Making sense of ROI in a world where nothing works alone
We've all been there...
The boss leans back in his/her chair and asks: "Which campaign actually drove those sales?" The room goes quiet. Everyone knows the truth, but no one wants to say it. You have no idea which single thing "worked" because nothing worked alone.
It's because here's what happened: A potential customer saw your LinkedIn post, ignored it, then saw your CEO speak at a conference, got retargeted with an ad, received an email, visited your website twice, talked to a colleague who mentioned you, and six months later, they picked up the phone. Oh, the joy of B2B marketing!
So which touchpoint gets the credit?
This is the convergence problem. And it's why the "prove it" mindset, whilst completely understandable, often leads us down the wrong path.
The attribution trap
We've all been sold the dream of attribution models. First touch, last touch, multi-touch, algorithmic magic that promises to tell us exactly what worked.
However, attribution models don't actually tell you what caused the sale. They tell you what touched the sale. There's a massive difference.
Think about it like this: if you got a promotion at work, was it because of the presentation you gave last Tuesday? Or was it the three years of consistent performance, the relationship you built with your manager, the project you rescued two quarters ago, and yes, that presentation that finally crystallised everything in your boss's mind?
B2B buying works the same way. Multiple things converge over time to create the conditions for a purchase. Trying to isolate one thing and say "that's what worked" isn't just difficult - it's often misleading.
Why finance (and 'the boss') wants you to "Prove It"
Before we get frustrated with the "show me the ROI" brigade, let's acknowledge something - they're not wrong to ask.
Finance (or 'The boss') allocates scarce resources. They need to know if investing $50,000 in marketing will generate more value than investing it in product development, or sales hiring, or keeping it in the bank, or taking that 'business trip' to the conference in Tahiti. That's their job.
The problem isn't the question. The problem is the assumption that marketing works like a vending machine: put money in slot A, get sales out of slot B, calculate the ratio.
That's not how influence works. It's not how trust builds. It's not how complex buying decisions happen.
The Multiplier Effect: A different way to approach 'the numbers'
Instead of asking "which tactic worked?", the Multiplier Effect asks a better question:
"What combination of activities created enough value to change buyer behaviour?"
This shifts us from isolation to integration. From credit assignment to system thinking.
Here's a real example: A company runs brand awareness campaigns, creates product content, does outbound prospecting, hosts customer events, and has an active MD on LinkedIn. When someone becomes a customer, which thing worked?
All of them? None of them? It depends.
What we can say with confidence: the system created enough touchpoints, trust signals, and problem-solution connections that buying became the logical next step.
How to actually measure this
Alright, but how do you report to the MD, CFO, or Head of Marketing, without just saying "trust me, it all helps"?
Here are three approaches that respect The Multiplier Effect whilst providing useful insight:
1. Marketing-influenced revenue (not marketing-sourced)
Stop trying to claim that Marketing "sourced" every deal where you touched the opportunity first. Instead, track revenue where Marketing significantly influenced the outcome.
This matters because in B2B, Sales often closes the deal. Marketing creates awareness, educates the market, builds brand, qualifies interest, and warms up the conversation. All crucial. None of it happens in isolation from what Sales does.
Track deals where Marketing had meaningful touchpoints, but don't pretend you can mathematically separate Marketing's contribution from Sales' contribution. They multiplied each other's impact.
2. Incrementality over attribution
Rather than asking "which touchpoint caused the sale?", ask "what would have happened if we hadn't done this?"
This requires actual experiments:
Run campaigns in some geographic regions but not others, and compare results
Test increased investment in a channel for a quarter, then measure the downstream impact on pipeline velocity
Pause certain activities (scary, I know) and see what happens to your baseline
Incrementality thinking accepts that things work together but tries to isolate the marginal impact of changes to the system.
3. Velocity metrics over point-in-time attribution
One of the smartest ways to measure marketing's impact is to look at pipeline velocity:
How many opportunities × win rate × average deal size ÷ sales cycle length
This metric captures whether the combined system is getting faster and more efficient. It doesn't pretend to isolate variables, but it does tell you if things are improving.
When you segment this by vertical or cohort with, for example, opportunities created before and after a major campaign, you can start to see patterns without needing perfect attribution.
Five metrics that respect The Multiplier Effect
Ok, you still need to report something to your boss. You still need numbers on a dashboard. But those numbers should reflect outcomes, not attempt to assign credit to isolated activities.
These metrics let you demonstrate value without pretending you can mathematically separate what works in isolation:
1. MQL to SQL Conversion Rate
This measures the quality of your marketing engine without requiring you to attribute sales to specific touchpoints.
Are the leads marketing generates actually worth Sales' time? If your conversion rate from Marketing Qualified Leads to Sales Qualified Leads is improving, it means something in your system is working better - even if you can't pinpoint exactly what.
The insight: is your marketing getting better at identifying and nurturing genuine opportunities, or are you just generating noise?
Track this by source if you like, but remember that someone might have seen your brand campaign before clicking your LinkedIn ad. The channels multiply each other's effectiveness.
2. Qualified Pipeline Created
This is the dollar value of opportunities Marketing helped create for the next two quarters.
Present this by cohort, when the opportunity was created, and show forward pipeline by expected close quarter. This gives your MD visibility into whether future revenue targets are even coverable by the demand you're generating together with Sales.
The beauty here is that you're not claiming Marketing "caused" these deals. You're showing that Marketing touched them meaningfully. That's honest, and it's useful.
3. Pipeline Velocity
Here's that formula again: (Number of opportunities × win rate × average deal size) ÷ sales cycle in days
This is brilliant because it captures four variables at once. When this number improves, something got better. Maybe you're generating more opportunities. Maybe your win rate improved because brand strength made you the obvious choice. Maybe deals are getting larger, or closing faster.
You don't need to isolate which variable changed or which tactic caused it. You just need to know: Is the engine getting more efficient?
Track this by segment and by source every quarter. Use medians rather than averages to avoid skew from outlier deals. When you see improvement, dig into which of the four factors moved. That tells you what's working in your system.
4. CAC Payback Period
How many months does it take to recover your fully loaded marketing and sales costs from the gross margin of new customers acquired?
Critical point: fully loaded means everything. Media costs, content production, tools, team salaries, agencies, events - all of it. And revenue should be on a gross margin basis, accounting for churn within the first 12 months.
This metric respects The Multiplier Effect because it doesn't require you to isolate individual tactics. It measures whether the entire engine is efficient enough to scale.
Typically, if you're recovering costs in under 12 months, you should invest more. If it's taking over 24 months, you've got a problem. The specific tactics don't matter as much as the system-level efficiency.
Calculate this by major channel if you must, but acknowledge that even "channel-specific" results are influenced by everything else you're doing. Your event leads convert faster because you've been running brand campaigns. Your outbound works because your content educated the market first.
5. Marketing-Influenced New Annual Recurring Revenue
If someone asks "what's the ROI of marketing?", this is your answer - but frame it carefully.
Track new revenue (or gross profit) where Marketing either sourced the opportunity or significantly influenced it. The key word is "influenced," not "caused."
In B2B, Sales closes deals. Marketing creates conditions where deals become closeable. Both matter. Both contributed. The Multiplier Effect is the whole point.
If you have the analytical maturity, use incrementality experiments or marketing mix models. If not, multi-touch attribution is acceptable, but only if you can validate it with geographic tests on major channels. Don't just implement a fancy model and assume it's telling the truth.
Most importantly, be honest that you're measuring influence, not claiming sole credit. There's another variable in this equation called Sales effectiveness, and it's unwise to treat that as a constant when you're doling out credit to Marketing.
Addressing the "Prove It" mindset constructively
When someone asks you to prove marketing's ROI, here's how to respond without either capitulating to bad measurement or dismissing their concern:
Acknowledge the legitimacy: "You're right to want to understand if marketing investment is creating value. Let me show you how we're measuring that."
Reframe the question: "Rather than isolating individual tactics, which is misleading, here's what we can demonstrate: the speed at which we're converting interest into revenue, the efficiency of our investment recovery, and the incremental impact of scaling our efforts."
Show trends, not single points: "This quarter, our payback period improved from 18 to 14 months. Pipeline velocity increased by 22%. When we increased investment in this segment by 30%, we saw a 45% lift in SQL generation within two quarters."
Be honest about uncertainty: "Can I tell you exactly which touchpoint 'caused' each sale? No, because that's not how buying works. Can I show you that when we systematically invest in these combined activities, we get predictable improvements in business outcomes? Yes."
Some executives want a simple answer because simple answers feel controllable. They want to cut the "thing that doesn't work" and double down on the "thing that does."
But in a convergence model, cutting something that appears to have low attribution might actually collapse the entire system. Your brand campaigns might not show up as last-touch very often, but they might be the reason your sales calls get returned at all.
This doesn't mean you can't make trade-offs or prioritise. It means you need to:
Test changes systematically
Look at system-level outcomes
Accept that some uncertainty is irreducible
Focus on velocity and efficiency metrics that capture combined effects
The Multiplier Effect
The Multiplier Effect isn't about avoiding accountability. It's about having accountability for the right things.
Don't measure individual touchpoint attribution when you know buying doesn't work that way. Do measure whether your combined system is creating enough value to change buyer behaviour at an acceptable cost and speed.
Don't pretend you can mathematically separate marketing's impact from sales' impact. Do measure whether the pipeline you're creating together is healthy and growing.
Don't ignore the "prove it" mindset. Do educate your colleagues, or boss, about what can actually be proven, and what kinds of evidence should shift investment decisions.
In the end, the best question isn't "which campaign worked?"
A better question is: "Is our integrated system of marketing and sales activities creating sufficient value, quickly enough, and efficiently enough, to meet our growth objectives?"
That's a question you can actually answer (OK, following a big deep breath!)
And it's the one that matters.
