How to Measure the Five Forces: 25 Indicators Every Analyst Should Check

A useful Five Forces analysis should show more than five labels and five scores. It should show what was measured, why it matters, and how the evidence changed the conclusion. This guide sets out 25 indicators that we check before rating an industry. There are five indicators for each force. They are diagnostic signals, not a formula. A high concentration ratio may matter greatly in one market and very little in another. A low switching cost may exist on paper while buyers still refuse to move. The analyst's job is to connect a measured fact to a competitive mechanism. The score comes after that reasoning, not before it.

The framework follows Michael E. Porter's Five Forces article and Harvard Business School's official framework overview.

The indicators help an analyst ask better questions. They do not remove judgment. Each one tests a reason that profit may be competed away, negotiated away, supplied away, entered away, or replaced by another solution.

Some indicators are numbers. Market shares, customer concentration, capacity use, and entry costs can often be estimated. Others require observed behavior. Discount approvals, failed supplier changes, lost tenders, and customer migration can reveal more than a clean spreadsheet.

Our rule is simple. Never record a number without recording its mechanism. If customer A represents 18 percent of a supplier's sales, what can that customer obtain because of its size? A lower price? Longer payment terms? Custom service? The share alone is not buyer power. The concession, or a credible threat that produces one, is the evidence that matters.

Several indicators can point in opposite directions. That is normal. A concentrated industry may have gentle rivalry if firms are differentiated and capacity is tight. A fragmented industry may have brutal rivalry if offers look identical and demand is shrinking. The tension belongs in the analysis. It should not be averaged out without explanation.

Define the market before collecting indicators. State the offer, customer group, geography, and time period. These choices determine the denominator in a market share, the buyers who count, and the alternatives that qualify as substitutes.

Suppose an analyst measures the concentration of the global software market, then uses the result to rate rivalry among cloud accounting tools for small UK businesses. The number may be correct. It is still useless for that question. Its product, customer, and geographic boundaries do not match.

The boundary is a testable hypothesis. It may change when evidence shows that customers switch more widely than expected. If it changes, recalculate earlier measures. Do not keep a convenient concentration figure from the old boundary.

The U.S. Department of Justice explains why shares depend on sound market definition. The European Commission's 2024 notice also describes product and geographic market tests.

Rivalry is the pressure created by existing competitors. Count firms, but do not stop there. The sharper question is how they compete and whether the structure keeps that pressure alive.

Two calculations are useful here. CR4 is the combined share of the four largest firms. HHI is the sum of every firm's squared market share. If four firms each hold 25 percent, the HHI is 2,500. Both measures describe structure. Neither proves how firms behave.

Legal concentration screens are not Five Forces scoring thresholds. The DOJ uses HHI in merger analysis for a specific legal purpose. Copying those thresholds into an industry score would confuse two different tasks. Choose a measure that reflects the market, then test it against prices, capacity moves, customer losses, and internal selling behavior.

The DOJ's official tools explain the use and limits of market shares and concentration, including the need to reflect market realities.

Buyer power is not the same as customer importance. A large customer has power when it can credibly reduce, delay, redirect, or internalize its purchase. The result should be visible in terms, price, service, or risk transfer.

Customer concentration should be calculated at the level being analyzed. A supplier may serve thousands of end users but negotiate with three national distributors. In that case, distributor concentration may matter more than the number of end users.

Switching cost also needs proof. Contract length is only one part. Data migration, staff habits, quality approval, compatible equipment, and fear of failure can keep a customer in place after a contract expires. Survey intent is weaker than a completed switch.

Purchase importance cuts both ways. Buyers bargain hard over a large, standard cost item. They may bargain less over a small input that could stop production or damage their brand. State which effect dominates and show why.

Useful behavioral evidence can include switching records, win and loss reports, discount approvals, customer surveys, and testimony. These examples appear in the DOJ's guidance on evaluating competition.

Supplier power asks whether providers of an input can take value from the industry. Price increases are the obvious sign. Allocation rules, weaker service, tighter payment terms, lower quality, and forced redesign can matter just as much.

Use qualified supply, not the total number of companies in a directory. Ten firms may make a component. Only two may meet the required tolerance, delivery location, volume, and certification. The effective supplier market has two firms until another one qualifies.

Capacity data need a time dimension. A six-month shortage may create strong negotiating power now but little structural power if new capacity opens next year. Record announced projects, cancellation risk, and the time needed to reach stable output.

Forward integration should be credible. A supplier needs access to customers, sales capability, service capacity, and an economic reason to move. A vague management comment is not enough. Look for direct sales, channel conflict, hiring, product launches, or investment.

Company filings and risk disclosures can help identify dependency and supply constraints. Public filings are available through SEC EDGAR. Claims still need comparison with contracts, capacity data, and buyer behavior.

Entry is a process, not a registration count. Forming a company or launching a website is not meaningful entry if the firm cannot win customers at a viable cost. Measure the path from intent to competitive scale.

Past entrants are valuable tests. Record how long they took to win meaningful demand, how much capital they used, and whether they stayed. Failed entry can expose a barrier that looks small in a desk study. Successful entry can show that an old barrier has weakened.

The threat can be strong even before a new firm arrives. Incumbents may hold prices down because entry is credible. The relevant question is whether potential entrants have the capability, incentive, and path to affect competition within the chosen period.

Scale should be measured at the activity that drives cost. It might be a factory, a route network, a customer data pool, a local service team, or a brand campaign. A national market share can hide a large minimum scale in one region.

The DOJ separates an entrant's capability, incentive, and likely competitive impact. The OECD's Competition Assessment Toolkit discusses scale, sunk cost, and other entry barriers.

A substitute solves the same customer problem in a different way. It may sit outside the analyst's industry classification. That is exactly why it is easy to miss.

Start with the job, not a product list. Video meetings can substitute for some business travel. Repair can substitute for replacement. Renting can substitute for ownership. None belongs to the same narrow product category, yet each may cap demand or price.

Relative price needs a full cost view. Include setup, maintenance, downtime, training, disposal, and risk where they matter. A cheaper sticker price may produce a more expensive customer outcome.

Observed response is stronger than resemblance. Search data, customer interviews, transaction changes, lost-sale notes, and natural experiments can show whether demand actually moves. If no movement is visible, explain whether the substitute is weak or the evidence is still too thin.

Begin with an evidence sheet, not prose. Give every observation a source, publication date, measurement period, geography, unit, denominator, and force. Mark whether it is a fact, estimate, company claim, or analyst judgment.

Normalize only when the comparison remains meaningful. Currency can be converted. Reporting periods can sometimes be aligned. Product definitions cannot be repaired with arithmetic. If two sources count different things, keep them separate.

Use three comparison points when possible. A current measure shows the present. A time series shows direction. A relevant benchmark shows whether the level is unusual. One year may reflect a shock. One peer may have a different business model.

Industry codes are useful starting points, not final market definitions. Official datasets may group businesses that customers do not see as alternatives. They may also split businesses that compete directly.

Official baseline data can come from the U.S. Census Bureau's Economic Census or Eurostat's structural business statistics. Each figure still needs a scope check.

Do not give every indicator one point and calculate an average. The five indicators are not independent or equally important. A binding regulation may dominate the entry threat. A severe input shortage may dominate supplier power. Weight comes from the mechanism and evidence, not from a fixed spreadsheet cell.

Write the force conclusion first. State the direction, the two or three decisive drivers, the strongest counterevidence, and the expected time period. Then assign a pressure score from 1 to 5.

A force can score 5 with low confidence. That means the available evidence points to strong pressure, but the conclusion is fragile. A force can score 2 with high confidence. That means weak pressure is well supported. Confidence is not a reward for an attractive answer.

When evidence is mixed, use 3 and explain the split. Do not invent decimals to hide uncertainty. If a force cannot be scored, say so. If two or more forces have low confidence, we do not publish a combined industry attractiveness index.

If all five forces are adequately supported, our optional index uses this formula: 100 times the quantity of 5 minus the average force score, divided by 4. It reverses the pressure scale. A higher index means lower combined structural pressure. The index is a summary, not a forecast or investment recommendation.

One fact can affect several forces. A proprietary technical standard may raise entry barriers, increase switching cost, and strengthen a supplier. Reusing the fact is allowed. Counting the same economic effect three times is not.

Trace each use to a different mechanism. If the standard delays new entrants, record the delay under entry. If it makes customer migration costly, record the migration burden under buyer power. If one supplier controls the standard, record its licensing leverage under supplier power. State the interaction when the mechanisms reinforce one another.

A clean number can still be wrong. Concentration estimates often depend on missing private-company revenue. Switching estimates often ignore failed migrations. Announced capacity may be delayed or cancelled. Put the weakness next to the finding, not in a footnote that nobody reads.

Markets also contain segments. Large customers may have strong buyer power while small customers have little. Urban rivalry may differ from rural rivalry. A single score is acceptable only when the article explains the variation that matters.

Research on industry analysis warns that average industry measures can hide large differences among firms and segments. See the American Economic Association's review of industry heterogeneity.

A finished sheet should let another analyst reproduce the logic. It does not need to contain every fact gathered. It does need the decisive facts, their sources, their limits, and the reasoning that connects them to each force.

The final test is plain language. Can a reader answer four questions? What was measured? Why does it affect profit pressure? How strong is the evidence? What could change the conclusion? If any answer is missing, the score is not ready.

These 25 indicators are a starting discipline. They make shallow analysis harder. They also make disagreement more useful. A reader can challenge the boundary, the evidence, or the mechanism instead of arguing over an unexplained number.

Our wider scoring and source rules are documented in the site's editorial methodology. Google also recommends clear authorship, sourcing, and production context in its people-first content guidance.