Do you know the quality threshold in SEO? You should… Little mentioned in France, the notion of quality threshold in Google deserves some attention. Its interest, in case it is used by Google? Crossing a threshold in the eyes of the search engine.
It’s not necessarily obvious to know the different ranking factors of a page in the Google SERPs, the secrets of the search engine algorithm oblige. But a little-known concept deserves some attention: the threshold of quality. “When we talk about exceeding the quality threshold, this represents the fact of submitting content to Google that exceeds the level required to be included in the main index”, summarizes Jean-Baptiste Bessière, SEO consultant at Pixalione. “If the gap between the quality threshold and the quality of our content is large, we can expect a dramatic increase in our content positions as we move into the primary index.” How to achieve this? Here are some ideas for better understanding and using the quality threshold.
“A still vague concept”
The quality threshold was largely developed by the late SEO pioneer Bill Slawski, known for being the first to talk about Google’s primary and secondary index, and Koray Tuğberk Gübür, creator of the site holisticseo.digital. It implies a score to be ranked for a given query or query network. This would allow Google to rank pages more efficiently and crawl them faster.
However, this concept is not well known, especially in France. “It’s a bit of a vague idea,” explains Sylvain Peyronnet, CEO of Babbar.tech. “Google Search analyst Gary Illyes talked about it in a Q&A where he discusses a score below which a page isn’t indexed.” “Unlike predictive ranking and Google’s proven primary and secondary indexes, the quality threshold has only been theorized, recalls Jean-Baptiste Bessière. It is not a metric used by Google, but rather the name that designates the fact of creating content whose criteria far exceed those required to pass Google’s main index for a query.
Are there any other clues to the existence of this quality threshold? Hard to say. For example, in 2012 Google filed for the Site Quality Score patent. According to this document, the score could be determined mainly by user actions. But just because Google files for a patent doesn’t necessarily mean it uses it for its search algorithm.
Another element, John Mueller, webmaster trends analyst, responding in 2021 to a question about the existence of this metric, said that he believed that this notion should not be quantifiable, as is the case with the Google Ads quality score. However, he did indicate that a quality metric could appear in Search Console in the future. But this hypothesis seems unlikely to him, in particular because his exhibition could open the door to “abuse according to him”.
A complex operation
According to one of the promoters of the quality threshold, Koray Tuğberk Gübür, broadly speaking, in order to set the quality threshold, Google would group queries and documents into a thematic cluster. This term refers to “how Google will categorize and group different queries according to their theme”, per Jean-Baptiste Bessière. Pre-selected documents for each theme are ranked when a query belonging to a cluster is used.
For Koray Tuğberk Gübür, the order in which the documents of the main index are placed in this type of thematic cluster is currently determined by a “triplet” classification method. This takes into account the subject of the request, the central element of the request and the subject of the request. Different quality thresholds are determined, for different formats of nouns, verbs and other types of words such as adjectives. Therefore, depending on how a query is formulated, the results will vary to best fit the user’s query.
Naturally, the more similar documents the corpus has, the higher the quality threshold will be. Conversely, the smaller the corpus, the lower the quality threshold for relevance. In this case, Google would rank sources with high PageRank, “even if their web pages don’t perfectly match the specific query. They are relevant to the overall topic,” writes Koray Tuğberk Gübür.
In other words, for Fabien Raquidel, SEO consultant, if no expert content appears on a given query, the first one in the SERP may have an average page, but above all it will have a significant Trust. “That’s why many media sites are in the top positions on many queries, even if their content is not perfect, Google trusts these sites and they meet the quality threshold, in at least reassuring Internet users. Another example, they prefer to have Lelynx compared redemption-de-credit.com in the first results of the SERP on the corresponding query.By displaying these results, Google assumes no risk.
Elements to be improved on a case-by-case basis
For those who think that the quality threshold exists, various actions can be implemented, first you need to know what type of Serp you are dealing with. “Some SERPs give internet users satisfaction. The quality threshold then seems ‘fixed’, as for the ‘real estate appraisal’ request. It is then very difficult to position yourself among the top 4”, estimates Fabien Raquidel. “However, other Serps, apart from those related to the notion of freshness, show a fluctuating quality threshold. The top positions change the results regularly, as currently on queries related to CBD. In this case, it is easier to rank in the top positions. “
Serps can also return different query styles. “For example, the query ‘SEO consultant’ currently returns results on profession and training, but also on the service provided by freelancers.” Thus, for Fabien Raquidel, “the threshold of quality means nothing if we do not add intentionality”.
Once you figure out your SERP type, another difficulty is knowing what changes to make. “Topics with a high impact on users’ lives, such as YMYL queries (Your money Your life, with topics dealing with finance, savings, health…, ed) can have much higher EAT criteria (Competence – Authority – Reliability, ed). On less serious subjects, such as journalists, the criteria can promote popularity,” observes Jean-Baptiste Bessière.
“The quality threshold means nothing if we don’t add intentionality.”
But how to know the elements to modify for each request? “Semantically, we can analyze the competition, to see on a group of requests what similarities we will find between the competitors in the lead on the subject, as some agency tools already do”, develops Jean-Baptiste Bessière. “We can also use the method proposed by the Peyronnet brothers. It involves crawling the SERPs on a theme. And analyzing the results using a machine learning algorithm, such as XGBoost or Random Forest. They allow you to get predictions on the most important criteria to rank on the topic in question. The method can be automated using the data science platform Dataiku.”
These processes remain difficult for people with limited resources to implement. “And, although precise, the results remain variable and the quality threshold of Google, through predictive ranking, always changes,” recalls Jean-Baptiste Bessière.
For his part, to pass the quality threshold, Koray Tuğberk Gübür essentially recommends making short and informative sentences, producing a coherent text, highlighting the EAT or even providing authoritative sources in your field. More in depth, he advises for example to complete a single topic with all the details, even if they do not appear in the questions and documents of the competitors.
#quality #threshold #exist
Comments
Post a Comment