Figure 1From “The thirty-percent rule for sales and marketing”
The measured investment share of SG&A by industry
Comparison of standard BEA-HH assumptions versus empirical market-based exit parameters across major industries.
Figure 1The measured investment share of SG&A by industry
Source: Data from Ewens, Peters & Wang (2019, revised October 2023), NBER Working Paper No. 25960, Table 1 (p. 47).

Source and evidence
Each line names a claim related to this exhibit, its source, and how it was checked.
- A EPW24-C5 R&D depreciation is estimated, not assumed, and it moves with R&D intensity: "we find a significantly higher depreciation rate for the two industries with the highest R&D intensity: healthcare (33% vs 17%) and high-tech (42% vs 28%)"Ewens, Peters & Wang (2019, revised 2023), Table 1
- A EPW24-C6 The SG&A share is the number the 30% rule stands on, and the paper states both the estimate and its spread: "we find that our 28% estimate of the fraction of SG&A representing invested capital is similar to that used in earlier work (30%)", while "this ratio varies dramatically across industries, from 20% (consumer) to" 51% (healthcare)Ewens, Peters & Wang (2019, revised 2023), Table 1 and section 1
Grades: A, verified against the printed page of the primary source · B, primary source, text layer only · C, authoritative secondary · D, reported.
From the essay
Growth that compounds
The thirty-percent rule for sales and marketing
Finance models assume 30% of SG&A builds an asset. Market exit prices prove the real share ranges from 20% in consumer goods to 51% in healthcare.
Read the essayThis exhibit belongs to an essay: the numbers live here, the argument lives there.
Cite or embed
Related exhibits

Table 1Table
Every published depreciation rate, and what it was measured on
Every growth budget is a gross number

Table 1Table
Commercial budgeting matrix: operating cost vs capital creation
The thirty-percent rule for sales and marketing

Table 3Table
Which empirical transfer test must precede deploying a predictive model across cohorts?
One customer model cannot predict every outcome
Contact
Bring the question you're working on.
Research, a guest lecture, or a commercial question.