Marketing Metrics
The reference book on marketing math. Not a cover-to-cover read, but the one you keep on the desk when you need a metric defined and computed correctly.
What it is
Four academics set out to define, derive, and stress-test the metrics marketers actually use - and largely succeeded. The book is encyclopedic: margin and contribution, customer lifetime value, ROI and return on ad spend, share metrics, channel and pricing math, each with its formula, a worked example, and the traps that produce wrong numbers.
It reads like a well-organized reference rather than a narrative, which is exactly what makes it durable. More than a decade on, the definitions still hold because the math does not change when the channels do.
What's strong
Its insistence on getting the denominator right is a gift to anyone who has watched a team confuse gross and net, or quote a ROAS that ignores margin. The chapters on margin and customer value are particularly strong, and the recurring 'data sources, complications, and cautions' notes are where the real expertise shows - they tell you how each metric gets gamed or misread.
For a performance team, it is the antidote to dashboard metrics that look precise but mean nothing.
Where it stops
It is academic in tone and not light reading; you consult it, you do not curl up with it. The digital-specific examples are dated, and it predates the privacy and attribution upheavals that complicate measurement today - so pair the timeless math with current thinking on incrementality and blended measurement.
Who should read it
Anyone who reports marketing numbers to a CFO, analysts who want their formulas above reproach, and operators building a metrics framework. It pairs directly with our ROAS, CAC and LTV calculators.
How RGM uses it
This sits on the desk as our tie-breaker whenever two stakeholders define a metric differently. When someone quotes a ROAS that quietly ignores margin, or a CAC that omits a real cost, we go to the relevant chapter, agree on the formula, and move on. The 'cautions' notes are the part we lean on most - they are a catalogue of the ways each number gets misread or gamed, which is exactly the institutional memory a measurement framework needs. We map its definitions onto our calculator suite so the math a client sees in a tool matches the math in the canonical reference, with no quiet discrepancies.