3 Data Science Methods for SEO

September 22, 2020 - 2  min reading time - by Julie Quintard
Accueil > Infographics > 3 Data Science Methods for SEO

At a time when data science is becoming more and more present in the habits of marketers and companies, OnCrawl would like to show how data science can be a real game changer for SEO. This infographics presents 3 ways to use data science to take your SEO strategy to the next level.


Feel free to share on your website



1. Prediction

What would you do if you could look into the future and see your SEO traffic for the next 6 months? I bet you would make some changes to your strategy!

With data science, it is now possible to forecast keyword performance tendencies, future long-tail trends, and even organic traffic. Using the Facebook Prophet algorithm and your Google Search Console data, you can easily build a predictive model to forecast Google hits.

Great for:

  • Reassuring your C-suite of the value of an investment
  • Balancing expenses between organic and paid channels

2. Text Generation

This is the dream of many SEO professionals: to get SEO-optimized quality content at zero effort.

Data science now makes it possible to automatically generate text for a page. Using the deep learning framework PyTorch and GPT-2, you can train this machine learning algorithm to learn your lexical field and writing style and automatically generate content on a given topic.

Great for:

  • Creating anchors for internal linking
  • Mass-creating variants of title tags
  • Mass-creating variants of meta descriptions
  • Building rich product pages

3. Anomaly Detection

Immediately be alerted to drops and peaks in your site’s SEO performance, based on your favorite KPI metrics? It’s now possible!

Using the Robust Random Cut Forest (RRCF) algorithm, you will be able to detect any change on given KPIs that is unusual and requires your attention. This algorithm will even take seasonal events into account, along with gradual changes to the website over time. Examining anomalies can also reveal whether certain metrics are key to a website’s SEO.

Great for:

  • Knowing if your site is affected by an algorithm update — before a human can identify the change
  • Detecting technical problems before they create a lasting impact
  • Identifying real-world events that have an impact on your site
  • Automating your SEO monitoring

All these machine learning algorithms are ready-to-use on the OnCrawl Labs.

Julie is working as a Marketing Manager at OnCrawl. She is a real marketing and digital passionate and regularly writes articles about SEO and OnCrawl news. You can reach her on Twitter.
Related subjects: