N-grams is a fancy term in Natural Language Processing and Understanding, which is what machines like Google use to read "tokenized" words.

They are basically 2 to 6-word phrases that are most commonly found in a body of text once tokenized.

The goal of this step is to identify the most commonly used words and phrases among the #1 ranking pages in your niche.

1. From the Dashboard, click on the big button labeled N-grams.
2. On the N-grams page, select 2 through 6 as the length and click Go
3. Now you’ll see a list of all the N-grams it has found. Your goal with this step is to copy and store all of the ones that stand out to you as unique to your niche and skip the generic ones such as “of the” or “to be” but make note of the ones that contain a word from your niche. For example, in the dog niche, the #1 n-gram was “your dog” which I thought was interesting, so I wrote that down. But the #1 was “of the” so I skipped it. You can also download them as a spreadsheet from SketchEngine and, just to go through that, delete the irrelevant ones.

I’d aim for at least 25 to 100 words to add to your list. Anything that’s been mentioned 10 times or more.

Having a hard time choosing which n-grams/words to include?

Download/export all the n-grams from SketchEngine and upload them to ChatGPT's Data Analysis plugin with this prompt:

Attached is a list of n-grams. Analyze this list. Give each word a relevance score based on itss relevance to the main topic of [insert your niche here]. Then return a table or csv of each word and its relevance score.