Google’s MUM Algorithm Update: What Does It Mean For SEO In 2022?
- Imo Gee
- May 3, 2022
- 3 min read
The Google MUM (Multitask Unified Model) algorithm update was announced at the Google 1/O event in May 2021. So why am I talking about it now? MUM is a relatively new AI from Google which aims to answer complex questions and will be advancing and learning over the next couple of years.
Pandu Nayak, Vice President of Search at Google
So what is MUM?
To get to the point, it’s Google’s new-ish AI-powered search algorithm that is focusing on interpreting and understanding not just written content but also images, videos and podcasts across multiple languages to deliver the best results for the user. It aims to solve users from needing to type out many queries for the answer they require. Because this new technology can understand various languages, the user should receive better results through a transfer of worldwide knowledge.
My predictions on what this mean for SEO
From my understanding, this will mean a push on thorough, long-form content. The algorithm essentially looks for expert information. This made me think of the E-A-T concept from Google‘s Search Quality Guidelines, as I mentioned in my Blue Array Technical SEO course review blog post.

As it focuses to expand on the understanding of video, audio and imagery. I expect that content that includes imagery, videos, a thorough internal linking plan and helpful external links will be rewarded within the SERPs. Essentially in-depth content which displays expertise. In Google's blog post they stated that this algorithm will be able to point the user to other helpful resources like articles, videos and images. As the SERPs gain more competition, you want Google to see you have all the information in one place. I also expect SERPs to gain some new features. Featured snippets, knowledge panels and video carousels have all become well integrated within the search results pages. With the new understanding of imagery, audio and content, I expect Google (as always) to be testing new features within the SERPs.
With it's improvement in understanding imagery, I expect an increase in the use of tools like Google Lens amongst users in the next couple of years as the AI advances.

A lot of people were also early adopters of the smart speakers, at one point even Spotify teamed up with Google to give out Google Nest Speakers. But why would they do this? It has been theorised a lot that Google and other tech giants are listening to you to improve their products. With this, audio interpretation can only get better and the results that will feedback to the user will be more helpful and take them on a journey to find all their answers/needs.
As MUM is much more capable to identify the relationships between the keywords used within a phrase, the SERPs will improve for more complex questions and long-tail keywords. With its deeper understanding of language it can help transfer knowledge across the world to the user.
Pandu Nayak, Vice President of Search at Google
MUM Summary:
It is multimodal - meaning it can understand information across text and images. In the future, it will expand to video and audio.
It is trained in 75 different languages and can generate its own.
Acquires deep world knowledge to collect and translate for the end-user
1000x more powerful than BERT
But wait, what is BERT?
BERT was a previous Google search query algorithm launched in 2018. This algorithm focused on user intent, processing the keywords entered into Google in relation to the other words and phrases within it, rather than one by one. The example below from Google shows how the algorithm can now match user intent better. Queries used to interpret 'adult' with 'young adult', now it can understand that 'young adult' was taken out of context and wouldn't be the correct item the user is looking for. Resulting in the SERPs now displaying a more helpful result.

What are your thoughts?
Do you have any extra thoughts or comments on the MUM algorithm update? Have you already seen movement in your data which could be caused by it? Drop a comment or message on the contact page.
Comments