Comparing Product Lists: Selection, Variants & Photo Evidence

The spreadsheet as an editorial object

Google Sheets supports sorting, filters and individual filter views. Those features explain part of the appeal of a shared product table: the same underlying records can be rearranged around different interests. Someone focused on footwear can isolate shoes; someone thinking about a budget can change the price order. The format is useful because it makes a collection negotiable rather than forcing everyone through the same sequence.

Yet a filter cannot recover a product that was never selected. If a list contains many colour variations from a small group of brands, its apparent size can exaggerate the range of ideas it offers. Selection happens before sorting. That makes a spreadsheet an editorial object as well as a data object: the person compiling it determines which silhouettes, materials, price points and visual identities receive attention.

Consider a hypothetical 24-row clothing edit. If twelve rows repeat one T-shirt and twelve repeat one hoodie, it contains two model families. Another 24-row edit might contain eight model families in three colours each. Both can serve a reader who values colour choice, but the second offers more garment-level alternatives. Neither row count should be presented as a complete measure of breadth.

A collection can be broad in one dimension and narrow in another

A black T-shirt in three sizes is not the same kind of variety as three differently cut T-shirts. Google Merchant Center makes a related distinction in its product-data model: individual variants have their own identifiers, while an item-group identifier connects related versions such as colours and sizes. That commercial data convention is useful background for understanding why product counts require context; it is not a claim that this directory participates in that programme.

For a discovery page, displaying every variant separately can make a narrow assortment look expansive. For an order record, however, collapsing variants can erase information that matters. The two views answer different questions. Discovery is concerned with the range of possible choices. A transaction needs an exact item, variant, destination and price. Confusing those purposes produces both repetitive browsing and ambiguous comparisons.

Our category edits contain 24 references each. In the frozen 2 November 2026 selection, the T-shirt category spans 17 brands and the bag category spans 16. Brand count is one check against repetition, not a measure of every design difference. Two different labels may still supply similar black chest-logo shirts; one label may supply several substantially different cuts.

Use the product categories to compare a focused selection rather than treating a larger row count as a better choice.

Discovery, listing information and order evidence

Fishgoo describes its service as purchasing from supported Chinese marketplaces, followed by warehouse inspection and international shipping. A third-party directory occupies a different position in that process. It can introduce an item, explain a design or link to fuller information. It does not become the seller, the purchasing account or the warehouse record merely because a reader arrived through a Fishgoo-related search.

Three information layers are worth keeping separate. A discovery card gives a quick visual and editorial identity. A product listing supplies the selected offer and its available details. An order-specific warehouse record concerns a particular purchase after intake. A polished card is not evidence that the warehouse has received that item, just as a photograph attached to a historical listing is not necessarily a photograph of a future order.

This separation makes the directory more useful. The reader can value the first layer for inspiration without asking it to prove something only the later layers could establish. A good introductory page creates a clear route toward the next relevant source instead of implying that all uncertainty has already been resolved.

The shared QC gallery analysis examines the gap between a photographed example and the exact variant being considered.

A comparison of information layers

The distinction can be expressed without treating any one format as a complete shopping system. Each layer has a useful role and a different limit.

A dated record needs field-level context

One updated date can conceal several different events. An editor may change a title without rechecking a price; a source may update stock while retaining older photographs. A useful date identifies the observation it belongs to. It should not make an old image look newly taken or turn a copied price into a fresh transaction quote.

Suppose an editor checks a listing on 3 November and finds that only the black variant is available. A photograph saved on 2 November can still illustrate the design, but it should not be described as a new stock check for every colour. Likewise, a red variant marked unavailable is a specific observation; a blank stock field simply means the information is absent. The two states should remain distinct.

The distinction also improves later corrections. If a destination changes its title, the original product ID can preserve continuity. If the seller changes the offer itself, matching the old name is insufficient. A record that keeps identity, source, selected option and observation date together makes that difference visible instead of burying it in a general refreshed label.

In our frozen study, 203 selected product IDs included 200 records with QC photographs and 159 with a non-null representative parcel-measurements object. Those counts describe field coverage. They do not establish inspection outcomes, measurements for every variant or the accuracy of a future delivery. The study is useful precisely because its scope is narrow enough to state clearly.