The More You Analyse Competitors, the More Your Position Is Fixed in Their Coordinate System

Every instrument for analysing competitors demands the same thing. A list of the parties to be treated as competitors, prepared by you.

3C, SWOT, Five Forces, VRIO, the value chain. They arrange things differently, but a list sits at the input of all of them. Since it is an input, the output of the analysis is a function of the list. Change the input and the conclusion changes; leave the input alone and no amount of rigour changes the direction of the conclusion. And how the list is built is written down in none of the frameworks.

What can be settled here, and what cannot, is worth separating first. Three questions can be settled here. What the frameworks take as given. Where the axes of comparison are supplied from. And where the material for replacing an axis is held. All three follow from the nature of the instruments and the nature of comparison itself, so they hold their shape in any field, and an outsider can set them down.

Some things cannot be settled. Who your real competitors are, what you should stop doing now, how many records are enough before you judge. These cannot be written. The material for that judgement sits not in this article but in the jobs you were given and the ones you lost. The line between what is writable and what is not runs there. And ignoring that line to write your real competitors are of this kind would produce one more list handed to you from outside.

Whether damage is being done shows in where decisions start. If what you do next is decided by what a rival started, the coordinate system sits on their side. Every move is then a move behind, and being behind is invisible from the outside, because the analysis table is always full.

The cost side carries the same shape. Retaking your position each time a rival moves is work without an end. Catch up on one axis and the next axis appears. Every catch-up costs something, and the fact of having caught up accumulates into nothing. The following year starts from the same position.

What an instrument produces is your position on an axis of comparison. The axis itself gets placed outside the instrument, by somebody. As long as that somebody is a rival, remeasuring the position never moves the starting point. What remains at the end is not a list of competitors but the ability to say where your list comes from.

What the Five Competitor-Analysis Frameworks All Demand as Input

The five frameworks receive the range of comparison as an input and do not handle whether that range is sound.

3C analysis lays out market, competitors and your own firm, and looks for the position to take. SWOT sorts information into strengths, weaknesses, opportunities and threats. Value chain analysis breaks down the steps where value is produced and looks for the step where an advantage sits. VRIO judges whether a resource meets four conditions: value, rarity, inimitability and organisation.

Five Forces reads the profitability of an industry from five forces: the threat of new entrants, the threat of substitutes, the bargaining power of buyers, the bargaining power of suppliers, and rivalry among existing firms. The management scholar Michael E. Porter set it out in his 1980 book Competitive Strategy.

All five function in practice. Information gets structured and a basis for discussion appears. That benefit is real.

What goes unasked is fixed for each instrument. 3C does not ask what to write in the competitor column. SWOT does not ask whom to write under threats. The value chain does not ask what range counts as one chain. VRIO’s rarity does not ask rare among what. Five Forces does not ask the outline of the industry the five forces operate in.

This is not a defect in the instruments. Deciding a range belongs to design rather than to analysis. Five Forces was presented to analyse industry structure in the first place, and an instrument whose object is an industry has no reason to ask where the industry ends. The instruments work correctly within their own scope.

Nowhere do the instruments state that division of labour. Fill in a framework and the fact of having filled it in produces a feeling of having considered the matter. Once the blanks disappear, it looks as though there is nothing left to think about. Yet the range itself, the variable that decides the conclusion, sits outside the boxes from the start.

The mismatch comes from the side that carries the instrument in. In a business run by one person or a few, industry often fails to match the reality of the work. The structure in which large firms divide a market with rivals differs from the structure in which one person faces particular buyers directly. In the first, the list of rivals is more or less fixed; in the second, it changes buyer by buyer.

And a table leaves the person who filled it in with the feeling of having reached a judgement. Once the table is complete, a conclusion appears to have been drawn. What comes out of a table is the tidied form of its input. Tidying is not judging. However much judgement feels finished, that much chance to question the range is lost.

The Line Called an Industry Is Drawn by Participants Watching Each Other

The range of comparison usually arrives under the word industry. And that line is not a natural object.

A classification set by government to gather statistics, a boundary set by a trade body to define membership, or the set of firms a search result lined up. Each is a line drawn by someone for some purpose. Different purposes put the line in different places. Statistical classifications are drawn for the convenience of the survey; trade body boundaries are drawn around the unit that is admitted to membership. None of them expresses what that business’s buyers actually compare it with.

There is research that looked inside an industry to see how the line is made. The management scholars Joseph Porac, Howard Thomas and Charles Baden-Fuller studied the Scottish knitwear industry and examined whom the owners treated as competitors (Porac, Thomas & Baden-Fuller, 1989, Journal of Management Studies, 26(4), 397–416). What they observed was a process in which owners watched one another, grouped those who looked alike as rivals, shared that grouping across the group, and strengthened it by sharing it. The line was not given from outside; it was made on the side of the participants.

The object studied was one industry in one region, not a statistically generalised law. What is borrowed here is only the description of a mechanism: the outline of an industry can arise from the mutual perception of its participants.

Parties being considered by the same buyer at the same time do exist. One reading asks what is wrong with calling those parties competitors. The question is how you come to know who they are.

In most cases, the list of competitors is built from those you take to be in the same trade. People with the same job title, people in the same association, people appearing above you under the same search term. This is your classification, not the buyer’s. What buyers actually set side by side often crosses trades. Outsource it, hire someone, buy a tool, or do nothing at all. These belong to no single industry, and they compete for the same budget.

Where regulation fixes the range, the industry line carries real weight. For work that requires a licence, the set of licence holders is the range of comparison. The dividing line is whether that line actually restricts the buyer’s options. If it does, the range is real; if not, it is only your own classification.

Search results are another reason the line looks real. Enter a term and similar job titles appear in a row. But a search result is an answer to the person who typed that term, not the whole of a buyer’s consideration. The same buyer searches another term on another day and sees an entirely different set. The rows differ term by term and merge inside the buyer’s head.

Something is lost here. Treat only the parties in the search result as competitors and the layer that does not search that term — the layer that does not yet know the method exists — drops out of the comparison set and out of the design of delivery at the same time. The range of the analysis becomes the range of the business.

The More Competitor Analysis You Do, the More Your Position Is Fixed in Their Coordinate System

Using the instruments does not move the judgement of the user. It moves the place where that judgement sits.

The output of competitor analysis is a set of axes. Price, delivery time, scope, number of cases, specialisation. Firms are laid out along those axes and you see where you win and where you lose.

Where do the axes come from? Mostly from what competitors are already advertising. A rival displaying three hundred clients makes case count an axis; one displaying three days at the fastest makes delivery time an axis. The axes of comparison are the ones competitors prepared. And analysis is the work of measuring your position along them.

After measuring, there are two moves: fill a weak axis, or extend a strong one. In both cases, the axes stay where the rival put them.

Keep measuring and your moves become subordinate to theirs. A rival begins a new claim and one axis is added; you measure again along the new axis and decide whether to respond. Through that repetition, the starting point of every decision sits on their side.

It shows in how time is spent as well. Chasing axes never ends, because axes keep multiplying. And while you chase them, the work of putting into words what your business treats as important goes to the back. Nobody is inconvenienced by that, so it stays at the back indefinitely. The same structure appears in Thought patterns are not something you fix. A form you received stays invisible for as long as it remains a form.

The axes rivals advertise are advertised because buyers care about them — so adopting them should be rational.

Part of this is right. Anyone advertising an entirely irrelevant axis leaves the market, so the surviving axes reflect some buyer interest.

The mistake is to take the advertised axes for the whole of that interest. Advertising gathers around what is easy to compare. Things that can be counted, stated in a period, listed. What is hard to compare goes unadvertised even when it matters, because advertising it puts you at a disadvantage in comparison.

So the set of axes in the market is not the set of buyer interests but the set of interests shaped into a comparable form. What is lost during the shaping stays lost. And among the lost items sit elements that decide how a price is judged.

Differentiation Is Movement Along an Axis

Differentiate to avoid competing on price sounds right. In many situations it works. Place the operation precisely, though, and differentiation is taking a different position along an axis that already exists.

If the axis is scope, you differentiate by narrowing or widening it. If the axis is specialisation, you differentiate by restricting the field. Both are movements along an axis.

Being a movement, it brings two properties.

The first is that comparison continues. Different position or not, while you stand on the same axis you are laid side by side. Laid side by side, the final judgement becomes a total score. And when total scores are close, the deciding factor drops to price. Differentiation leaves price as the final judge for this reason.

The second is that you get caught up with. A position along an axis can be moved into by others. Where a position proves effective, people arrive later. The lower the barrier to entry, the faster this happens.

Differentiation that cannot be caught up with does exist. It exists where imitation requires long time or heavy investment. Equipment, accumulated data, legal protection. What VRIO treats as inimitability sits exactly here.

However, a business run by one person or a few finds that kind of barrier hard to build. It is not a field where the scale of capital investment creates the gap. So the differentiated position stays inside the range that can, in principle, be copied. The interval before being caught up with is itself the source of profit. This is not an explanation of why you cannot earn; it is a description of a constraint to design around. A plan built on never being caught up with breaks the moment you are.

And a differentiated position is announced to the outside the instant it works. Since you are advertising it, it is visible, and what is visible can be copied. Finding an effective position and publishing that position are both contained in the act.

So a design built on differentiation requires you to keep looking for the next position. While you keep looking, the centre of the business sits on where is currently vacant rather than on what we treat as important.

Looking for vacant positions is not in itself a bad act. Finding where demand exceeds supply is basic work. The danger is when it becomes the only principle of design. Vacancies fill. Move each time one fills and nothing accumulates. A business with no accumulation competes on the same terms in year ten as in year one.

Replacing an Axis Means Answering Three Questions

If movement along an axis is differentiation, changing the axis itself is changing the unit of comparison.

A unit is what counts as one. Choosing a unit is the same act as choosing who you stand beside (Do not begin pricing from cost). Changing it consists of three questions.

The first question. What is the buyer buying this for? Not the surface request but what changes as a result of having it. When the answer to this changes, the comparisons change.

The second question. For that purpose, what other means is the buyer considering? Not other firms in the trade. Other ways of meeting the same purpose. Hire someone, buy a tool, do it internally, do nothing. What appears here is the real comparison set.

The third question. Within that comparison set, what condition do only you meet? It resembles the question of differentiation, but the comparison set differs, so the answer differs. The strength you hold against your trade and the strength you hold against other means usually do not match.

The three questions have no framework. There are no boxes to fill. Tidying the information you already hold does not produce the answers. They come from records of what the buyer actually considered.

The second question does not come up as long as you look only at your own trade. Other means of meeting the same purpose usually swing far cheaper or far dearer. Doing it yourself looks like zero monetary outlay. Hiring someone adds a digit. The swing is far larger than the few tens of thousands that separate you from your trade.

Once that shape is visible, the meaning of the price changes. Being twenty per cent dearer inside your trade is close to noise inside a comparison with other means. When what the buyer is really weighing is commission it or do it ourselves, a twenty per cent gap decides nothing.

A limit attaches to the third question as well. What you write there is a condition, not a strength. A strength against your trade can be written as relative merit. What can be written against other means is the condition only you satisfy. And having written the conditions out, the conclusion is sometimes that another means is the sound choice. The correct move there is not persuasion but a change of destination.

The comparison set differs by the stage at which contact happens. Where the buyer arrives having already decided to commission the work, the comparisons revert to your trade. If your enquiries come only after that decision, comparison with the trade cannot be escaped. Escaping it requires contact before the decision. Here the question of the axis becomes continuous with the question of delivery.

What You Stop in Competitor Analysis Is Not Gathering Information but Receiving the Axis

One misreading needs setting aside. This is not a case for ignoring competitors. There are two things to stop.

The first is receiving the axes of comparison from competitors. What a rival advertises is useful information. Whether to adopt it as your own criterion is a separate judgement. Seeing something as information and using it as a criterion are different acts.

The second is putting the list of competitors at the start of the analysis. The start belongs on the side of what buyers considered. The list of competitors comes out as a result. Reverse the order and the list decides the conclusion.

There are also things not to stop. Do not stop tracking the going rate. Without knowing the reference point a buyer carries in, you cannot predict what happens when you name a figure. Do not stop observing what rivals advertise, either. Without knowing what language a buyer is bathed in, you cannot know how your own language will land.

In the early period, competitor analysis is effectively the only source of information. There are no buyer records yet. Told to begin from what buyers considered, you would have no material.

What to take here is not a ban on analysis but a label saying provisional. Begin with the axis borrowed from competitors, and as transactions occur, record what buyers actually compared you with and replace the borrowed axis. Whether the replacement ever happens depends on whether you are aware that it is borrowed.

The trouble sits here. A borrowed axis comes to feel like your own as you use it. An explanation repeated many times becomes a natural claim to the person repeating it. And a chance to doubt an axis that feels like your own almost never arrives. The trace of an axis having belonged to someone else disappears at the moment it is adopted. Written into a box in a table, it becomes our strength, and the origin is gone.

So the replacement runs on procedure rather than on feel. List the axes you use now and write beside each one where it came from. From a buyer record, from a rival’s claim, or from your own impression. Mixed together they cannot be told apart; write the origin and they separate. Once separated, what should be replaced becomes visible.

Stop watching rivals and you will be left behind — this concern has two parts. Being left behind is real where the criteria of the field are updated externally. Regulation changes, required specifications change, tools are replaced. You will not know if you stop observing. A change in what a rival advertises is not an update to the criteria. An advertisement is a record of that rival retaking their own position. The first is worth following; the second is not. Bundle the two into one industry news and the reason for chasing the second is supplied indefinitely by the first.

What to Keep as a Record of the Buyer

The material for replacing an axis exists only inside conversations with buyers. It will not arrive on its own, though. Three kinds are worth keeping.

What they tried before enquiring. Before the request arrived, the buyer was doing something. Tried it themselves, asked someone else, bought a tool, left it alone. The means that come up are, as they stand, what the buyer was comparing.

What else they considered while deciding. Not merely compared with other firms, but who those firms were and where the hesitation sat. The point of hesitation is the axis. And it often fails to match the axis you assumed.

What decided it, when you were turned down. Of the three, this record is the hard one to obtain, and it is also where the deciding factor gets stated. The first reason offered for a refusal is usually price.

Price stands in because it is the reason that causes no friction. Saying it did not fit or some doubt remained costs more than saying the budget. A word that is safe for the person refusing carries little information for the person refused.

Start keeping records and for a while it feels as though your material has shrunk. Competitor analysis fills a table in a day; buyer records accumulate only as fast as transactions do. What fills up quickly looks more correct.

And not every record is trustworthy. The reasons buyers give are shaped into something explainable on the spot. Where the real decider was hard to put into words, an easier reason appears instead. Price is the leading example.

Even so, records have a property competitor analysis lacks. Being wrong eventually shows. A list of competitors that is wrong leaves no trace of being wrong. A buyer record that contradicts your assumption stays as a contradiction.

Keeping them requires a timing. All three change shape if you recall them after the work is over. One line straight after receiving the request, one line straight after being turned down. That is enough. Aim for a tidy record and the effort of tidying arrives first, and the practice stops.

Asking is also genuinely awkward, especially for the third. Move the position of the question, though, and the burden falls. Instead of asking for reasons in reply to a refusal, ask during the proposal: what other approaches are you weighing? At that moment the buyer is still deliberating, so answering helps them order their own thinking. Part of what you would learn after a refusal can be obtained before it.

Read Records as a Distribution, Not One by One

Records become usable when they are read as a body rather than one at a time.

One person saying it was expensive tells you nothing. Of twenty cases, how many named another means and how many named your trade — that ratio maps the position of the axis accurately. A single reason depends on circumstance; a distribution depends on structure.

Two readings are available. One is the composition of the comparisons. Sort the parties named into same trade and other means and count. The higher the share of other means, the further you stand outside your trade’s coordinate system. The other is the composition of the hesitations. Is price the only thing named, or do scope and approach appear? More of the latter means the axis of judgement has moved away from price.

Reading a distribution requires a count of cases. This is one weakness of the method. Competitor analysis finishes in a day; a distribution of records takes as long as the transactions take. It loses on speed. Whether you adopt it anyway depends on how many years the business is built to run.

And an accumulated record sometimes reveals what the earlier analysis was measuring. The case is when a party you had treated as a competitor turns out never to have been named as a comparison. At that point, what the analysis had been measuring becomes clear.

Where the count is still small, ratios cannot be read. What can be read then is the change in composition. Until last month only your trade was named; this month doing it ourselves appeared. With three cases, that change still carries information. A ratio shows a level; a change in composition shows a direction. The direction is readable earlier.

There is a constraint. A distribution can only be taken from the buyers who are already arriving. What the people who never arrive compare you with never appears. Design from records alone and the outline of your current audience is fixed in place. Records are material for replacing a coordinate system; they are not material for widening an audience. The two are separate jobs.

And records carry one route to misreading. Where a distribution is skewed, whether the skew belongs to the market or to your own delivery is not decided by the records. Only your trade being named may mean that buyers consider only your trade, or that you are only found by people who have finished considering it. Separating those two requires looking at the route by which you are found, not at the records.

A Changed Axis Shows in the Buyer’s Language and in How They Refuse

That an axis has been replaced does not show up in revenue straight away. It appears in the words buyers say.

The first is the vocabulary inside enquiries. Whether the words you use have begun to appear in the text of enquiries. If they have, buyers are receiving your classification. If not, and only the general terms of the trade appear, you are still inside the existing frame.

The second is what they declare comparing. Whether the answer to where else did you look includes anything outside your trade. If only your trade appears, the coordinate system has not moved.

Three is how they refuse. On the same axis, refusal takes the form somewhere cheaper was available. With the axis moved, refusal takes the form we decided to do it ourselves this time or we will revisit the timing. The latter is evidence that the comparison moved from your trade to another means.

All three measure what the buyer said rather than what you did. Changes on your own side — new messaging, rebuilt materials — are records of your workload, not records of arrival. Use workload as an indicator and you look busiest exactly when nothing is moving.

And all three lag. Between changing your language and seeing it in the text of an enquiry, time passes. During that time the indicators say nothing. So decide the interval of judgement in advance. Measure without deciding how many months form one block and the reading wobbles each time, and moves change in response to the wobble. A pattern of moves changing on a short cycle is indistinguishable, from outside, from having no axis at all.

How many records make a distribution, how many months until the vocabulary shifts. Give those two a number and the number becomes the next borrowed thing. The form of measuring your position with a borrowed ruler would stand up again on the side of the indicators.

A competitor-analysis framework demands a list, the list decides the conclusion, and the conclusion decides the move. And the work of building the list sits inside none of the instruments. So stopping competitor analysis does not mean discarding the instruments; it means moving the work of building the list to the outside of them. The instruments work accurately once it has been moved. With the range coming from your own judgement, looking at a table to see where you are weak inside that range regains its meaning.

In the end, one procedure is available today. Write out, in a single column, the axes you currently use to explain yourself. Beside each line, write where it came from. Did a buyer say it, did a rival display it, or did you simply feel it. While the three are mixed, which of them to doubt cannot be decided either. Once separated, the party to check with next is the buyer.

Where to begin when you build the axis from your own side — the question of what to put first — is handled in How to make a concept. How the axes are made and maintained outside any industry is taken apart in How social common sense is made.

Who holds the coordinate system is one face of the design question of where the source of value sits. The overview from the design side is in Stepping down from price competition.