Google’s New Neural Matching Update in Local Search. What is it?

Sometimes, SEO can seem like trying to shepherd cats. Just when you think you’ve found a calm strait on these usually roiling waters, Google throws a rock your way. As we all know, Google is in a constant state of both revolution and evolution. As frustrating as it may be for digital marketing agencies and SMEs alike, in order to ensure better quality searches for users Google is constantly implementing new algorithms in its never-ending quest to bring users the most relevant entries on Search Engine Results Pages (SERPs).

Last week, Google confirmed via its Search Liaison Twitter account what many of us have long suspected. That change is afoot in the world of local searches. The neural matching that the search giant implemented last year has extended to local searches. Readers can take a look at the whole thread on Twitter below;

 

In early November, we began making use of neural matching as part of the process of generating local search results. Neural matching allows us to better understand how words are related to concepts, as explained more here: https://t.co/ShQm7g9CvN

— Google SearchLiaison (@searchliaison) December 2, 2019

 

Agencies and on-the-ball small business owners may have already noticed the effects of this update, which has been unofficially named the “Bedlam Update” by local SEO expert Joy Hawkins. And while not all have reported positive results at present, the update has some interesting and potentially exciting implications in an era where over 45% of all Google searches are local searches. 

Unprecedented Reveal

The reveal of the Bedlam Update is unprecedented. Not because neural matching is brand new, but because Google is rarely this forthcoming about sharing the internal workings of their algorithms. Which, when you think about it, is just good practice on their part. The more they divulge about how any given algorithms work, the more vulnerable they make themselves to unscrupulous parties who will attempt to game the algorithm rather than provide value for the user. 

What’s refreshing about this reveal is that it doesn’t imply a set of hoops that ranking-hungry businesses will have to jump through. Google is simply detailing a bold step toward better quality searches through neural matching.

What is Neural Matching?

As stated previously, neural matching has been a part of SEO for a little over a year now, but what does it mean. Well, it’s all a matter of semantics. Quite literally, as it happens. Neural matching is all about helping search engines understand concepts rather than words. While search engines have historically relied on keywords to bring their users relevant results, neural matching parses not just searched terms but searched terms and their synonyms. So they won’t be precluded from a potentially high-quality match just because it didn’t happen to have the right keywords. 

How does this work in terms of local searches?

What does neural matching mean for local searches? In theory, it could be a colossal boon!

Until now, the way Google has worked has created some bottlenecking of results for some local businesses. All it has been able to use to rank businesses in SERPs is their names and the sparse data available in descriptions. Needless to say, this has disproportionately favoured businesses that happen to feature the function of their businesses in their business names. 

Neural matching aims to help users find what they’re looking for rather than marching search terms to keywords within a geofenced area. 

Global multilingual rollout

While the Bedlam Update is concerned with local searches, its rollout is huge in scope. When it was rolled out in November, it affected searches all over the world in a wide array of languages. A rollout of this scale implies great confidence in the update’s ability to provide higher value searches, even if the current state of SERPs doesn’t necessarily reflect that. 

Is this the same thing as BERT?

If you read our blog on the rollout of Bidirectional Encoder Representations from Transformers of BERT in October, you’ll likely see some similarity between these developments. However, while this Neural Matching Update and BERT have a very similar endgame, Google’s Danny Sullivan insists that they are not the same thing. 

However, both are geared towards helping search engines to better understand the people who use them rather than the other way around. This should, then, be encouraging news for businesses and agencies alike. Both the Bedlam update and BERT are geared towards helping Google users find high-quality relevant content. So if you’re already making efforts to be both, the combination can only work in your favour. 

In the same way that BERT supplements Rankbrain rather than supplanting it, the Bedlam Update will coexist alongside BERT. 

With algorithms working harder to leverage semantics to help users find relevant content, this is especially useful for businesses who want to find success in their local markets but whose business names were not chosen with SEO-friendliness in mind (like those who opened the door long before search engines, or even the internet, existed).

Previous algorithms lent an edge to those who employed keyword stuffing in business names, but a greater reliance on neural matching will help discourage this practice. 

Preparing for bedlam: What businesses need to do

It’s entirely understandable that agencies, their clients and small businesses who like to keep their SEO in-house might wonder how best to prepare for the ongoing impact this update will have on their pages. 

As with BERT’s announcement in October, whenever a new algorithm is announced, agencies and businesses can start panicking about slipping in rankings as the system fluctuates and the algorithm finds its footing. Indeed, many small business owners are already growing frustrated that they’re slipping in rankings (in some cases losing ground to the exact kind of spammy content that the update is intended to avoid). Unfortunately, this kind of flux is to be expected while the AI and the teams behind it get to grips with its proper implementation.  

Google’s official advice is much the same as its advice on BERT. Essentially, Keep Calm and Carry on with your existing strategy.

 

The use of neural matching in local search doesn’t require any changes on behalf of businesses. Those looking to succeed should continue to follow the fundamental advice we offer here: https://t.co/tPkyuyMjsP

— Google SearchLiaison (@searchliaison) December 2, 2019

 

While neither a small business or an agency like being told to sit on their hands while their page rankings jitterbug up and down SERPs, the truth is that this temporary state of flux needs to be ridden out.

The best things that businesses and agencies can do is stick to known best practices like ensuring NAP data is up to date, building links with trusted local sources and directories, incentivising customer reviews and turning out high-value content on a regular basis. 

By all means, keep an eye on local SERP results and if you happen to see any decidedly spammy or low-value content like stuffed keywords or hidden text, report it. Use “suggest an edit” and then “remove this place”.  

Whatever you do, resist the urge to commit any cardinal SEO sins in the hope of climbing up the SERPs while the algorithm finds its feet. Don’t mistake the current volatility as a tacit admission that Google has begun rewarding spammy SEO tactics. 

Have faith that when the dust settles, the spam will be consigned to its proper place and the digital cream will float to the top once again.