Insights
Published:
July 28, 2026
Author:
This is some text inside of a div block.

Case Study: Developing a Web Data Scoring Pipeline to Support GEO Evaluation

Share:
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.This is some text inside of a div block.

Creating a tool to gain insights to evaluate brand visibility across AI content.

Working with Evolved, Data Scientists from the Hartree Centre North East Hub created a tool to help provide insights to evaluate brand visibility across large language model generated content.

Challenge

Evolved is a digital marketing agency that serves clients across the UK and Europe, helping businesses to grow through a range of digital public relations and search engine optimisation (SEO) services. As consumers increasingly turn to large language models (LLMs) such as ChatGPT and Gemini, rather than search engines such as Google, Evolved have now moved into the emerging area of generative engine optimisation (GEO) that improves client brand’s visibility in LLM generated content. The goal of the project was to develop reliable ways to evaluate and explain brand visibility within LLM results, investigating how the LLM uses online sources to retrieve information for users.

Approach

The Hartree Centre North-East Hub team worked with Evolved to explore a web data scraping and scoring pipeline. This utilised query expansion to broaden search results to related fields around the initial query a user might input, and explored a variety of innovative scoring methods, combining occurrences of client and competitor brands with information on the semantic relevance of associated web content. Through exploring topics across multiple market sectors and correlating scoring metrics with LLM response market share data from Evolved’s own state of the art AI visibility tools, key indicators of brand performance in LLM responses were identified.

Benefits

The code deliverable and report produced by Hartree Centre North-East Hub team have provided Evolved with insights into the features that determine brand visibility in LLM generated content. This will enable Evolved to integrate new proof-of-concept features into their GEO tools so that LLM visibility reports explain as well as evaluate brand performance. This will enable Evolved to give evidence-based advice on steps to increase brand visibility, maintaining a competitive advantage in the rapidly evolving SEO and GEO landscape.

"Working with Data Scientists from the Hartree Centre North East Hub has been really helpful. I have definitely learnt a lot. The information they have put together, and the process they have put together will put us in good stead moving forward. I am really appreciative of what they have been able to achieve in the time that they have had. It is difficult for small businesses to get this type of fully funded support, and we are grateful to have been successful in recieving it. Thank you to everyone that has been involved in the project."

- David Watts, Evolved

Further Information

This work was completed as part of one of our collaborative data projects. The projects are up to 12 weeks in duration and give you access to a wide range of expertise across our team of data scientists and data engineers. We will work alongside your team to scope your data science or engineering project, build a prototype solution, and explore options to deploy it within your organisation. You can learn more about them on our webpage here.

If you would like to learn more about the Hartree Centre North East Hub or our collaborative data projects, please get in touch with us at: hello@hartreenortheast.uk