How Market Brew Predicts Search Rankings
With a strongly correlated model, you can predict what will happen next.
Introduction to Search Engine Models:
Market Brew search engine modeling technology allows users to define, create, and deploy statistical replicas of any search engine environment. It does this by starting with a "generic" search engine model, complete with all the modern families of algorithms, representing everything from on-page to off-page ranking factors. Each part of the search engine model's biases/weights are then finely tuned by an artificial intelligence process called Particle Swarm Optimization.
At the conclusion of this A.I. process, the "generic" search engine model has been transformed into a very useful and highly calibrated replica of the real thing.
- How does Market Brew work? Think of it as a search engine that can calibrate its own settings using machine learning to behave like any search engine you want.
- Users can test their website changes in the model and can predict how their actual ranking results will be affected, months before those changes show up in their rank trackers.
Users can then deploy their changes on the search engine models in order to predict how the target search engine environment (TSEE) will react when those changes are introduced. As a result, users can also copy these biases/weights from their production search engine model to various testing search engine models, in order to conduct A/B testing prior to the production deployment of those changes.
Track Search Engine Algorithmic Updates:
One of the hardest parts of SEO today is understanding how algorithmic updates affect your specific search results. Market Brew automatically detects when the models diverge from the target search engine environment. When this happens, Market Brew re-calibrates itself, and because the search engine models store their bias/weight settings, users are able to compare the settings before vs. after to determine which parts of the search engine model are now more or less emphasized.
- Users track and characterize algorithmic updates for any search engine by comparing bias/weight settings on each model.
Our Patented "Search Engine Model" Approach:
- Has been regression tested each week for over a decade, and is constantly updated based on the the latest macro algorithmic updates.
- Is automatically self-calibrating, and like a successful poker player with a positive expected value on every hand, the search engine model forms a basis for a real-time environment that can analyze ranking distance vs. statistical gap analysis and sort its TODO list by the most traffic or revenue per work unit. This gives the user a positive expected value on every optimization they make, starting with the highest return.
- Is fully customizable, allowing each user to mimic ANY target search engine environment (TSEE), like the US version of Google. They can upload their own metrics like revenue, conversions, and more to make the simulations as locally accurate as possible.
Market Brew's Search Engine Model Is Self-Calibrating, To Any Search Engine Environment.
I think what they're doing is really smart.
Steve Hoffman, Angel Investor & Captain of Founders Space, Inc. Magazine's Top 10 Incubators List
This technology is gold. Of all of the technologies in the SEO sector, Market Brew is the only one that I consider the 'right' approach.
Bryon Sheffield, Senior Director at FICO
White Papers about the Technology
Implementing ARTIFICIAL INTELLIGENCE to Attack SEO
Market Brew was selected unanimously as the best startup by the panel of judges. Market Brew's usage of big data with analytics to deliver search metrics to CMOs was very impressive.
Tom Plunkett, Lead Author: Oracle Big Data Handbook, on seeing Market Brew as one of the top big data companies out of the Silicon Valley
There are some key technical innovations that enable Market Brew to accurately and efficiently predict search rankings for the major search engines:
Families of Penalty Algorithms
A sample of core search algorithms that are modeled.
- Market Brew includes families of penalty algorithms that accurately model the algorithmic penalties applied by search engines.
- When search engines adjust their algorithms, sites that are close to the penalty thresholds are often dramatically impacted.
- Market Brew keeps you in the “safe zone” where the penalties do not significantly impact your search performance, and where you are much less likely to be penalized by future updates.
Neural Network Crawlers
You tell the system what target search engine environment (TSEE) to inspect, and market brew does the rest.
- Market Brew includes a neural network algorithm that automatically crawls the part of the Internet that affects your site.
- Because of this smart-sampling, the self-learning crawler is lightning fast, typically refreshing your rankings within 100 minutes.
- Combined with using industry standard data, this lets us accurately model both On-Page and Off-Page metrics that affect your ranking.
Transparent Search Engine
Fully Transparent Search Engine Model.
- Our patented technology exposes the key values and ratios that drive all search engine algorithms.
- We integrate relevance, link flow, and algorithmic penalties to determine a query score just like all other search engines do.
- This lets us accurately model the distances between ranked pages and run predictive algorithms that identify the highest ROI opportunities.
- Because the search engine model is transparent, every part of the search engine model is accessible. Users can fix, test, and verify changes easily and rapidly.