Growth Decomposition (Granularity of Growth)
When you need this method
Your growth rate is known, its cause is not. Until you separate how much came from the market, how much was won against competitors, and how much came from acquisitions, you will act on the wrong lever. A company riding a growing segment looks healthy on almost every metric, even while it is losing share in that very segment.
Approach
- 1Cut the business into fine cells, meaning product by customer segment by region, rather than into divisions or reporting units.
- 2Determine market growth per cell for the period under review and state where that figure comes from and how reliable it is.
- 3Compute the market pull: the cell's revenue base times the cell's market growth. Summed across cells, this is the portfolio contribution.
- 4Isolate the share effect as a difference, namely own growth per cell minus the market pull of that cell.
- 5Report acquisitions separately so that acquired revenue does not end up counted as organic performance.
- 6Add the three contributions and reconcile them against the reported growth rate. A large residual means the cell structure or the market data is wrong.
- 7Attach action to the largest contribution: shift the portfolio, fight for share, or steer acquisitions differently.
Typical application
A typical B2B SaaS in the HR space grows about thirty percent for three years running and reads this as proof of a strong sales system. Decomposing by product, segment, and country tells a different story: most of the increase comes from two cells where the market itself grows at double digits, while the company's own share slips slightly in both. In a third cell, a flat market, the company gains share clearly. The conclusion is not more sales headcount but the question why what demonstrably works in the flat cell does not carry over into the growing ones.
Limits and counter-indications
The method lives and dies by market data per cell. Where segment growth is merely estimated, you are computing precisely on imprecise numbers. The share effect emerges as a residual and absorbs every error in market data and cell definition, a point regional economics has long made about the related shift-share calculation. The decomposition explains a past period and is not a forecast: firm growth carries a dominant stochastic element, and the mix of one period does not reliably repeat in the next. For a company in a single, poorly bounded, or newly emerging market there are no meaningful cells and therefore no market pull to net out. The originators' empirical claim that share gain contributes least at large corporations rests on their own non-public sample of very large firms and cannot be recomputed from outside; for smaller vendors the share effect carries visibly more weight.
How to measure impact
Report the three contributions in percentage points of the growth rate, check that they reconcile to the reported rate, and track across several periods what share of growth comes from your own share gain.
Related methods
Sources
- 1.Seyfried: Shift-Share Analysis, Business Quest (University of West Georgia), 1996 (opens in a new tab) · Business Quest, Richards College of Business, University of West Georgia · 1996 · academic and scholarly literature · describes the methodCarries the decomposition arithmetic the method rests on: measured growth is split additively into the pull of the overall market, the structural effect of the segments a unit operates in, and a competitive effect, with an explicit formula per component, and the three components must account for the observed rate. Boundary: the treatment comes from regional economics, has no separate term for acquisitions, prescribes no level of cell granularity, and states explicitly that the competitive effect is a residual.
- 2.Viguerie, Smit, Baghai: The Granularity of Growth, John Wiley & Sons (opens in a new tab) · John Wiley & Sons (Nachweis über Google Books) · 2008 · practitioner source · provides the contextEstablishes authorship, publisher, and year, along with the claim that growth must come from multiple sources and that markets should be viewed both broadly and granularly. Boundary: the publicly available description names neither the three drivers nor the authors' study figures; both sit inside the book text, and the underlying company sample is not public.
- 3.Coad: Firm Growth. A Survey, Documents de travail du Centre d'Economie de la Sorbonne, 2007 (opens in a new tab) · Université Panthéon-Sorbonne (Paris 1), Centre d'Economie de la Sorbonne · 2007 · academic and scholarly literature · limits the methodBounds the reach of the method: this survey of the empirical literature records that firm growth is dominated by a stochastic element and is therefore hard to predict, and that research into its determinants has had limited success. Boundary: the paper says nothing about this method as such, it only contradicts reading a decomposition as a forecast.
Origin: McKinsey (Viguerie/Smit/Baghai)