Reverse Engineering Google’s Algorithms with Market Brew

At the Advanced Search Summit last week, Brewco CTO Scott Stouffer presented his journey in developing AI to reverse engineer Google’s search engine algorithms. What Google isn’t telling you, Market Brew, Scott’s creation, can and will. Scott’s presentation, which you can view here, tells the origin story of Market Brew.

After an initial career building high-scalability systems for the financial world, Scott was hired by an early SEO firm to build a system for indexing search results on a nationwide basis. He went on to parlay his experience into building one of the Internet’s earliest and most successful link networks, gaming the algorithms Google used at the time to drive substantial revenues to his customers.

But Scott realized that this was not a good business model. Google would make changes to their algorithm, and Scott would reengineer his, again and again, until he grew tired of the cat and mouse game he was playing. Scott set out to create a different way of doing things, using search engine modeling as a new way to understand how to rank best on Google and other search engines.

He built a reference search engine that incorporated all of the key algorithmic and rule-based pillars of Google at the time, and he and his colleagues would hand adjust the settings and weights of his search engine to develop highly accurate models of how their search engine would rank sites for a given keyword. As they released new algorithms through their major updates, Scott and team added these factors into his search engine.

When the requirements of adjusting parameters become too difficult, due to both complexity and Google’s increasing use of AI systems, Scott adopted an AI approach using a form of genetic algorithms, particle swarm optimization, to machine learn the key parameters that determine rankings for a given search on a given search engine. And today’s Market Brew was born.

It would be impossible to create Market Brew today, or for anyone else to craft a system like it. Separate from the patent protections the system enjoys, it is only because Market Brew has been adding to its model, Google algorithm update by Google algorithm update, that it can continue to adapt and adjust to new algorithm updates like the current Google updates.

How did this all come about? Why is AI so essential to figure out best SEO practices? Check out Scott’s video to learn more.