Incrementality testing and diminishing returns work together, since one test only produces a point estimate at a single spend level. This matters because budgeting requires knowing spend levels you haven’t tried, which a point estimate can’t answer. Without curve-based thinking, marketers assume doubling spend doubles revenue, when platforms taper off first. This video explains how to quantify that curve so doubling spend ties to a real multiplier, not a guess.
Why Incrementality Testing and Diminishing Returns Both Matter for Budgeting
Incrementality testing and diminishing returns both matter for budgeting, because a single incrementality test can only tell you what happened at the spend level you actually tested, not what would happen at a level you haven’t tried yet. In this video, John Wallace, CEO of LiftLab, explains that this is officially called a point estimate: a snapshot of revenue driven at one specific spend level, useful but incomplete once you’re trying to actually plan a budget.
Planning requires answering a harder question than a point estimate can: what happens at future or different spend levels? That’s where diminishing returns and curve-based thinking take over from linear assumptions. Wallace is candid that this moves past math most people haven’t touched since school, but marketers feel the underlying reality intuitively: platforms don’t scale infinitely, and performance tapers as spend goes deeper. The goal isn’t just to say that directionally, it’s to quantify it, so that doubling spend might be known to produce 1.6x revenue, or 1.3x, or whatever the real answer is for that specific channel, a calculation LiftLab makes available directly inside a customer’s instance.
In this video, you’ll learn:
Why incrementality tests offer limited insights for budget planning
The difference between point estimates and curve-based performance
How ad platform saturation impacts marketing returns
The importance of quantifying revenue growth beyond linear projections
How to make better long-term decisions with advanced data
TIME STAMPFull Video Transcript: Incrementality Testing and Diminishing Returns Explained
+–
In this video, John Wallace explains why a single incrementality test isn’t enough for budget planning, and why curve-based thinking about diminishing returns fills the gap.
[0:00 – 0:13] The Limitations of Point Estimates
Why can’t we stop at understanding incrementality? Officially, when we run an incrementality test, what we’re going to get is what’s called a point estimate. It’s going to be an understanding, at that spend level, how much revenue did we drive.
[0:13 – 0:28] Why We Need More Information
But to answer harder questions, when we sit down to actually plan out a budget, make hard decisions about where we’re going to spend more and where we’re going to spend less, the incrementality numbers don’t get us enough information. We need to know, at different spend levels, at future spend levels, what would happen.
[0:28 – 0:53] Embracing Curve-Based Understanding
And that’s why we embrace diminishing returns and curve-based understanding. This takes us out of an area where we’re thinking in a linear basis. I know that myself, you might be like me, I haven’t worked a derivative in several decades. But we feel this as marketers. We know that there’s saturation. We know that the ad platforms don’t scale infinitely, and that tapering of our performance needs to be taken into account.
[0:53 – 1:17] Quantifying Revenue Growth
We can’t say that when we double or triple our spend that we’re going to necessarily double or triple our revenue. And while I just articulated that directionally, we need to quantify it. We need to know that when we double our spend, we can increase our revenue by 1.6, or 1.3, or whatever that answer is. And that’s exactly what we’re making available to all the marketers that are asking these hard questions, right inside of their LiftLab instance.
Key Lessons: Moving From Point Estimates to Curve-Based Budget Planning
Beyond point estimates: Incrementality tests only show performance at one specific spend level, leaving gaps in long-term strategy.
Diminishing returns: Marketers must accept that ad performance tapers off, as platforms cannot provide infinite scalability.
Linear thinking limits: Relying on simple linear math often results in inaccurate predictions when doubling or tripling budgets.
Strategic quantifying: Moving to a curve-based model enables more precise revenue forecasting for different future spend scenarios.
