The Box Shop: user behavior and its effects.
The Box Shop is a small, finely curated Munich store for special gift packaging. The analysis examines how findable the shop is via search engines, what effect measurable user behavior has on Google rankings — and how much RankBrain influences the picture.
Time on page, page views, bounce rate.
Can the poor placements be explained by user behavior? Parallel coordinates show: no clear pattern.
Page views
The home page is viewed much more often than the blog articles from 2016. The question is whether the number of views really determines the ranking.
Time on page
The home page achieves much better placements with at times significantly shorter dwell times. Under identical scoring rules, time on page therefore cannot be a decisive ranking factor.
Bounce rate
All ranked blog articles from 2016 show a high bounce rate (up to 85%) — but for blog articles that is nothing unusual. No clear correlation here either.
Entries
Even documents that are directly accessed far less often achieve good placements. The number of entries is not a decisive metric.
Load time, click depth, linking.
Structural aspects show clearer effects — but are still not the main cause.
Page speed
No top placement is generated by documents delivered slower than ~2.5 seconds. Fast delivery is a baseline requirement — but on its own it is no guarantee of good placements.
Click depth
Documents reachable from the home page in fewer clicks tend to receive better placements. The blog articles from 2016 require at least two clicks to reach.
External links
The home page has the most independently referring domains. The poorly placed blog articles have very few inbound links.
Internal linking
Even documents with few internal links can reach top positions. The sharp demotion of the blog articles cannot be explained by this.
Text length
More than half of all placements come from documents containing between 1,300 and 1,600 words — but good placements range from 600 to 3,400 words. What counts is which words appear in them.
The most important factor turns out to be RankBrain — Google associates the "packaging" context with The Box Shop only as a second choice.
When Google classifies the context differently.
Comparing the slots above and below the blog-article placements shows: the "gift box" context achieves very good placements. For "packaging" and "crafting" there is not a single good placement.
Gift box
Very good placements — Google clearly assigns this context to the shop.
Packaging
Other documents, including the home page, are also demoted in the "packaging" context. Google does not consider this context a sufficient fit.
Do-it-yourself
The crafting component of the blog articles is not sufficiently linked semantically with gift packaging.
How we lift the placements.
Structural measures first — then rethink content strategy.
Delivery
Ensure fast delivery of all documents — not just the most important ones.
Accessibility
Important documents should be reachable quickly and in a few clicks.
Internal linking
Internal linking should be reviewed and intensified.
External promotion
Promotional campaigns with external links are never wrong.
Consider spinning off
If these measures are not enough, it may make sense to move the crafting content to a separate (sub-)domain. Google's top results for "packaging" only feature pure shops or pure crafting guides — the hybrid of The Box Shop has no place there.