Being the cheapest is usually a mistake you cannot see

Cheapest is a position in a list. Almost nobody chose it on purpose.

Picture a category manager opening a view of her catalogue that shows, for every line, where her price sits against the competitors she tracks. Not the gap in dollars. The position. Cheapest, equal cheapest, fourth of seven, dearest.

She is expecting a spread. What she gets is a column of "below market" running down two hundred rows of a category she has never once described as a discount category. Cheaper than everybody, on nearly everything, for most of a year.

None of those prices is wrong the way an auditor would mean it. Each came off a costing sheet, a supplier list, or a rule somebody wrote in good faith, and each is perfectly defensible on its own. The problem only appears when you stop reading them as numbers and start reading them as positions. Read that way, the catalogue says something nobody in the business has ever said out loud: we are the floor of this market.

The number is yours, the position is not

A price is something you set. A rank is something that happens to you.

That distinction sounds academic until you notice which one the customer experiences. Nobody standing in front of a search results page evaluates whether $189 represents a fair return on your landed cost. They see an ordered list and they see where you are in it. The number is your input. The rank is the output, and it moves without you touching anything, because it is a function of eight other businesses' decisions as much as your own.

Most retailers manage the input obsessively and never look at the output. There is a margin report, a cost review, a promotional calendar. There is rarely a document that says: in tapware we intend to sit mid-field, in accessories we intend to sit at the top, and in the four lines everyone uses as a price check we intend to be within a dollar of the cheapest.

[NEEDS FIGURE — the share of a catalogue that already sits below every tracked competitor at the moment monitoring is switched on, measured across production tenants. The sentence it belongs in: "When a catalogue is first connected, about [X] per cent of lines are already sitting below every competitor being watched, and almost none of that was a decision." Computable from the 'below' state in lib/price-position.ts over each tenant's first complete walk.]

Rank one has nothing underneath it

Every position except cheapest is defined by a price somebody else has already published. If you want to sit third of seven, you need to know two prices and stay above them. If you want to hold a dollar under the cheapest, you need to know one price and stay a dollar under it. Both of those have a stopping point built in, and the stopping point belongs to the market rather than to you.

Cheapest has no such reference. It is defined only by the absence of anyone lower, which means the person who defines it is whoever turns up next with a smaller cost base, a clearance run or an automated rule. You have committed to a position and handed the pen to a stranger.

The practical consequence is that being cheapest is the most expensive rank to hold. To keep it you have to know every price, every day, forever, and be willing to go under any of them at any hour. To hold fourth of seven you have to know roughly where the field is and check occasionally. One of those is a business. The other is a treadmill you have volunteered for.

There is a second consequence that is quieter. When you are the cheapest, the market gives you no feedback at all. Being too dear produces silence and lost sales, which at least eventually shows up in a conversion rate somebody investigates. Being too cheap produces nothing. The units still move. The margin percentage still reads acceptably against the cost you paid. The customer who was always going to buy from you simply pays less, says thank you, and leaves.

You are only cheapest of the sellers you chose in 2023

Here is the part that undermines the whole exercise if you get it wrong. Your rank is measured against a field, and the field is a list you assembled at some point and have probably not revisited.

Track four sellers out of the nine who actually appear on the first page of a search and "below market" means below four. You may be fifth. A retailer who confidently reports being the cheapest in a category is, with no bad intent anywhere, reporting a fact about their own competitor list.

This cuts the other way too, and more often. Old competitor lists tend to be full of sellers who no longer matter — a marketplace account that stopped restocking, a specialist who exited the category, a chain whose online arm now only ships to two states. Every one of those still drags your measured position around and still gets a Monday morning of somebody's attention.

So before you argue about what rank you want, settle who is in the field. Take the list, go line by line, and ask whether you would genuinely change your price because of what that seller did. If the answer is no, they are not a competitor, they are a habit.

Where you want to sit changes by category

The rank that is right for a line depends on things you already know and have never had to write down.

How comparison shopped is it? A commodity consumable with a barcode that any customer can paste into a search bar behaves nothing like an installed appliance where the quote includes removal of the old unit. The first one is priced against a list. The second is priced against a proposal.

Does the customer know what it should cost? People have a reference price for a kettle and no reference price at all for a thermostatic mixing valve. Where the reference is strong, position matters and small gaps get noticed. Where it is weak, the market has far more room in it than most retailers use.

What comes with it? Stock on the floor, next-day delivery, a trade account, someone who answers the phone on a Saturday, an install crew. Every one of those is a reason to hold a rank above the cheapest and be entirely comfortable about it.

What does it pull through? A line that sits at the front of a basket is worth being aggressive on in a way that a line at the back is not. This is where deliberately taking rank one earns its keep, and it works precisely because it is a short defended list rather than an accident spread across the catalogue.

Turning a position into a rule

A position you hold in your head is a position that lasts until the person holding it takes annual leave. The version that survives is written down and applied by something that runs every day.

That is what pricing rules are for, and it is why ours are scoped by category and brand rather than set once for the whole catalogue. A rule can target the cheapest competitor, or a fixed distance under it, or the third-lowest price in the field — the position, not a number you have to keep re-deriving. Underneath it sit the guards: a margin floor that will not be crossed whatever the market does, and a ceiling that stops a rule quietly walking a price upward. When a recommendation comes back, it says which competitors counted, which rule fired, and which guard bound.

Then the catalogue view tells you where you actually landed. Below market, equal lowest, within market at fourth of seven, above market. Not a gap in dollars. A position, on every line, that you can look at and say yes, that is the one we picked. See how it works.