Competitor analysis: what to copy and what to ignore

The usual version of this exercise produces a document nobody reopens: a list of rival businesses, their websites, their social media follower counts, and a vague sense that one of them is doing well. It fails because none of it connects to a decision. A follower count does not tell you whether to change your price, and knowing a competitor posts more often does not tell you whether posting more would sell anything.

Useful competitor analysis is built backwards from the decisions you actually face: what to charge, who to serve, what to promise, and where to be different. That shortens the list of things worth recording to a handful — the price a customer actually pays, what is included at that price, which kind of customer the business is built around, what it promises about delivery and returns, and what it visibly does not do. Everything else is context you can ignore without loss.

Draw price on one axis and quality on the other, where quality means whatever your buyers actually judge — durability, finish, speed, expertise, reliability, service. Place each competitor where their customers would place them, not where their marketing does. Include yourself, honestly. Two axes are enough because they capture the trade-off every buyer makes, and a two-axis picture can be read in a glance in a way a spreadsheet of attributes cannot.

What emerges is usually a cluster, because businesses copy each other, and empty regions around it. The cheap-and-basic corner is normally crowded, and so is the mid-market, where everyone who could not decide has ended up. Genuinely empty space tends to sit in the expensive-and-excellent corner, which requires capability you may not have, and in a spot that is harder to name: an ordinary price with one dimension executed unusually well. Before you get excited about a gap, ask the next question.

Some gaps are empty because nobody wants what sits there. Cheap-and-excellent is empty because the arithmetic does not work, not because nobody thought of it. A gap can also be empty because reaching the customers in it costs more than they are worth, because serving them requires a licence or capability that is genuinely hard to acquire, or because someone tried, failed, and left no trace you can see. The absence of a competitor is weak evidence about the presence of demand.

The check is to look for the customers rather than the gap. Can you name actual people or businesses who want this and are currently either not buying, buying something ill-fitting, or complaining about what they bought? Complaints in the reviews of businesses adjacent to the gap are the best available signal here, because a complaint is a customer describing an unmet need in their own words, at no cost to you. A gap you can populate with real, findable buyers is an opportunity. A gap that is merely geometrically vacant is a hunch.

The largest competitor in your category has the lowest costs per unit, the most recognition, the deepest supplier relationships and the most tolerance for a bad quarter. Competing with them on the dimension they have optimised for years means fighting where every structural advantage is theirs. If they win on price, they can go lower than you and wait. If they win on range, they can hold stock you cannot afford. Attacking a leader's strength is a race decided by resources you do not have.

The alternative is to compete on something they are structurally unable to prioritise. Scale forces standardisation, which means a large competitor typically cannot offer genuine customisation, cannot let a customer talk to someone who remembers their last order, cannot serve a small segment with specific needs profitably, and cannot turn something around at short notice for one person. These are not consolation prizes; they are the things scale gives up in exchange for cost. Choosing one and doing it visibly better is a defensible position, because matching you would require them to dismantle their own advantage.

Not everything needs to be different, and treating differentiation as a goal in itself produces exhausting businesses. Copy freely anything that is a solved operational problem rather than a source of preference. If every competitor offers a clear returns window, sends an order confirmation, publishes their prices, states delivery timelines, or takes the payment methods buyers expect, those are baseline expectations and being different is simply being worse. Buyers do not reward novelty in the mechanics; they notice its absence.

The distinction worth holding is between hygiene and position. Hygiene is what a customer would be annoyed to find missing, and there is no advantage in it — only a penalty for lacking it. Position is the one or two things you are deliberately better at, and it should be narrow enough to state in a sentence and true enough that a customer would recognise it. Copy all the hygiene, quickly and without pride. Spend the effort you save on the small number of things you have chosen to be known for.

Almost everything genuinely decisive about a competitor is invisible from outside. You cannot see their margins, so you cannot tell whether a low price is efficiency or desperation. You cannot see their volumes, so a busy shop and a profitable shop are indistinguishable to you. You cannot see how long their cash lasts, whether the owner is funding it from elsewhere, or whether the business is quietly for sale. Any figure you have about a competitor's performance is either published by them for a purpose or estimated by someone with an incentive, and both deserve the label.

So keep the conclusions to the shape of the market rather than the health of the participants. Where prices cluster, what is bundled at each level, which customer each business is built for, what the reviews complain about, and which positions are unoccupied — all of that is observable and stable enough to plan against. Whether a specific competitor is doing well is not knowable from where you are standing, and plans built on an assumption about it are built on a guess wearing the clothes of a fact.