Domain preference: personalised ranking by the sites a user favours
Personalised ranking has a domain preference twiddler that annotates each result. It reflects one user's preferences, not the authority of a site. How the preference is learned is not known.
01What the schema shows
Two objects sit in the quality_twiddler namespace: P13nRankDomainPreferenceTwiddlerParams, the settings, and P13nRankDomainPreferenceTwiddlerResultAnnotation, an annotation on each result. P13n is the usual shorthand for personalization: a p, thirteen letters, an n. RESONEO describes it as the domain preference of personalised ranking, written onto each result. Like every twiddler, it only recommends; the framework in Superroot, the layer that merges what Google's indexes return, decides. The parent article on twiddlers covers that framework.
02What Google says about personalised results
Google's help page says that with personalised recommendations on, Search uses information such as the user's Search services history. Google's 2022 post on personalisation adds that its systems personalise only when doing so gives more relevant results, and gives shopping results from brands a user likes as an example.
Neither page mentions domains or this twiddler. Whether domain preference is learned from clicks, from explicit choices, or from both is unknown. It may be related to Pref, which RESONEO links to Preferred Sources, but no source shows that.
03E-E-A-T
A domain preference is easy to misread as site authority. It is not. It records which sites one user favours, not how trustworthy a site is for everyone. There is no E-E-A-T bearing. The only lever for a site owner is the slow one: people who return to a site because it serves them well.
04References
Other (5)
- [1]CRESONEO, "Twiddlers: what still moves a result after ranking", schema list ↗
- [2]CRESONEO, "Twiddlers: what still moves a result after ranking", family "Freshness, repeat queries and preferences" ↗
- [3]OGoogle Search Help, "Personalization & Google Search results" ↗
- [4]OGoogle, "Helping you find and control personalized results in Search", 28 Sep 2022 ↗
- [5]Ceeat-mechanics, own reading