GE-McKinsey Nine-Box Matrix
- Introduced by
- General Electric & McKinsey & Company; Arnoldo Hax & Nicolas Majluf
- Year
- 1983
- Type
- Multifactor portfolio-classification pipeline for resource allocation
- Family
- Strategic management / corporate portfolio planning
What does it do?
The GE-McKinsey nine-box matrix is a multifactor portfolio-analysis tool that positions a company's business units in a three-by-three grid defined by two composite dimensions: the attractiveness of the industry the unit competes in, and the unit's competitive strength within it.
Developed by General Electric with McKinsey & Company in the early 1970s as a richer alternative to the BCG growth-share matrix, it replaces single proxies (market growth and relative share) with weighted indices built from many underlying factors. Hax and Majluf's 1983 Interfaces article gave the matrix a systematic methodological treatment, and Wind, Mahajan, and Swire's 1983 Journal of Marketing study empirically compared it with other standardized portfolio models, showing how much business positions depend on model choice. The nine cells map onto invest-grow, selectivity, and harvest-divest zones that guide resource allocation.
The intuition behind the method
The BCG matrix is fast but blunt: it judges a market by its growth alone and a business by its relative share alone. The GE-McKinsey matrix keeps the same basic question — is this a good business in a good market, and how should we fund it? — but answers it with more nuance. To rate a market's attractiveness it weighs size, growth, profitability, competitive intensity, regulation, and more into one score; to rate a unit's strength it weighs share, cost position, brand, technology, and capabilities into another. Plotting the two on a three-by-three grid yields a richer picture: the top-left corner is invest hard, the bottom-right is harvest or exit, and the diagonal is a zone of selectivity where you choose carefully. The cost of the added realism is judgment: someone must pick the factors and weights.
History
The nine-box matrix arose in the early 1970s when General Electric, managing a vast portfolio of businesses, worked with McKinsey & Company to develop a screen more discriminating than the BCG growth-share matrix, which GE found too reductive. The tool spread across large diversified corporations and was echoed in Shell's contemporaneous directional policy matrix. By the early 1980s it had a substantial literature: Hax and Majluf's 1983 Interfaces article provided a careful methodological account of constructing and using the industry attractiveness-business strength matrix, while Wind, Mahajan, and Swire's 1983 empirical comparison of standardized portfolio models exposed how much a business's apparent position depends on which model and weights are chosen. Together these works established both the matrix's value as a multifactor portfolio tool and the caution required in interpreting it.
When to use it
Use the GE-McKinsey matrix when a diversified firm needs a portfolio assessment richer than the BCG matrix can give — when industry attractiveness and competitive strength each depend on several factors that a single proxy would miss, and when stakeholders can supply informed judgments and weights for those factors. It suits resource-allocation and prioritization decisions across business units, especially where industries differ in profitability and structure, not just growth. It is most credible when the factor list and weights are agreed transparently and tested for sensitivity. It is less appropriate when quick directional triage suffices (the BCG matrix is simpler), when the scoring would be too subjective to trust, or when the real question concerns the internal cash-flow succession the BCG matrix foregrounds rather than multifactor attractiveness.
Limitations and considerations
- Heavily dependent on subjective choices of factors, ratings, and weights, which can be manipulated to justify a desired conclusion.
- The composite scores obscure the underlying drivers, making it hard to see why a unit sits where it does.
- Wind, Mahajan, and Swire showed positions can shift substantially across portfolio models, so results are not model-invariant.
- Like all such matrices it offers directional prescriptions, not quantified financial forecasts, and can mislead if treated as decisive.
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Captures industry attractiveness and competitive strength as multifactor composites, far richer than the BCG matrix's single proxies.
- Accommodates industry-specific factors and weights, letting analysts encode domain knowledge into the assessment.
- Its nine cells give finer-grained, zoned guidance (invest, select, harvest) than a four-quadrant scheme.
- Bubble sizing conveys the scale of each business alongside its position, aiding portfolio prioritization.
Applications
- Prioritizing investment across the business units of a diversified corporation using multifactor scores.
- Assessing portfolio balance when industries differ in profitability and structure, not merely growth.
- Identifying units in the harvest-divest zone and those warranting selective, earned investment.
- Encoding industry-specific knowledge into attractiveness and strength factors for tailored portfolio review.
- Cross-checking BCG growth-share conclusions with a richer multifactor portfolio assessment.
How it works
How it works
The vertical axis is a composite index of how attractive the unit's industry is. The analyst lists the factors that make an industry attractive — market size and growth, profitability and margins, competitive intensity, cyclicality, regulatory and technological conditions — assigns each a weight summing to one, rates the industry on each factor, and combines them into a weighted score. Hax and Majluf describe this construction in detail, noting that the choice and weighting of factors is where industry knowledge enters the analysis. The result is a single attractiveness value that is far richer than market growth alone.
Formula
A_j = \sum_{m} w_m\, a_{jm},\quad \sum_m w_m = 1,\ a_{jm}\in[1,5]\ (\text{size, growth, margins, intensity})The horizontal axis is a parallel composite index of the unit's competitive strength within its industry, built the same way from factors such as relative market share, cost position, product quality and brand, distribution, and technological and managerial capability, each weighted and rated and combined into one score. Hax and Majluf emphasize that strength should reflect the unit's standing against the specific competitors it faces. This composite replaces the BCG matrix's reliance on relative share alone with a multidimensional view of competitive position.
Formula
B_j = \sum_{n} v_n\, b_{jn},\quad \sum_n v_n = 1,\ b_{jn}\in[1,5]\ (\text{share, cost, brand, capability})Each unit is then placed on the grid according to whether its attractiveness and strength scores fall in the low, medium, or high band, producing a three-by-three matrix of nine cells. Bubbles are typically sized by the unit's revenue or the size of its market, so the grid conveys not only position but scale. Hax and Majluf treat this placement as the synthesis step that turns two abstract indices into a spatial portrait of the whole portfolio, with each cell carrying a distinct strategic meaning.
Formula
(A_j,B_j)\mapsto \text{cell}\in\{L,M,H\}\times\{L,M,H\}\ \Rightarrow\ 3\times 3\text{ grid}The nine cells collapse into three strategic zones along the grid's diagonal. Units in the high-attractiveness, high-strength corner sit in the invest-and-grow zone and warrant priority funding; units in the low-low corner sit in the harvest-or-divest zone; units along the diagonal occupy a selectivity zone where investment should be earned and targeted. Hax and Majluf lay out these prescriptions while cautioning that they are directional guidance, not mechanical commands, and that the underlying scores must be interrogated before acting.
Formula
\text{zone}= \begin{cases}\text{invest/grow}& A+B\ \text{high}\\ \text{selectivity}& \text{diagonal}\\ \text{harvest/divest}& A+B\ \text{low}\end{cases}Because the matrix's output depends on chosen factors and weights, a disciplined analysis tests how robust the placements are. Wind, Mahajan, and Swire demonstrated empirically that the same businesses can land in different positions under different standardized portfolio models, so good practice checks sensitivity to weighting assumptions and compares positions across models such as the BCG and GE-McKinsey matrices. Agreement across models and across reasonable weight choices increases confidence; large swings warn that the conclusion is an artifact of modeling decisions rather than a stable feature of the portfolio.
Formula
\text{corr}\big(\text{position}_{GE},\,\text{position}_{BCG/other}\big);\ \text{check sensitivity to } w,vCommon pitfalls
- Choosing factors and weights after the fact to push a favored business into the invest zone, biasing the analysis.
- Hiding wide uncertainty in single point scores and presenting placements as more precise than they are.
- Reading the nine cells as automatic commands rather than directional zones requiring further analysis.
- Skipping sensitivity and cross-model checks, so a placement that is an artifact of the weighting passes for a robust finding.
Frequently asked questions
How does the GE-McKinsey matrix differ from the BCG growth-share matrix?
Both place business units on a two-axis grid to guide allocation, but the GE-McKinsey matrix uses composite multifactor indices instead of single proxies. The BCG matrix rates market attractiveness by growth alone and competitive position by relative share alone; the GE-McKinsey matrix builds industry attractiveness from size, growth, profitability, intensity, and more, and business strength from share, cost, brand, and capabilities. It also uses a three-by-three grid with invest, selectivity, and harvest zones rather than four quadrants. The trade-off is richness for added subjectivity in choosing factors and weights.
How are the factors and weights chosen, and why does it matter?
The analyst selects the factors that determine industry attractiveness and competitive strength, assigns each a weight summing to one, rates each unit, and combines them into composite scores. This is where domain expertise enters — but Hax and Majluf and others stress it is also where bias can enter. Because Wind, Mahajan, and Swire showed that placements can change materially under different models and weights, the factor and weight choices should be made transparently and tested for sensitivity so the conclusions are not artifacts of those choices.
What do the three diagonal zones mean?
The nine cells group into three zones. The upper-left cells (high attractiveness and high strength) form the invest-and-grow zone, deserving priority resources. The lower-right cells (low attractiveness and low strength) form the harvest-or-divest zone. The cells along the diagonal form a selectivity zone, where the business is mixed and investment should be targeted and justified rather than automatic. Hax and Majluf present these as directional guidance to structure allocation debates, not as mechanical decisions, since the scores behind them carry judgment and uncertainty.
Related methods
Sources
Hax, A. C., & Majluf, N. S. (1983). The Use of the Industry Attractiveness-Business Strength Matrix in Strategic Planning. Interfaces, 13(2), 54-71.
Open DOI referenceCopied.Wind, Y., Mahajan, V., & Swire, D. J. (1983). An Empirical Comparison of Standardized Portfolio Models. Journal of Marketing, 47(2), 89-99.
Open DOI referenceCopied.