Selling Guide Methodology and Evidence Labels

What the selling guides measure

Each guide begins with a dated Reach Dog category profile. A category cluster count is the number of distinct product clusters carrying the stored category term. What, Where, When, Who, and Why rows are overlapping associations around those clusters.

Cluster counts are not monthly search volume, sales, revenue, market size, conversion rate, or profitability. The guide does not add overlapping dimension counts together.

Four evidence labels

Observed category association means the relationship appears in the dated Reach Dog category profile. It describes the category evidence, not every product inside the category.

Modeled buyer wording means Reach Dog turned a category association or product constraint into a natural shopping question for merchant review. Unless a question is explicitly accompanied by search-volume evidence, it is not presented as a measured query.

Merchant-supplied product fact means the fact comes from the merchant’s product record, manufacturer documentation, certification, testing, or another source the merchant can verify.

Unsupported means the available category and product evidence does not justify the statement. Unsupported ideas should be removed or held for verification.

How the five dimensions are used

The five buyer dimensions organize the questions a listing may need to answer:

What: the requested use, activity, or product function
Where: the setting, room, surface, or occasion
When: timing, season, duration, or moment of use
Who: the buyer, user, recipient, or household context
Why: the supported benefit or result being compared

The dimensions organize evidence. They do not turn an association into a fact about an individual product.

Why the questions require product verification

A category profile can show that office chairs are associated with ergonomic support. It cannot establish that a particular chair has adjustable lumbar support. A mattress category can be associated with back support without proving that an individual mattress relieves pain.

For that reason, every selling-guide question names the product facts a merchant must verify before using the recommended content. Missing facts remain missing; an AI assistant should not supply them from general category knowledge.

Research dates and updates

Reach Dog began this research program in January 2024. Each guide carries its category snapshot date, content-review date, and content fingerprint. A page date should change only when its evidence or public guidance changes materially.

Search volume, CPC, and rolling keyword history appear only when the public evidence package contains those fields and dates. Their absence is not replaced with an estimate.

September 8 qualifier subsets

The lipstick-with-SPF, sensitive-stomach dog-food, and pet-carrier guides use a separate cluster check dated September 8, 2026. Membership starts with derived_categories tags on the clusters themselves. A title filter then requires the product and qualifier wording described on each guide. This extra filter excludes unrelated records that also carry a matching tag.

These are bounded subsets, not complete category totals. The lipstick check found one matching cluster, with no populated buyer-dimension tags. The food and carrier pages count dimension labels within their own screened clusters, once per label per cluster. Labels overlap and are not added. The results do not establish label accuracy, health effects, or suitability for any individual buyer or animal.

The same check did not establish a flat-feet running-shoe subset. That guide retains its clearly labeled August 10 broad-category evidence pending a reviewed mapping. An empty result under this filter is not evidence that such products do not exist.

The reproducible query, cluster identifiers, selected titles, and dimension counts are stored in the repository’s qualifier_cluster_evidence.json. The guide’s original publication and review dates are preserved; the new evidence date is stated separately. No keyword demand figures are published from this export. Where demand is used for internal topic ranking, a keyword is counted once at its best match score within the topic, rather than summed again for every matching cluster.

Choose a category to apply this method

Open a guide to check its source evidence and work through its product-specific questions.

Pets

Browse the pets collection, or choose a product:

Beauty and personal care

Browse the beauty and personal care collection, or choose a product:

Women’s apparel

Browse the women’s apparel collection, or choose a product:

Footwear and fitness

Browse the footwear and fitness collection, or choose a product:

Travel and everyday carry

Browse the travel and everyday carry collection, or choose a product:

Outdoor and recreation

Browse the outdoor and recreation collection, or choose a product:

Home and work

Browse the home and work collection, or choose a product:

See these ideas against your own catalog.

Read my store See the commerce map