Announced by Google at their most recent 2021 I/O event, they define MUM as a new milestone for understanding information – and it’s going to change the game.
While it’s still in its early stages, Google’s new Multitask Unified Model has the potential to revolutionize the way we use search engines and how companies create and perform SEO strategies. The goal is to provide a faster, more satisfying, and accurate answer - almost as if you have an expert right at your fingertips.
Imagine getting incredibly accurate results for a search such as:
Typically, this sort of search would yield little to no results that would truly answer your questions in a well-rounded manner. You may need to break down the question and perform separate searches for weather conditions, elevations, appropriate gear, and health conditions. You think to yourself, “if only I could just ask a hiking expert,” who could answer this right away and give you exactly what you needed. Similar to how we enjoy Reddit, MUM can do just that, too.
MUM can transform the way you complete complex tasks by using the T5 text-to-text framework, where the input and output are always text strings. It is 1000X more powerful than BERT (Bidirectional Encoder Representations),which is a neural network-based technique for natural language processing (NLP) introduced and open-sourced by Google in 2018.
What does this mean?
This technology will change the way we use search engines and affect SEO and keyword strategies, including breaking language barriers that can be incredibly hindering towards information gathering and sharing. With more accurate and specific results, your audience will have less searches and fewer clicks before completing their search, thus increasing competition towards getting your products and services in front of your audience first.
We’re always adapting to new technology and playing catch-up with our fast-moving digital world, and this is just one of the latest examples of how we’re improving communication and information interpretation.