Answer 07 of 12
How do search and tags work for these nature images? – Cheap Nature Images
Plain keyword matching, a page for every keyword, and the three habits that find the right frame fastest.
Search matches words in titles, descriptions and keyword sets, and orders matches newest first. There is no relevance model inferring intent, no personalisation, and nothing is promoted for money.
That is a limitation and an advantage. It will not rescue a vague query, and it will never quietly hide a result you asked for.
Tags
Every keyword on every image becomes a tag with its own page listing all of its images. Nothing is dropped for being rare, which is precisely what makes the narrow tags valuable: a page for a specific plant or a specific weather condition with a dozen images is more useful than a page for the word "nature" with everything on it.
The tag index is paginated. One page holding every tag in a large catalogue would be slow, unreadable, and would divide its own weight so finely that none of its links benefited.
Three habits
Search the noun, then pivot on tags. Find one image that is close, open it, and use its tags to move sideways. The tags on a picture you already like are a far better filter than a second guess at a phrase, and hardly anyone uses them this way.
Search conditions, not adjectives. Fog, frost, backlit, overcast, golden hour, long exposure, aerial, macro. All recorded, all discriminating. "Beautiful", "stunning" and "peaceful" are recorded nowhere and match nothing.
Use the specific name. If you know the species, the terrain type or the region, type it. Common and scientific names are both indexed where they were documented.
Empty results
The catalogue is nature. A query outside that returns nothing, which is honest rather than broken. Nothing is padded with loosely related images to make a page look full.
If a phrase should match and does not, it is probably missing from the keyword sets. Send it; that is the easiest thing here to fix.
Browsing instead
Three routes when search is not the right tool. The tag index lists every keyword. The collections group images by price. The HTML sitemap, linked in the navigation, lists every page on the site including every image, paginated.
Related images
Each image page carries a bounded set of related frames chosen by keyword overlap. It is capped deliberately: a related block that grows with the catalogue turns every detail page into a directory.
Why there is no filter panel
No sliders for orientation, colour or people count. Filters look helpful and they generate an enormous number of nearly identical pages, each with its own address and almost no content of its own.
Tags do the same job with pages that are worth existing: one address per real concept, each listing a genuine set of images. If you want landscape-orientation frames of a particular subject, the subject tag plus your own eye is faster than a filter panel and does not leave a few thousand thin pages behind it.