Automatic grouping with keyword planner's 'refined keywords' update

Automatic grouping with keyword planner's 'refined keywords' update

Keyword research is a vital part of any SEM & SEO strategy, but the time required to manually sift through all related keywords to understand the semantic groups – themes, subthemes, brands, etc. – can be a massive time sink.

Headshot of Luke Ashmore-Delany

Published

18 August 2020

Categories

Acquisition

SEO

Luckily (and quietly!) Google has recently updated their Keyword Planner tool to include a beta feature called “Refine Keywords.”
This feature uses Google’s learning algorithms to understand the type of keyword/s being entered automatically, then sort them into a number of different categories to better understand which keywords can be aligned with which brands, product features, industries, and more.

The ‘Refine Keywords’ tool

Where to find it

To use this Keyword Planner feature, you must use a Google Ads account with Expert Mode enabled instead of Smart Mode.

Then, access the “Tools” menu in the top right of the interface and select “Keyword Planner.”

Once in the tool, click the “Discover new keywords” option and enter one or more keywords to begin your keyword research with. For our example here, we’ll use the term “smart watches”.

On the right-hand side of the interface, you’ll now see the “Refine keywords” tab. Here, you can see how Google has categorised the related keywords. For our research, Google has recognised “smart watches” as both a fashion and technology item, so we can see groups like “gender” as well as “technology” and “device”:

An example of the keyword categories for the keyword “smartwatch”

Expanding out the Brands tab, for example, allows us to choose specific keyword groups to include or exclude based on the brand that they contain:

Tick/untick individual items to add or remove them from your keyword research

The same keyword can be in multiple categories simultaneously. If your original keyword entered is too narrow or niche, you’ll only get the Brand/Non-Brand option, or if volume/related terms are too sparse, you’ll be left without results.

Advantages & Limitations

Beyond in-interface manipulation, this feature is also present when downloading keyword reports, which is a great boon as you can sort the information into your preferred data application like Excel or a visualisation program like Tableau or PowerBI. From there, it’s simple to see which competitors have the highest/lowest presence and difficulty, as well as perform calculations to see where the cheapest bidding opportunities lie.

Despite the clear advantages, the refined keywords are far from perfect and often miscategorise or split out groupings incorrectly. For example, both “iPhone” and “ios” are discrete categories in our research above, or “watch” was placed in “Other” instead of “Brands”.

That said, especially using the spreadsheet download, it’s a lot easier to review Google’s categorisations and fix whatever errors are present than categorising all of the terms from scratch.

The verdict ...

This feature is in beta and naturally doesn’t cover the full breadth of its potential. Still, it represents a very exciting step in Google’s ability to granularly understand and categorise search data to help advertisers and optimisers understand the lay of the land.

For SEM, this can help provide extremely valuable data on competitor activity and the difficulty of bidding. At the same time, SEOers can easily save time when categorising keywords when trying to understand the rankings of their site and competitors.

The dog and bone.
Subscribe and be the first to hear about news and events.
Headshot of Luke Ashmore-Delany

Written by

Luke Ashmore-Delaney
The dog and bone.
Subscribe and be the first to hear about news and events.
View our last posts

Indago Acquired by Nunn Media – October 2024

Gary Nissim - 2 min read
Google AI Overviews funny parrot example

An Overview of ‘AI Overviews’

Lewis Torossian - 9 min read
Realistic robot pondering how to steal traffic with GA4 logo

How to Identify AI-Referred Traffic in Google Analytics 4

Preet Singh - 5 min read
Team members thinking together
Get in touch
Ready to get the ball rolling? Drop us a line.
First name*
Last name*
Email address*
Phone number*
Company*
Your message