Reference

Frameworks: the intellectual foundation in depth

Part of the Market Competitive Intelligence skill · loaded on demand from SKILL.md

Read this when you need the original method behind a stage of the skill: the reasoning Gadiesh and Gilbert give for each of the four mapping steps, the drivers Porter lists under each force and the pitfalls he warns against, the value-chain decomposition, the mechanics of triangulated sizing, and the components of a competitor analysis. SKILL.md gives the condensed version; this file gives enough to apply each method with judgment when the situation does not match the template.

Contents

  1. Profit Pool analysis (Gadiesh & Gilbert, 1998)
  2. Porter's Five Forces (1979; 2008)
  3. Value-chain mapping (Porter, 1985)
  4. Market sizing by triangulation
  5. Competitor fact-base construction (Porter, 1980)
  6. How the methods fit together

1. Profit Pool analysis (Gadiesh & Gilbert, 1998)

Sources

Orit Gadiesh and James L. Gilbert, both of Bain & Company, published two companion articles in the May-June 1998 issue of Harvard Business Review: "Profit Pools: A Fresh Look at Strategy" (the concept and its strategic uses) and "How to Map Your Industry's Profit Pool" (the method). Read together, the first article says why to look at profit rather than revenue and the second says how to build the map.

The claim

A profit pool is the total profit earned in an industry at all points along the value chain. The central observation is that the distribution of profit across activities, customer segments, channels, products and geographies is usually very different from the distribution of revenue. Some activities with modest revenue earn most of the profit; some activities with most of the revenue earn very little. Companies that manage to the revenue map (pursuing the largest segments, growing volume in the activity they already occupy) routinely miss where the industry's profit actually sits, and are then surprised when a competitor who mapped the pool reorganises around the deep parts.

The first article illustrates this with truck rental, where the rental activity itself earns thin margins while the surrounding activities (accessories, insurance, trailer rental, storage) earn most of the profit, and with the automotive value chain, where downstream activities such as financing, leasing, insurance, used vehicles and service earn a large share of the industry's profit on a small share of its revenue. The pattern generalises: after-sales, financing, consumables, data and services are the usual deep pools around a product.

Why it works (the mechanism)

The map forces three separations that ordinary industry analysis blurs:

  1. Revenue from profit. Plotting activities as bars whose width is revenue and whose height is margin makes profit visible as area. A wide flat bar and a narrow tall bar can hold the same profit; managers who have only seen the revenue view have never had to confront that.
  2. The industry from the company. The pool is defined over all participants performing each activity, not over the client's own lines of business. This exposes activities the client does not perform, which is where entry, adjacency and integration opportunities live.
  3. Current from future. Adding the trend in each pool (five years back, five years forward) turns the map into a statement about where profit is migrating, which is the input strategy actually needs.

The authors also identify the uses of the map: choosing where to compete within the chain, deciding whether to expand into adjacent activities, spotting which customer or product segments to emphasise, seeing where a new entrant or a business-model change is attacking the pool, and rethinking the economics of the business (for example moving from selling a product to capturing the downstream service pool).

The four-step mapping method

The second article describes the method as four steps. The wording below paraphrases the article; the reasoning is what to retain.

Step 1: Define the pool. Decide which activities are inside the value chain being mapped, which products, customers and geographies are included, and what profit measure will be used. The authors advise defining the chain broadly enough to include activities adjacent to the company's own, because the point of the exercise is to see profit the company does not currently touch. They also warn that the boundary should follow the economics of the customer's purchase rather than the company's organisation chart. Choose the profit measure to be comparable across activities: operating profit is the usual choice; use return on invested capital or economic profit where the activities differ sharply in capital intensity, because a 6 percent margin on a capital-light activity may be more attractive than a 12 percent margin on a capital-heavy one.

Step 2: Determine the size of the pool. Estimate the total profit of the whole chain. The article suggests starting from whatever aggregate is most reliable (total industry revenue times a defensible blended margin, or the sum of the profits of the largest participants plus an estimate for the rest) and producing a rough total before any activity-level work. The purpose is to have a control total against which the activity estimates in step 3 are reconciled.

Step 3: Determine the distribution of profits. Estimate the profit in each activity. This is the hardest step because company data do not align with activities: companies report by legal entity or reportable segment, not by value-chain stage, and most sizeable participants perform several activities. The article's own words are that financial data are rarely reported in bundles that correspond to value-chain activities and that translating them takes considerable creativity; it does not prescribe a single technique. In practice three approaches cover most cases, used in whichever combination the data permit: aggregate the activity-level profits of the companies that participate (where pure plays or segment data exist), work from a sample of representative companies and scale up, or infer margins from the economics of the activity where no company data exist. Where the largest players are diversified, split them by activity using their own disclosures or by analogy to a pure play.

Step 4: Reconcile the estimates. Sum the activity profits from step 3 and compare with the control total from step 2. Investigate any gap: usually an activity has been double counted (the same profit attributed to a manufacturer's segment and to a distributor's), an activity has been missed, or one of the margin proxies is not representative. Iterate until the two totals agree within a tolerance appropriate to the decision. The article stresses that the map does not need to be precise to be useful, but it does need to be consistent; a map whose parts do not add up cannot be defended.

Dimensions along which a pool can be cut

The value-chain activity is the primary dimension and the one this skill maps first, but the authors note that the same pool can be cut by customer segment, product line, channel or geography, and that the most useful map is often the one that combines two dimensions (activity by customer segment, for example). Cut the pool a second way when the decision concerns a segment rather than an activity, and check that the two cuts sum to the same total; a pool that adds to $20bn by activity and $23bn by segment has a double count in one of them.

Reading the map: recurring patterns

Four shapes recur and each has a standard strategic reading:

  • A wide, shallow bar next to a narrow, deep one (the client is usually in the wide one). The deep pool is an adjacency to consider entering, or a pool a competitor will use to subsidise an attack on the client's activity.
  • Depth concentrated upstream (manufacturers or IP holders). The channel is a cost centre for the chain; distributors compete on cost to serve, and the strategic question is scale.
  • Depth concentrated downstream (service, financing, consumables). The product is the entry ticket; the strategic question is who owns the customer relationship after the sale.
  • A pool that was deep and is flattening. Structure is changing; run the forces on it before assuming the trend continues, because the flattening is often cyclical.

Judgment points the method does not settle

  • Revenue at different stages overlaps (the manufacturer's revenue is the distributor's cost of goods), so the revenue bars do not sum to end-customer spend. State which revenue basis each bar uses. Profit, by contrast, is additive across stages, which is why the pool is measured in profit.
  • Services that are invoiced separately from the product should be a separate activity with their own revenue; services bundled into product price should be estimated as a margin uplift on the product activity, not added as a new bar, or they will be double counted.
  • Normalise the year. A pool measured at a cyclical peak overstates the pools that are most cyclical.
  • A pool is not a segment of the client's P&L. Do not let the client's own margins stand in for the activity's margin unless the client is representative.

2. Porter's Five Forces (1979; 2008)

Sources

Michael E. Porter, "How Competitive Forces Shape Strategy", Harvard Business Review, March-April 1979 (vol. 57, pp. 137-145), introduced the framework; his book "Competitive Strategy" (1980) developed it. Porter, "The Five Competitive Forces That Shape Strategy", Harvard Business Review, January 2008, restated the framework with new material on how to apply it, what is not a force, and the common mistakes. The 2008 article is the version to work from; the 1979 article establishes the original claim.

The claim

The profitability of an industry over the medium and long run is determined by its structure, and structure is captured by five forces: rivalry among existing competitors, the threat of new entrants, the bargaining power of suppliers, the bargaining power of buyers, and the threat of substitute products or services. The strongest force or forces determine the industry's profitability and are therefore the most important to strategy. Industry structure is relatively stable; it explains why some industries earn high returns for decades while others, equally large or fast-growing, earn little.

Why it works (the mechanism)

Each force names a party with which the industry must share the value it creates. Rivalry transfers value to customers through price competition and to the cost base through arms races in marketing and capacity. Entrants force incumbents to hold prices down or invest to deter entry. Substitutes cap the price customers will pay. Powerful buyers extract lower prices or more service for the same price. Powerful suppliers raise input costs. The forces are therefore an accounting of where the pool leaks, which is exactly what the profit-pool map cannot tell you: the map shows the pool is shallow, the forces show who is draining it.

The framework also explains the client's position only indirectly. Porter is explicit that industry structure determines the average profitability available to participants; a company's own profitability depends additionally on its position within the industry. Confusing the two is the most common misreading and the reason SKILL.md requires the client's share of revenue and share of profit to be reported alongside the forces.

The drivers under each force (2008)

Threat of new entrants depends on barriers to entry and expected retaliation. Barriers:

  • Supply-side economies of scale.
  • Demand-side benefits of scale (network effects).
  • Customer switching costs.
  • Capital requirements.
  • Incumbency advantages independent of size (proprietary technology, best locations, established brands, cumulative experience).
  • Unequal access to distribution channels.
  • Restrictive government policy.

Retaliation is expected where incumbents have responded forcefully before, hold resources to fight, are committed to the industry, or where growth is slow so that an entrant's volume must come from incumbents.

Power of suppliers is high where:

  • The supplier group is more concentrated than the industry it sells to.
  • The supplier group does not depend heavily on the industry for its revenues.
  • Participants face switching costs in changing suppliers.
  • Suppliers offer differentiated products.
  • There is no substitute for what the supplier provides.
  • The supplier group can credibly threaten to integrate forward.

Power of buyers is high where:

  • There are few buyers, or each buys in volumes large relative to a seller.
  • The industry's products are standardised or undifferentiated.
  • Buyers face few switching costs.
  • Buyers can credibly threaten to integrate backward.

Buyers are price sensitive where the product is a significant fraction of their cost structure, they earn low profits or are under cash pressure, the product has little effect on the quality of their own product, or it has little effect on their other costs. Note that price sensitivity is itself an outcome of these conditions; cite the conditions, not the sensitivity.

Threat of substitutes is high where the substitute offers an attractive price-performance trade-off relative to the industry's product and the buyer's cost of switching to it is low. Substitutes perform the same function by a different means and are often overlooked because they come from outside the industry. Treat a different way of buying the same product (a marketplace, a direct programme) as a substitute for the incumbent's role when it meets the customer's need without the incumbent's activity; treat it as an entrant when it performs the same activity by the same means.

Rivalry is intense where:

  • Competitors are numerous or roughly equal in size and power.
  • Industry growth is slow.
  • Exit barriers are high.
  • Rivals are highly committed to the business and have aspirations beyond economic performance.
  • Firms cannot read each other's signals well.

Rivalry is most damaging to profitability when it takes place on price, which is likely where products are nearly identical and switching costs low, fixed costs are high and marginal costs low, capacity must be expanded in large increments, or the product is perishable. Rivalry on dimensions other than price (features, service, delivery, brand) can expand value rather than destroy it, because different rivals serve different segments better.

Industry structure versus company position

Porter's framework is an account of the profit available to an average participant, not a verdict on any one firm. Two consequences follow for this skill. First, a strong-forces industry can contain a highly profitable firm (one positioned where the forces bear least) and a weak-forces industry can contain a failing one, so the forces never substitute for the client's own share-of-revenue and share-of-profit figures. Second, the consistency test in step 4 of the typical steps cuts both ways: if the more profitable participants are those better placed against the forces you have identified as binding, the analysis is probably right; if the most profitable firm in the industry is fully exposed to the force you called binding, the analysis has mislabelled the force or the industry.

Factors, not forces (2008)

Porter names four things commonly mistaken for forces and explains that each acts through the forces:

  • Industry growth rate. Fast growth does not guarantee profitability. It can strengthen suppliers, and combined with low entry barriers it draws in entrants; several fast-growing industries have been among the least profitable.
  • Technology and innovation. Advanced technology or rapid innovation is not by itself a source of industry attractiveness; some low-technology industries are far more profitable than high-technology ones. What matters is what the technology does to the forces.
  • Government. Government is not inherently good or bad for profitability; specific policies affect specific forces (patents raise entry barriers, procurement rules alter buyer power).
  • Complementary products and services. Complements affect demand but their effect on profitability depends on how they alter the forces (for example by raising or lowering entry barriers or switching costs).

Rate none of these as a force. Record each as a factor, and trace which force it moves and in which direction.

Typical steps in industry analysis (2008)

  1. Define the relevant industry: the products in it and the geographic scope of competition.
  2. Identify the participants and segment them into groups: buyers and buyer groups, suppliers and supplier groups, competitors, substitutes, potential entrants.
  3. Assess the underlying drivers of each competitive force to determine which forces are strong and which are weak, and why.
  4. Determine the overall industry structure and test the analysis for consistency: why is profitability what it is? Which forces control it? Is the analysis consistent with actual long-run profitability? Are more profitable players positioned more favourably in relation to the forces?
  5. Analyse recent and likely future changes in each force, positive and negative.
  6. Identify aspects of industry structure that might be influenced by competitors, by new entrants or by the company itself.

Step 4 is the one most often skipped. The consistency test (does the structure you have described actually explain the observed profitability?) is what turns a list into an analysis, and it is why SKILL.md's stage 5 checkpoint requires the forces to explain the pool map.

Common pitfalls (2008)

Porter lists these explicitly:

  1. Defining the industry too broadly or too narrowly.
  2. Making lists instead of engaging in rigorous analysis.
  3. Paying equal attention to all of the forces rather than digging deeply into the most important ones.
  4. Confusing effect (price sensitivity) with cause (buyer economics).
  5. Using static analysis that ignores industry trends.
  6. Confusing cyclical or transient changes with true structural changes.
  7. Using the framework to declare an industry attractive or unattractive rather than to guide strategic choices.

How the forces shape strategy (2008)

Porter gives three ways a company can respond to structure, and this skill uses them in stage 7 to generate options:

  • Positioning the company where the forces are weakest: choosing the segment, customer group or activity in which buyer power, rivalry and substitution bear least.
  • Exploiting industry change: anticipating a shift in a force and moving first.
  • Shaping industry structure: acting to weaken a force or redirect rivalry, such as raising switching costs, standardising in ways that reduce supplier power, or shifting competition from price to service.

He also warns that redefining the industry (finding a boundary within which the company faces a better structure) is a legitimate use of the analysis, provided the redefinition reflects real differences in the forces rather than a preference for a flattering picture.

3. Value-chain mapping (Porter, 1985)

Porter's "Competitive Advantage" (1985) decomposes a firm into primary activities (inbound logistics, operations, outbound logistics, marketing and sales, service) and support activities (firm infrastructure, human resource management, technology development, procurement), and the industry into a chain of such firms from supplier to buyer. For this skill the decomposition of interest is the industry-level chain, because the profit pool is plotted on it. Practical rules:

  • Decompose by activity, not by company. An activity is a step that adds cost and could in principle be performed by a different firm.
  • Include activities beside the main flow: financing, insurance, logistics, after-sales, data, and any consumable or accessory tied to the product.
  • Six to twelve activities is the workable range; fewer hides the deep pools, more cannot be populated with data.
  • For each activity record the performers (types of firm), the revenue basis at that stage (what is invoiced to whom) and whether the same firms perform adjacent activities.

4. Market sizing by triangulation

Triangulation is consulting craft rather than a published framework, but its logic is worth stating. Three estimation routes share no common source of error: top-down errors come from proportional allocation assumptions, bottom-up errors from unit counts and per-unit spend, comparable errors from the analogy. When the three agree, the agreement is evidence; when they disagree, the disagreement identifies the factor that is wrong, provided each route has been written out factor by factor.

Top-down. Start from a verified aggregate (national spend, an industry pool, an official statistic). Apply share factors for geography, segment, product, channel. Each factor is a proportional assumption unless evidenced; mark the ones that are least likely to be proportional (adoption of a new service is rarely proportional to spend).

Bottom-up. Count the units (customers, sites, households, machines). Multiply by adoption rate (share of units buying at all), spend per adopting unit, share of that spend within the definition, and yield (the provider's revenue per unit of spend under scope, where the sized market is a fee or margin rather than the whole spend). The client's own customer data usually supply spend per unit; check whether the client's customers are representative of non-customers.

Comparable-based. Find a market where the quantity of interest is observable (a peer region, a market further along the adoption curve, a listed company's segment) and apply its ratio. Name the structural point on which the analogy could fail.

Reconciliation. Write the three results side by side. For each pair, name the factor that explains the gap. Prefer the route whose factors are best graded, state the range spanned by the routes, and give a growth rate with its own evidence. Do not average; averaging discards the information in the spread. Present the sizing sheet with every factor, source and grade visible, so that a reader who disagrees with one factor can see the effect of changing it.

5. Competitor fact-base construction (Porter, 1980)

"Competitive Strategy" (1980), chapter 3, sets out four components of competitor analysis: future goals (what drives the competitor, at all management levels and in multiple dimensions), current strategy (how the competitor is competing now), assumptions (what the competitor believes about itself and about the industry), and capabilities (strengths and weaknesses). From these four a response profile is derived: the competitor's likely offensive moves, its defensive capability, and the moves by others to which it is most vulnerable or that would provoke the strongest retaliation.

The practical value of this ordering is that most competitor profiles stop at capabilities and current strategy (what the competitor is and does), which describes but does not predict. Goals and assumptions predict. A competitor that believes scale in logistics is the source of advantage will keep acquiring; one that believes local relationships hold accounts will keep opening branches; each will respond to the client's moves in ways consistent with those beliefs. Evidence for goals and assumptions comes from stated strategy in filings and investor material, patterns of capital allocation, the backgrounds of senior management, and the competitor's public explanations of its past moves.

For this skill the profile template in templates.md adds the quantitative fact base (size, growth, margin, footprint, customer base, cost position) with sources and grades, so that the profiles feed war-gaming and value-driver-modeling as well as strategic-choice.

6. How the methods fit together

Run in the order SKILL.md gives, each method supplies an input the next needs:

MethodAnswersFeeds
Value-chain mapWhat are the activities?The axes of the profit-pool map
Profit poolWhere is the profit, and where is it moving?The pools to size; the hypotheses the forces must explain
Market sizingHow large is the pool the decision concerns?The value at stake for value-driver-modeling
Five forcesWhy is each pool deep or shallow, and will it stay that way?The binding constraints on the client's margin; the option-generation logic
Competitor fact baseWho is positioned to capture the pools, and what will they do?war-gaming and strategic-choice

The consistency tests between methods are as valuable as the methods themselves: a pool that the forces cannot explain, a market size that does not fit inside its pool, a competitor whose margin is inconsistent with the activity it is placed in. Each such inconsistency is a finding, and usually the most useful one in the fact base.