers that hold up under scrutiny.

That means being upfront about the limitations of your attribution model. It means presenting ranges rather than false precision when the data genuinely doesn’t support a single number. It means including fully-loaded costs even when that makes the figures look less impressive. And it means distinguishing clearly between what you can measure directly and what you’re estimating.

Early in my career, I worked with a finance director who was deeply sceptical of marketing numbers. Not because he was anti-marketing, but because he’d seen too many presentations where the numbers changed depending on who was in the room. The moment I started presenting ROI with explicit assumptions documented and acknowledged limitations, the conversations changed entirely. He started engaging with the analysis rather than challenging the inputs. That’s the outcome you want.

The goal of an ROI presentation isn’t to make marketing look good. It’s to give decision-makers accurate information about where marketing investment is and isn’t working, so they can make better resource allocation decisions. If your ROI numbers are doing anything other than that, they’re serving the wrong purpose.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions