Comparison sites are inventory dressed as judgement

This works through comparison and aggregator sites in the order the parts actually depend on each other.

What a comparison site actually is

Most comparison services are advertising businesses that earn from clicks or commissions rather than from readers. The set of products shown is therefore the set with a commercial relationship, which is a subset of what exists.

Presenting that subset as a table implies completeness that the underlying data does not have. Default sort orders are a commercial decision and are rarely explained on the page. This does not make the data wrong, but it makes the boundaries of the data invisible.

Where the specification data comes from

Fields are usually populated from supplier feeds, which are prepared for commerce rather than for accuracy. Errors propagate widely, since many sites use the same feeds and none of them verifies against the manual.

The measurable part is this: missing values are common and are often displayed as absence of a feature rather than as absence of data. Units and conventions differ between manufacturers, so a column can compare quantities measured differently. Checking any decisive figure against the manufacturer's own documentation is a necessary habit.

Filters and the shape they impose

Filters exist only for fields the database holds, which pushes decisions toward whatever happens to be recorded. The attributes that predict durability, such as serviceability and parts availability, are never filterable.

After the warranty ends, this channels buyers toward specification numbers, which is precisely the failure mode that produces vanity metrics. Range filters with fixed bands can exclude products just outside a boundary for no meaningful reason. Being aware of what cannot be filtered is more useful than becoming skilled at the filters that exist.

Price comparison specifically

Prices shown may exclude delivery, may be conditional, or may be from sellers with limited stock. Availability data lags, and the cheapest listing is frequently the one least likely to be fulfilled promptly. Seller reputation, returns handling and local warranty support are worth real money and are rarely compared.

Judged against the category, grey imports appear alongside domestic stock without any indication that warranty terms differ.

Pricing strategy is a subject of its own, and this site deliberately leaves it to those who cover it properly.

Aggregated review scores on comparison pages

Scores are frequently imported from other sites and averaged, which compounds the compression already present. The source and date of imported scores are usually not shown, so an old review can dominate a current listing. Some aggregators generate summary text automatically from reviews, which introduces further distortion.

Judged against the category, a score shown without a link to its source cannot be checked and should be treated as decoration. Where sources are linked, following two or three of them is quicker and far more informative.

Manufacturer figures are measured under conditions the manufacturer chose.

Using them for what they are good at

They are genuinely good at establishing what exists in a category and at what price points. They are good for finding model numbers, which then let you search manuals, parts lists and forums. They are good for spotting the range structure of a manufacturer's line-up at a glance.

Compared like for like, they are poor at judging quality, durability and suitability, which are the decisions that matter most. This site does not aggregate listings, and using a comparison site as a starting index rather than a conclusion is the sensible approach.

The takeaway

Use comparison sites to find out what exists, then leave them before you decide anything.

The question is rarely which is best. It is which is enough.