Price Architecture Monitoring: Reading a Competitor's Price Ladder, Not Just Their Price Changes

Most competitor price monitoring answers one question well: did this product's price change, and by how much. That is the right question when you are defending a specific SKU, and it is the question alerting is built around.

It is also a question about a single point on a curve. It tells you nothing about the curve.

Price architecture is the shape of a catalog's prices taken together. Where does the range start, where does it end, where do most listings cluster, and where are the gaps. That shape changes far more slowly than any individual price, and when it does change it usually reflects a decision someone made in a planning meeting rather than a tactical response to a slow week. Which is exactly why it is worth watching.

A Price Change and a Price Position Are Different Reads

A competitor drops one product from 84.00 to 69.00. Your alert fires and you triage it. Depending on context it might be clearance, a promotion, a response to your own move, or an error on their side. The competitor price monitoring operator playbook covers that per event work, and most operators are reasonably good at it.

Now consider a different pattern. Over eleven weeks, that same competitor's cheapest listing moves from 84.00 to 44.00, their most expensive listing climbs from 210.00 to 290.00, and the middle of their catalog thins out. No single price change in that sequence would have looked urgent. Several of them probably never fired an alert at all, because they arrived attached to new products rather than to changes on existing ones.

Taken together they say something a price alert cannot: this competitor has stopped being a single tier brand and is building a ladder. They are opening a cheaper entry point to catch price sensitive shoppers and stretching a premium tier at the top to protect their average order value. That is a positioning change, and it will affect you long after the individual price moves are forgotten.

The per event view and the structural view are complements, not substitutes. You need the alert to react this week. You need the architecture to plan next quarter.

The Four Cuts Worth Tracking

You do not need a statistics practice for this. Four numbers per competitor, recorded on a regular interval, carry most of the signal.

The entry price point. The cheapest thing a customer can actually buy. This is the number that anchors a first time visitor's sense of whether a brand is expensive, and it is the one most likely to be a deliberate choice rather than an accident. Use the cheapest purchasable listing rather than the cheapest listing in the catalog, because a product that is visible but sold out is not really setting an entry point.

The head of the range. The most expensive listing. Movements here are usually about brand positioning or a new premium line. A ceiling that climbs while everything else holds steady often means a competitor is testing whether their customers will pay more, and that test usually shows up as a small number of new listings well above the existing range.

The center of gravity. Where the bulk of the catalog actually sits. Use the median rather than the average, because a handful of expensive outliers will drag an average around and tell you very little. The median is the price a shopper is most likely to encounter while browsing, which makes it a better description of how a brand reads than either extreme.

The gaps. Look for price bands where a competitor has nothing. Gaps are the most actionable part of this read, because a gap is either a deliberate refusal to compete in that band or an oversight. Either way, it is a band where you would meet less resistance. When a competitor closes a gap they had held open for a year, they made a decision, and it is worth knowing about.

Record these four per store, with a date. That is the whole instrument.

What a Shift in Architecture Usually Means

Structural moves fall into a small number of recognisable patterns.

Opening a lower entry point. A new cheapest listing appears meaningfully below the old floor. Often a smaller size, a starter kit, or a stripped configuration. This is usually defensive, aimed at a cheaper rival taking the price sensitive end of the market, and it frequently precedes heavier promotional activity. If you see it alongside deepening discounts, discount depth signals will tell you whether the entry point is a real structural addition or a temporary offer dressed as one.

Stretching upward. New listings appear above the existing ceiling while the rest of the catalog holds. This is a margin move, not a volume move. It tends to arrive with new products rather than reprices, which is why launch tracking and architecture monitoring read better together than either does alone.

Compression. The range narrows from both ends. Usually a simplification, sometimes a supply constraint, occasionally a brand that has decided it was trying to serve too many customers at once. Compression is easy to miss because nothing dramatic happens on any given week.

Thinning the middle. The floor and ceiling hold, but the median moves and the count of mid band listings falls. This is a classic good better best rebuild, and it usually means a competitor concluded their middle tier was cannibalising their premium tier. Check removals here: if the disappearing listings cluster in one band, catalog removal signals will confirm the middle was cleared deliberately rather than drifting.

Drifting up as a whole. Every cut moves together. Often just cost pressure being passed through, which is unremarkable when it happens across an entire category and much more interesting when only one competitor is doing it.

The pattern names matter less than the habit of asking which one you are looking at before you decide what it means.

Reading Architecture Against Signals You Already Track

Architecture is context, and context is most valuable applied to something else.

A price drop inside a band a competitor is actively building reads as investment. The same drop inside a band they have been thinning for two months reads as clearance. Same event, opposite conclusions, and only the structural view separates them.

Stockouts read differently too. Persistent sold out listings at the bottom of a range can mean the entry point is more demand than the competitor planned for. Persistent stockouts at the top usually mean a small production run rather than a supply problem.

And architecture pairs naturally with assortment. Category and vendor mix monitoring tells you what a competitor is choosing to sell. Price architecture tells you what they are choosing to charge for it. A store shifting into a new category at the top of its price range is making a very different bet than one entering the same category at the bottom, and you cannot tell those apart from either cut alone.

One caution. Rapid, broad downward movement across an entire range is a different situation with a different response, and it should not be treated as ordinary repositioning. Price war early signals covers that case.

A Monthly Review, Not a Weekly One

The most common mistake with this read is checking it too often. Price architecture moves on a scale of months. Reviewed weekly it produces noise, invites overreaction, and burns attention you need for signals that actually move fast.

Monthly, roughly twenty minutes. For each tracked competitor, record the four cuts. Compare against last month. Most months, nothing meaningful will have moved, and recording that is not wasted effort. It is what makes the months that do move legible.

When something moves, write a hypothesis. One line, before you look for confirming evidence. "Entry point dropped 30 percent, likely defending against the cheaper rival that entered in May." Writing it down first is what separates analysis from narration, because structural data is very easy to explain after the fact.

Quarterly, review the hypotheses. Some were right, some went nowhere. Keeping the misses visible is the part most teams skip, and it is the part that keeps the practice honest.

Feed conclusions into planning, not into this week's pricing. Architecture findings belong in assortment and pricing planning conversations. If a structural finding is changing your prices this week, you have probably mistaken a promotion for a repositioning. Your weekly rhythm should stay focused on events, which the weekly operator review playbook covers.

What Bonesaw Can Infer and What It Cannot

Being precise about the boundaries here matters, because this is a read that invites overreach.

Bonesaw observes publicly accessible storefront data: listing prices, variant prices, compare at prices where a merchant publishes them, availability, and product and collection metadata. From repeated observations it can show you how those values move over time. That is enough to build every cut described above.

It does not see cost, margin, units sold, revenue, conversion rate, or traffic. This is the limit that matters most. A competitor's median price tells you where their catalog sits, not where their business sits. If eighty percent of their revenue comes from three products at the top of the range, the median describes almost none of their actual economics, and nothing in public storefront data will reveal that. Treat architecture as a statement of merchandising intent, which is genuinely useful, and resist converting it into a claim about performance.

A few smaller boundaries are worth stating plainly. Published prices are not transacted prices, so cart level discounts, codes, and negotiated terms are invisible. Catalogs that are seasonal or heavily configurable will show range movements that are ordinary inventory rotation rather than strategy. And a single snapshot is close to worthless: the first review only establishes a baseline, and the read becomes meaningful once you have several months to compare against. That is an argument for starting before you need the answer.

Finally, this works best on competitors with a coherent catalog. A store with a few thousand listings across unrelated categories does not really have one price architecture, it has several, and the aggregate numbers will average away the thing you were trying to see. Cut by category in that case, or accept that the store is a poor candidate for this read.

Getting Started

Pick three competitors you already monitor. For each, write down four numbers today: cheapest purchasable listing, most expensive listing, median listing price, and any obvious empty band. Add the date. Ten minutes, total.

Come back in a month and do it again. The comparison is the entire product, and you cannot get it retroactively, which is the honest argument for starting the habit now rather than when a competitor's move has already landed.

If you want a broader starting point for what else belongs in a monitoring routine before you add this one, the competitor monitoring checklist covers the wider set. And if you are already tracking stores in Bonesaw, the price history you have been collecting is the raw material for this read, so your first monthly comparison may be available sooner than you think.

Frequently Asked Questions

How is this different from tracking competitor price changes? Price change tracking is event based and tells you what moved this week. Price architecture is structural and tells you where a catalog is positioned overall. A competitor can have heavy price change activity with a completely stable architecture, which means promotional churn rather than repositioning, and that distinction should change how you respond.

How many competitors can I realistically do this for? Three to five. This read rewards depth over coverage, and the value comes from consistent monthly records rather than from breadth. Tracking twenty stores badly is worse than tracking four well.

Should I use average or median price? Median. Averages are pulled around by a small number of expensive listings, so an average can move significantly when a competitor adds two premium products and changes nothing else. The median describes what a browsing shopper actually encounters.

Does this work for a single product brand? Not well. A store selling one product at three sizes does not have a meaningful price ladder. This read needs a catalog with enough listings for a distribution to exist, which in practice means a few dozen at minimum.

Can this tell me a competitor's margins or what is selling? No. Public storefront data contains prices, not costs, units, or revenue. Any read that claims to infer profitability or sell through from published prices alone is inferring well past the evidence, and building a plan on it is a good way to be confidently wrong.

What if a competitor's prices barely move at all? That is a finding, not a failure. A stable architecture over six months tells you a competitor is committed to a position, which is useful when you are deciding whether a band is worth contesting. Stability is information as long as you recorded it.

---

Bonesaw is a product of MoonsLink. Monitoring capabilities described in this guide reflect publicly accessible product and storefront data collected through standard web protocols. Bonesaw does not access private or authenticated data. All data collection respects robots.txt directives and site access policies.