How to apply data science to your SEO data

Blog banner

SEO is now so fiercely competitive that it’s no longer enough to rely on the standard techniques that everyone has mastered; you need to use data science techniques to unearth hidden opportunities that will give you the edge. This post offers a four-step framework for applying data science to your SEO data and transforming it into actionable marketing insights.

1. Decide on your data sources

Data insights will only ever be as good as your data sources. Where should you look?

Google Analytics and other descriptive analytics tools like Google Search Console, SemRush, Ahrefs, heatmaps and technical audit tools are strong contenders. However, as SEO gets more complex and integrates with other areas of digital marketing such as CRO, content marketing, CX management and ultimately sales, relying on one (or a few) of these solutions is no longer enough.

So how many data sources are necessary to get better SEO insights? The answer will depend on your current setup and post-adoption goals. Consider the areas where your visibility is limited and which data sources contain the answers you need. The next step will be to build a good harvesting engine/pipeline for those sources and to prepare your data for analysis.

2. Use data science to align SEO with other marketing initiatives

Your SEO gets stronger when backed by other marketing initiatives. A single team cannot optimise for all the 50+ search ranking factors without close collaboration with other specialists such as developers, UX designers, sales and customer support teams. Data science helps you figure out a universal set of SEO best practices every team can apply and adhere to.

To better understand what actions matter the most for your business, consider tracking the ever-changing relationships between the dependable and independent variables.

A variable is an event, a sales offer, a campaign or another activity that your company can measure. Data science allows you to pinpoint the relationships between different campaigns (or individual actions) executed and attribute their results to some KPIs (e.g. higher conversion rates).

To get a better understanding of how your SEO impacts other channels, consider capturing and analysing the following data:

  • Conversions and assisted conversions. The latter will help you identify the channels that don’t directly generate the conversions but play a part in the process. For instance, a customer discovered your website via organic search, browsed the products and later typed in the URL directly to make a purchase, or converted from a remarketing FB ad.
  • Top conversion paths. This data will give you more insights into how users interact with your website and other channels before becoming a lead or placing an order.

By gaining a deeper understanding of your customers’ journeys, you can create stronger alignment between all the marketing activities you deploy and attribute the results to individual campaigns with ease.

3. Focus on stories, not numbers

Apart from picking the right data sources/tools, you should also pay attention to the right metrics. Rapid growth in search traffic from Canada may seem like an SEO win, but is that traffic of any value for a business operating solely in the UK? The answer is clear.

Focus on tracking the metrics directly tied to specific KPIs – such as those reflecting conversions, repeat business, higher customer engagement etc. Make sure that you keep a strong focus on quantifiable, actionable metrics, not the vanity ones.    

In addition, it’s important to look beyond the SEO campaign numbers and dwell more on what drives those results. Insights are not just good data summaries. They are stories, explaining certain behaviours your customers are exhibiting and their correlation with your marketing campaigns.  

SemRush suggests focusing on the following metrics to create effective measurements:

What data should every marketer analyze to ensure their process helps them achieve their business goals?

4. Use data science techniques to visualise

Numbers stashed in spreadsheets can be hard to stomach for decision-makers. And by looking at your data hierarchically, you can miss an important story hidden between the lines.

Data visualisations can help you:

  • Accelerate knowledge discovery
  • Compare and contrast
  • Spot common trends and patterns
  • Digest large amounts of data at scale
  • Reveal questions that would otherwise be missed

Here’s a real-life example from our team. We leveraged data science during an SEO technical audit and received a lot of insights about the client’s website health and performance. That data told us all about the page authority, number of inbound/outbound links per page, rankings and a multitude of other factors. However, it didn’t provide a clear answer as to why some pages performed better in search results, while others lagged behind.

By visualising the website’s internal link structure and estimating the overall domain authority of each page on a 1-to-10 scale similar to Google, we could immediately see the areas for improvement and take proactive action.

Before

Visualising the website’s internal link structure - model output - Before optimisation

After

Visualising the website’s internal link structure - model output - After optimisation

On-page SEO optimisation is just one example of how marketers can combine visualisations with data science to gain better results. Visualisations help you make your SEO data even more actionable.

Ultimately, the goal of data science is to eliminate most of the guesswork from your SEO. Instead of presuming what will work and how a certain action impacts your goals, you switch to knowing what’s bringing you the results you need and how you can quantify your successes.  

Interested in using data science?

If you’d like to start applying data science to your marketing but aren’t sure where to start, give us a call on 02392 830281 or send us your details and we’ll call you.

George Karapalidis profile picture
George Karapalidis

George is an SEO Specialist and Data Scientist in the Portsmouth office. He has worked in E-Commerce and Digital Marketing across many industries, for and with companies all over Europe. Before joining Vertical Leap, George worked as Marketing Director for his own company, for which he managed to expand the company’s activities to 5 European countries. George has been creating websites for more than 10 years and he has in depth experience in designing and bringing optimised E-Commerce websites to market.

More articles by George
Related articles
Why data science is crucial for today's marketer: Expert interview

Expert interview: The importance of data science for today’s marketer

By Michelle Hill
A human face overlaid with a computer grid and multiple screenshots representing big data analytics

5 signs you should invest in data science

By George Karapalidis
Data cloud image to illustrate Data science for marketers (part 3): Predictive vs prescriptive analytics

Data science for marketers (part 3): Predictive vs prescriptive analytics

By George Karapalidis