Methodology

How the numbers are made

Every figure Course & Cloth shows you traces back to a dated snapshot of a public retailer catalog. This page explains the process end to end — including what the numbers can't tell you.

Sources: public catalogs, nothing else

We observe the product catalogs that golf retailers publish on their own websites — the listings any shopper can browse. Currently that set includes TGW, Worldwide Golf, TrendyGolf, Fairway Styles, and Carl's Golfland, and it is growing. We do not use point-of-sale feeds, retailer partnerships, panel data, or any private source. If a number appears in Course & Cloth, it came from a public page.

Snapshots: the catalog, dated

On a regular cadence, we record what each tracked catalog displays: the products listed, their brands and categories, and their prices — both the full price and any sale price the page showed. Each recording is a snapshot stamped with its observation date. Snapshots are kept, not overwritten, which is what makes comparison over time possible. When you see a figure in the dashboard or the brief, it carries the date it was observed.

Normalization: making catalogs comparable

Retailers describe products differently. One lists "Peter Millar Solid Performance Jersey Polo"; another lists the same garment under a different name, category tree, or color-variant structure. Before comparing anything, we normalize: brands are matched to a canonical brand list, categories are mapped to a shared taxonomy (polos, outerwear, headwear, and so on), and prices are parsed into full-price and sale-price fields. Normalization is conservative — where a listing is ambiguous, we keep it out of cross-retailer comparisons rather than force a match.

Comparison: where the intelligence comes from

The product is the diff. By comparing one dated snapshot to the next, we can report things a single catalog view can't show:

Assortment changes — products and brands appearing in a catalog for the first time, or no longer listed.
Price movement — listed prices that changed between snapshots, and how the price bands of a category shift.
Markdown pressure — the share of tracked products carrying a sale price, and how that share moves week over week.
New entrants — brands appearing at a tracked retailer where we had not observed them before.

What "growth" means here

When Course & Cloth says a brand "grew" at a retailer, it means one thing: the observed assortment grew — more of that brand's products were listed in the later snapshot than the earlier one. It is a statement about shelf space, not about sales. Assortment movement is often a meaningful leading signal (retailers tend to expand what they expect to work and mark down what they don't), but the inference is yours to make, and we label the observation so you can make it with open eyes.

The line we don't cross: Course & Cloth never reports or estimates sales, sell-through, demand, or inventory. A product that disappears from a catalog is recorded as a delisting we observed — it may have sold out, been pulled, or moved; we don't know, so we don't say.

Known limits

An honest methods section names its limits, so here are ours:

Coverage varies by retailer. Some catalogs expose rich, well-structured listings; others expose less. Depth of tracking differs accordingly, and cross-retailer comparisons are built only on fields we observe reliably at both.
Prices are as-displayed on the observation date. A price is what the public page showed when we snapshotted it. Member pricing, cart-level promotions, and short-lived flash discounts between snapshots are not captured.
Snapshots are periodic, not continuous. A change that appears and reverses entirely between two snapshots won't be observed. Cadence is chosen to make this rare for the signals we report, but it is a real property of the method.
Unknowns stay unknown. Where the data can't support a claim, the product says so rather than filling the gap with an estimate. The analyst chat cites the snapshots behind each answer for the same reason — so you can check us.

Why this approach

Enterprise research firms sell audited point-of-sale sell-through data, reported after the month closes, at prices set for enterprise budgets. That product is real and useful — and out of reach for most of the roughly 1,100 working golf brands. Catalog observation is the complementary layer: less authoritative about the past, far faster about the present. What's listed, at what price, and what just moved is observable today, and for pricing and merchandising decisions, today is usually when you need it.

Questions about the method? Write to hello@courseandcloth.ai — we answer methodology questions directly.