Research, authorship, and accountability
Every PortersFiveForce.com article must answer a defined question with evidence a reader can inspect. We begin by naming the industry, offer, customer, geography, time period, and exclusions. We then look for the economic mechanisms behind rivalry, buyer power, supplier power, new entry, and substitutes. A source is included because it supports a claim, not because it makes the page look researched. Company statements are labeled as company statements. Estimates remain estimates. Our own judgments are separated from facts. Scores summarize the written case and never replace it. We do not publish a page simply because a keyword exists, and we do not stretch one company template across a long list of names. If the evidence cannot support a distinct analysis, the article is delayed, narrowed, or not published.
These standards protect the reader from three common problems: a vague market boundary, facts with no analytical purpose, and confident conclusions that the evidence cannot carry. They also give our editors a clear reason to reject a draft.
A material factual claim should lead to a source that supports the right entity, period, geography, definition, and unit.
Readers should be able to see where evidence ends, where interpretation begins, and which assumption could change the conclusion.
An article must state what it does not measure. Industry structure is not company quality, a forecast, or investment advice.
Good formatting cannot rescue weak research. A complete-looking table may still hide mismatched definitions. A precise score may still rest on guesswork. We treat those as editorial failures, even when the page reads smoothly.
Research starts with a boundary statement. It names the decision, product or service, customer group, geography, time period, and important exclusions. Without that boundary, the five forces can change meaning from one paragraph to the next.
We ask what the customer is actually choosing between. We separate the user, buyer, payer, and beneficiary when their incentives differ. We also test whether regulation, distance, language, distribution, or service requirements create smaller local markets inside a broad category.
| Field | Question | Why it matters |
|---|---|---|
| Offer | What product, service, or use case is being assessed? | Prevents unrelated products from entering the same analysis. |
| Customer | Who chooses, uses, pays, and can switch? | Clarifies buyer power and realistic substitutes. |
| Geography | Where do the same competitive conditions apply? | Stops national evidence from being used for a local market without support. |
| Period | Which date or planning horizon governs the evidence? | Prevents old structure from being presented as current. |
| Exclusions | Which adjacent offers are outside the boundary, and why? | Makes the analyst's judgment visible and challengeable. |
Our boundary checks are informed by the U.S. Department of Justice tools for market definition and the European Commission’s market definition notice. These legal sources do not set our editorial conclusions. They help us ask disciplined product and geography questions.
Source quality depends on proximity, accountability, scope, and possible bias. We work outward from primary records. No source is accepted only because it ranks well, appears often, or agrees with the conclusion we expected.
| Priority | Typical sources | How we use them |
|---|---|---|
| Primary public record | Regulators, courts, government agencies, official statistics, public tenders | Rules, market conduct, definitions, volumes, concentration, entry, and observed events |
| Direct company record | Filings, annual reports, investor materials, contracts, official price pages | Reported operations, management claims, product terms, and company-specific facts |
| Independent analysis | Peer-reviewed research, credible financial press, established trade bodies | Context, conflicts, expert interpretation, and gaps not covered by primary records |
| Discovery only | Search snippets, directories, summaries, vendor posts, existing Five Forces pages | Leads to investigate. These do not carry a material conclusion by themselves. |
A primary source is not automatically neutral. A filing is useful for what a company reports, but it may frame risk in management’s preferred language. A regulator may focus on a legal question narrower than ours. We compare definitions and incentives before deciding how much weight a source carries.
Every final link must open and support the nearby claim. We reject a source when its entity, period, geography, unit, or definition does not match. A famous publication with the wrong scope is still the wrong source.
For the framework itself, we use Harvard Business School’s explanation of the Five Forces and Michael Porter’s 2008 restatement. Industry facts come from sources specific to the market being studied.
Each material claim is classified before it reaches the page. A reported fact, company claim, estimate, calculation, and editorial interpretation are not interchangeable. The label affects both the wording and the confidence we assign.
What is a fact?
A directly supported event, rule, amount, count, or condition with a clear source and scope.
What is an estimate?
A value produced by a source or transparent calculation. Its assumptions and period must remain visible.
What is our judgment?
An explanation of how evidence changes bargaining power, entry, substitution, rivalry, cost, price, or risk.
We keep a working evidence ledger for the claims that carry the conclusion. It records the claim, type, source, scope, period, force, conflict, and confidence. Sources that fail the scope test are recorded as rejected so they do not return later through a different draft.
The chain must be understandable without the score. Evidence shows a driver. The driver changes behavior, price, cost, demand, investment, or access. That change supports a force conclusion. If a link is missing, we have an interesting fact, not a finished analysis.
The evidence approach also draws on the DOJ’s discussion of documents, switching, win-loss records, customer behavior, and past entry in its guidance on evaluating competition.
A force score summarizes pressure on industry profit. It does not rate a company. We use a five-point scale only after the written explanation is complete. One means weak pressure. Five means very strong pressure.
A score must name the decisive mechanism and the counterevidence. Buyer concentration alone does not prove buyer power. A large competitor count does not prove intense rivalry. High capital needs do not prove that entry is impossible. We look for structure and behavior together.
What does confidence mean?
High confidence needs current, direct, well-matched evidence. Medium confidence leaves one important gap. Low confidence means another interpretation remains credible.
What does the index mean?
Our optional 0 to 100 Structural Attractiveness Index converts five pressure scores into a comparison aid. It is our editorial tool, not a formula created by Michael Porter.
We do not use decimal force scores. Decimal precision suggests a level of measurement the evidence rarely supports. If two ratings remain plausible, we explain what information would move the score instead of hiding the uncertainty.
AI-assisted tools may help organize research, compare coverage, check a calculation, find repetition, or test whether a draft answers its question. They do not become the author and they do not make the final analytical judgment.
Every material source must still be opened. Every factual claim must still be checked against that source. Scores, market boundaries, uncertainty, and final wording require human review. A generated citation is not evidence. A fluent paragraph is not proof.
Acceptable assistance
Research organization, structured drafting, coverage checks, table structure, calculation checks, editing options, link checks, and consistency review.
Not acceptable for publication
Invented facts, fabricated citations, copied analysis, raw generated drafts, unsupported scores, or pages produced only by changing a company name.
Oleksii Ryzhkov reviews the final scope, sources, conclusions, and copy before publication. If a claim cannot survive that review, it is removed or rewritten with the uncertainty made clear.
We do not publish a page merely to occupy a search result. Volume is not an editorial goal. A missing article is better than a confident-looking article built from weak evidence or a repeated template.
- No renamed templates. Each page needs its own market boundary, source set, causal argument, counterevidence, and conclusion.
- No unsupported market shares. A number must match the industry, period, geography, definition, and denominator used in the text.
- No competitor summaries as research. Existing strategy articles may reveal a lead, but they do not become our evidence base.
- No automatic conclusions. Growth, concentration, regulation, or technology does not determine a force without a mechanism.
- No hidden uncertainty. Missing evidence lowers confidence or removes the score. It is not filled with generic language.
- No favorable treatment for sale. A company cannot purchase a positive force rating or approve an independent editorial conclusion.
This rule makes publication slower. That is intentional. The aim is not to look complete. The aim is to make a conclusion that a skeptical reader can inspect and challenge.
We correct factual errors when we find them. We revisit an analysis when a structural event could change its conclusion. A new calendar year alone is not proof that an article has been reviewed.
Useful update triggers include a merger, meaningful entry or exit, regulation, a change in distribution, a new substitute, a capacity shift, or a supplier constraint. A score changes only when the mechanism changes. Link repairs and small factual corrections may leave the conclusion intact.
Readers can report an error through our contact page. A useful correction includes the page, disputed claim, reason, and a direct source when available. We review the evidence, not the tone of the message.
PortersFiveForce.com sells downloadable strategy resources. That commercial activity supports the site, but it does not give a company, source, or article subject the right to alter an editorial conclusion.
If a page contains a sponsorship, affiliate relationship, or another material commercial connection, it should be labeled on that page. A disclosure does not repair weak research. The same source and evidence standards still apply.
Editorial articles are educational. They are not investment, legal, tax, or management advice. Readers should verify time-sensitive facts and apply the analysis to their own decision, constraints, and market.
Editorial responsibility must belong to a person, not a tool or a brand name. Oleksii Ryzhkov is the author and editorial lead for this research program. He approves what is published and what is left out.
Oleksii defines the research question, reviews the evidence trail, challenges the force ratings, and takes responsibility for the final editorial judgment on PortersFiveForce.com.
View Oleksii Ryzhkov on LinkedInThese standards are a working commitment. If our process changes in a material way, this page should change with it. The date should move only after the text and the underlying process have both been reviewed.
For the purpose and people behind the work, read About PortersFiveForce.com.