Product Spreadsheet Research: Links, Prices & Evidence

Catalogue Coverage & Record Identity

Catalogue size describes the range available for discovery. A usable record needs a more specific identity: the seller listing, selected option, recorded price and the date associated with that information. These fields let you compare a candidate with an alternative instead of relying on its photograph alone.

Our saved 22 October 2026 snapshot contains 2,359 records, including 65 with a positive QC-photo count and 258 with an empty or generic brand label under the audit’s named-brand filter. This is the original research dataset, before subsequent display exclusions, and is not the current number of products shown on the website. The following sections explain what those recorded fields can contribute to a comparison.

Link Freshness

A catalogue refresh and a seller check answer different questions. Refreshing a stored entry shows when data was retrieved; it does not demonstrate that the seller still offers the pictured colour or that the same variant can be purchased at the recorded amount. An updated date is more useful when its meaning is explained.

The FTC’s comparison-shopping advice focuses on matching details such as the model, size, colour and shipping fees. Applied to spreadsheet research, that means two records with the same headline are not yet a like-for-like comparison. The selected option and current source information still matter.

Buyer Photos & Recommendation Evidence

A real Reddit moderation discussion criticises promotional finds lists that make strong quality claims without a sound basis. The post is evidence of a community concern, rather than proof that a particular directory or seller is unreliable. Its useful challenge is to ask what supports the recommendation beyond the link itself.

An attributable buyer photograph can add context that a presentation image lacks, but it still shows a particular item at a particular time. A photo count is weaker evidence than the images themselves. Clear source labels help readers separate an actual observation from a publisher’s inference.

Catalogue Signals & Remaining Questions

The table below separates useful discovery signals from conclusions they cannot support. This distinction keeps the shortlist practical: a strong image can earn an item another look, while an unanswered fit question remains unanswered.

Coverage and Missing Fields

The snapshot makes the difference between coverage and evidence visible. Positive QC counts appear on 65 of 2,359 records, or about 2.8%. That is a statement about one stored field. It does not mean that the remaining 97.2% failed an inspection, nor that the 2.8% passed one. Missing evidence and adverse evidence are different findings.

The same distinction applies to the 258 empty or generic brand labels, about 10.9% of the snapshot. They limit brand-based filtering but do not establish how an item was made. Conversely, a named brand makes a record easier to organise without independently verifying the supplied product. Brand completeness and product identity need separate treatment.

These percentages describe our saved selection on the stated date, not every product available through CSSBuy. A category filter also changes the denominator. A count across the whole directory should not be presented as the photo coverage of shoes, jackets or a particular seller unless that smaller group has actually been counted.

Product Identity and Variant Matching

A useful unit of comparison is the selected item version: seller listing, colour, size and package contents where relevant. A photograph and a recognisable headline can identify a design family while leaving several purchasable options underneath it. Comparing only the headline can combine prices or evidence that belong to different options.

Consider an invented example: two records show the same black hoodie, but one price belongs to a plain version and the other to an embroidered version. The difference is not yet a saving on the same item. The fair comparison begins after the versions are matched. The same issue arises when an electronics photograph shows a complete set but the lowest option is an accessory.

Repeated photographs deserve similar care. Two entries using the same image are not two independent observations of quality. They may still be useful links, but image repetition should not increase confidence as though two buyers had supplied separate evidence. Product identity, source identity and evidence independence are three different parts of the comparison.

Review Context and Commercial Incentives

The FTC recommends considering who wrote a review, where it appears and the reviewer’s history, rather than relying on a star score alone. It also highlights incentives such as a free product. That adds an important dimension to spreadsheet research: an attractive recommendation is more interpretable when its commercial context is visible.

For example, a buyer who names the selected size, supplies their own photographs and explains the comparison garment offers specific fit context. A promotional list that repeats “excellent quality” without explaining the basis offers much less. This does not prove the first opinion is correct or the second item is poor; it explains why the statements deserve different weight.

A useful review can also contain a narrow negative result. “The sleeve was shorter than my reference hoodie” is more actionable than a general dislike if the measurement method is clear. Its relevance still depends on whether the version, size and intended fit match your own shortlist. Popularity alone cannot make that connection.

Evidence-Based Shortlist Decisions

A practical comparison separates requirements from preferences. For a laptop bag, a suitable device compartment may be a requirement while colour is a preference. For a hoodie, body length may decide whether the item works under an existing jacket. A low reference price should be considered after the candidate’s essential role is understood.

Suppose candidate A has the preferred colour but no relevant dimensions, while candidate B provides comparable measurements and costs slightly more. The research conclusion can remain conditional: A needs more information; B is currently easier to evaluate. Giving both an invented numerical quality score would hide the reason for the difference.

The best next question is the one most likely to change the decision. Another exterior photograph adds little when the unresolved issue is laptop clearance or trouser rise. A specific missing dimension may resolve it. This is where editorial analysis earns its space beside a catalogue: it helps readers identify the information with the greatest practical value.

Shortlist Decisions & Missing Information

A useful shortlist assigns each candidate a decision status. A comparable item meets the essential requirement and has enough matching information for a price comparison. A pending item has a specific unresolved question, such as body length or included components. An unsuitable item fails a requirement even if its price or appearance is appealing.

For a hypothetical laptop-bag search, record the sleeved device dimensions as the requirement. Keep a bag pending when it has only external dimensions; remove one whose stated device compartment is too small. Compare prices among the remaining candidates using the same currency and included costs. This prevents a low headline price from dominating the decision before suitability is established.

Record the reason alongside the status, with the source and date of the decisive information. When a seller option changes, you can revisit that particular decision without rebuilding the whole shortlist. The research output is a small set of comparable choices and a clear list of questions worth resolving.