The same decade, the same economy, very different returns
Michael Porter's 2008 analysis of US industries between 1992 and 2006 put the median average return on invested capital at 14.3%. The tenth percentile sat at 7.0%. The ninetieth at 25.3%.
Airlines averaged 5.9%. Soft drinks and prepackaged software ran close to 37%, making them nearly six times more profitable than commercial aviation across the same period.
Same economy. Same fifteen years. Same national pool of management talent, capital, and technology.
Something other than execution quality is producing that spread, and it is worth knowing what — because it determines how much a business gets back for being well run.
The five forces, briefly
Porter's framework identifies five structural pressures that determine how much of the value an industry creates is retained by the firms competing in it.
Rivalry among existing competitors. Intensity and, more importantly, basis. Price rivalry destroys industry profitability. Rivalry on differentiated dimensions does not.
Bargaining power of buyers. Concentration, switching costs, price sensitivity, ability to integrate backwards.
Bargaining power of suppliers. Concentration, uniqueness of input, switching costs, ability to integrate forwards.
Threat of new entrants. Not whether entrants have arrived, but what stops them — capital requirements, scale economies, distribution access, regulation, switching costs.
Threat of substitutes. Different means of accomplishing the same job, which sets a ceiling on price.
Porter published the original framework in Harvard Business Review in 1979 and expanded it in Competitive Strategy in 1980.
The evidence, which is more interesting than the framework
This is the only empirically grounded framework in the Define set. It has been tested repeatedly by researchers who had no commercial interest in the result. What they found is more nuanced than either its advocates or its critics usually report.
Richard Rumelt's 1991 study partitioned variance in business-unit profitability. Stable industry effects accounted for roughly 8%. Stable business-unit effects accounted for around 46%. Corporate parentage was close to negligible.
Anita McGahan and Porter's 1997 study, using a different sample and method, found industry effects at 19%, business-specific effects at 32%, corporate-parent at 4%, and year effects at 2%.
Read those honestly and the conclusion is uncomfortable for both camps. Industry structure matters — it is not noise, and 8% to 19% of profitability variance is a great deal of money. But firm-specific factors matter roughly twice as much. Which industry you are in explains less about your returns than what you do inside it.
Two refinements make this more useful.
Industry effects persist longer. McGahan and Porter's later work found industry effects more durable than corporate or business-unit effects. Firm-level advantage erodes faster than structural advantage. A structural position is harder to win and harder to lose.
Industry effects vary enormously by sector. They explain a smaller share of variance in manufacturing and a larger share in services, wholesale and retail trade, lodging and entertainment, and transportation. For a consumer retail or services business, structure is doing more of the work than the headline averages suggest.
Why it is diagnosis and not prescription
The framework tells you what the category permits. It does not tell you what to do.
That distinction is where most misuse originates. A completed five forces analysis produces a verdict — this industry is structurally attractive, or it is not. Teams then look at the verdict for instruction and find none, because none is there.
What the analysis is actually for:
Setting expectations about achievable margin. If buyer power is high and switching costs are near zero, a business plan that assumes expanding margin needs a reason that is visible in the forces.
Explaining persistent margin pressure that operational fixes have not resolved. When three rounds of efficiency work have not moved gross margin, the cause is frequently structural, and continuing to optimise is expensive.
Testing a where-to-play choice before committing capital. A strategic choice that requires the business to earn returns the category has never produced needs an explicit account of what it will change.
Identifying which force to attack. This is the framework's most underused application. Structure is not fixed. A business can raise switching costs, reduce supplier concentration by qualifying alternatives, or change the basis of rivalry away from price. Naming which force is doing the damage makes it a target rather than a condition.
Where the framework breaks
The industry boundary is drawn wrong. This is the most common and most consequential error. Draw it too broadly and every force looks moderate. Too narrowly and the analysis excludes the substitutes that actually cap pricing. The boundary should be drawn where the economics are similar, not where the industry classification code sits.
It is a snapshot. Five forces describes a structure at a point in time. It has no mechanism for representing how quickly a force is changing, which is often the more important question. Pairing it with an evolution-based view addresses this.
Complements are missing. Porter's framework does not treat complementary products as a distinct force, though they materially affect profitability in software, hardware, and platform categories.
Platform and network dynamics fit awkwardly. Multi-sided markets, where the buyer on one side is the product on the other, strain a framework built for linear value chains.
Structural analysis alone does not explain firm performance. The variance research is explicit about this. Within any industry, the spread between firms is wider than the spread between industries. Structure sets the range. It does not determine where in the range a business lands.
How to run it properly
Four disciplines separate a useful analysis from a filled-in template.
Analyse forces, not lists. The output is not five bullet lists. It is a judgement about which force is most binding on profitability now, and why.
Look for the underlying cause. Buyer power is not an observation. It is a consequence of concentration, switching costs, or price transparency. The cause is what can be acted on.
Assess for the medium term. Short-run factors — the business cycle, input price spikes, a competitor's temporary discounting — are not structure. Structure is what still applies in three years.
State what would change the verdict. A structural analysis with no falsifying condition attached is a description. Naming what would have to shift makes it a monitorable position.
Diagnostic: is this a structural analysis or a summary?
Six tests.
The industry boundary is stated explicitly, with a reason for where it was drawn.
One force is identified as most binding, rather than all five described at equal weight.
Each force is explained by an underlying cause, not just rated high or low.
Substitutes include ways of accomplishing the job that are not products in this category, including doing nothing.
The analysis names at least one force the business could plausibly change.
The expected margin implied by the structure has been compared against what the financial plan assumes.
Test six is where structural analysis meets the budget. A plan assuming returns the category does not produce is not an ambitious plan. It is an unfunded assumption.
What this produces
A bounded expectation.
Structural analysis tells a leadership team what range of returns the category has historically permitted, which force is compressing them, and what would have to change for that to shift. It bounds the strategic choice without making it.
The research is clear that this is a partial answer — structure explains meaningfully less of the variance in profitability than firm-specific factors do. But it is the part that is hardest to fix later, and the part that persists longest once established.
Knowing what the category permits before choosing where to play is the difference between an ambitious plan and an expensive one.
Frequently asked questions
Is Porter's Five Forces still relevant?
Yes, with a clear understanding of what it does. It remains the reference framework for structural profitability and has held up across decades of independent research. It is a diagnostic tool, not a source of strategic recommendations, and it handles platform and multi-sided markets poorly.
How much does industry structure actually determine profitability?
Research puts stable industry effects at roughly 8% of variance in business-unit profitability in Rumelt's 1991 study, and 19% in McGahan and Porter's 1997 study. Business-specific effects were larger in both — 46% and 32% respectively. Structure matters and firm-level choices matter more.
Does industry matter more in some sectors?
Yes. Industry effects account for a smaller share of profitability variance in manufacturing and a larger share in services, wholesale and retail trade, lodging and entertainment, and transportation.
What is the most common mistake in a five forces analysis?
Drawing the industry boundary incorrectly. Too broad and every force looks moderate; too narrow and real substitutes are excluded. Draw the boundary where the underlying economics are similar.
Can a business change its industry structure?
Sometimes, and this is the framework's most underused application. Raising switching costs, qualifying alternative suppliers, or shifting rivalry away from price all alter structure. Naming the binding force converts it from a condition into a target.
Should Five Forces be used before or after customer research?
Structural analysis comes first in the Define sequence. It bounds what any choice can deliver. Customer outcome research then establishes what is underserved inside those bounds.
How often should structural analysis be refreshed?
When a force materially changes — a major entrant, a substitute reaching price parity, consolidation on the supply side, or a regulatory shift. Structure is durable, so calendar-driven refreshes add little.
Sources
Porter, M.E., How Competitive Forces Shape Strategy, Harvard Business Review (1979); Competitive Strategy (1980); The Five Competitive Forces That Shape Strategy, Harvard Business Review (2008), including US industry ROIC data for 1992 to 2006
Rumelt, R.P., How Much Does Industry Matter?, Strategic Management Journal (1991)
McGahan, A.M. and Porter, M.E., How Much Does Industry Matter, Really?, Strategic Management Journal (1997), and subsequent work on the persistence of industry effects
Structure your next phase
Zerologic runs market and structural analysis that establishes what a category permits — before a leadership team commits capital to a position inside it.
Talk to us: partners@zerologic.io · zerologic.io
