What Google’s ‘Bedlam’ Algorithm Update Means for Your Business

Brian Hart  |

In November 2019, panic ensued among SEO professionals. Without warning, Google changed its three-map pack algorithm, wreaking havoc on search result rankings. Two days later, the organic search algorithm also changed. SEO pros described the days that followed as “bedlam.” Formerly high-ranking businesses fell to mediocre or poor placements for no apparent reason.

By the end of the week, many businesses saw their rankings restored to their former glory, but not all. In response to the chaos, Google released a rare statement explaining that it had begun using “neural matching” to generate local search results.

This statement gives a rare glimpse into the black box of Google’s algorithm programming. It tells us that Google has made a fundamental shift in how its algorithms determine rankings. Before the change, it traditionally relied on keywords and businesses’ primary categorization to determine relevance, while also factoring in location and the prominence of a website or its owner.

Now, Google’s algorithm improves on the relevance component using machine learning (ML) not just to read the keywords, but to interpret the subtext of search terms. Google’s algorithm is now far from the stereotypical robot, taking everything 100 percent literally. It can now “understand” nuances of language and guess what is meant even when the searcher can’t quite describe it. As a result, the user receives search results that are far more relevant to his or her intention.

For business owners, the new algorithm presents a golden opportunity as long as they manage a few drawbacks, such as increased spam. By playing the game right, business owners can enhance web traffic, convert more leads and use the internet to grow their businesses like never before.

What is neural matching?

Google explains that neural matching helps its algorithms better understand how words relate to concepts. One Google engineer referred to it as a “super-synonym system.” The algorithm stores all the synonyms of words, and, when presented with a keyword search, it can combine all the possible meanings of each word and synthesize them to create multiple interpretations of the user’s possible meaning.

Though this may sound like an English professor with superhuman abilities, the process is much more mechanical. ML applications learn by repetition. They try one thing, then another, repeating what works and eliminating what fails.

Unlike a human, AI doesn’t truly understand the meanings behind the results it provides. It learns by rote. Because of massive computing power, it can use these rote lessons to dispense answers at speeds inconceivable for a human brain.

A neural matching example

To illustrate this concept, Google provided the example of the query, “Why does my TV look strange?” Surprisingly, one possible answer is the “soap opera effect,” which refers to the lifelike quality of HDTV that some viewers find similar to the videotaped look of soap operas.

Without ML, an algorithm couldn’t make this connection. It isn’t literal enough. Yet many searchers use this query without ever having heard of the soap opera effect. The new algorithm can suggest this answer even if the searcher doesn’t know how to express it. The soap opera effect is just one of many answers the algorithm can produce for this particular query.

How businesses can benefit

Businesses that provide multiple services can increase their local Google rankings. Before the update, a business’s primary category in Google My Business largely determined when it appeared in searches. A law firm with a primary business of personal injury appeared in searches for that type of attorney, but if the firm also practiced real estate law, it was less likely to show up in searches for real estate lawyers. Neural matching allows Google to better recognize all services, especially if businesses keep their website and My Business profile updated and detailed.

For this hypothetical law firm, the SEO strategy is clear: create more content for its real estate business, so it appears in local real estate law firm searches. Many businesses today have multiple subcategories. By creating content based on these subcategories for the new algorithm to search, businesses can appear in more local searches, receive more inquiries and win more business.

Users have highlighted spam as one drawback of the update. To reduce spam, report it to Google through the “suggest an edit” or “remove this place” feature. Common spam includes phony reviews, wrong locations and multiple listings for one business.

Businesses have a golden opportunity to enhance their visibility on the web. By creating content that maximizes potential neural matches, businesses appear in more searches. Companies with multiple lines can now create unique content for each business segment and appear in different types of searches. Despite the temporary uproar the new algorithm caused, businesses that design content with neural matching in mind stand to gain increased exposure and more business opportunities.

Brian Hart is founder and president of Flackable, a national public relations agency serving financial and professional services firms.

Original article at Flackable.com

_____

Equities Contributor: Brian Hart

Source: Equities News

DISCLOSURE: The views and opinions expressed in this article are those of the authors, and do not necessarily represent the views of equities.com. Readers should not consider statements made by the author as formal recommendations and should consult their financial advisor before making any investment decisions. To read our full disclosure, please go to: http://www.equities.com/disclaimer. The author of this article, or a firm that employs the author, is a holder of the following securities mentioned in this article : None

Watchlist

Symbol Last Price Change % Change
AAPL

     
AMZN

     
HD

     
JPM

     
IBM

     
BA

     
WMT

     
DIS

     
GOOG

     
XOM

     
BRK.A

     
FB

     
JNJ

     
WFC

     
T

     
NFLX

     
TSLA

     
V

     
UNH

     
PG