What is the keyword selection process for on page SEO

Excellent question--probably the most important question in SEO, since keyword strategy drives everything else. The first thing I do for my clients is create their keyword strategy. That drives key decisions in their website designs, blog designs, and ends up driving how they prioritize what content to create on an ongoing basis.

Unfortunately there is no tool that will do it for you. You need human knowledge, creativity and good judgment.

Here's a process you can use:

1 - Brainstorm elemental keywords and phrases you think your target audience might use when searching to solve the problems you solve--e.g., "beer," "brew," "fresh," "cold," etc.

2 - Figure out how to categorize those elemental keywords into categories that can be combined using "formulas" for how to mix and match the categories, so you end up with many (often thousands) of possible search terms people might use--e.g., formula = "benefit + drink" applied to keywords gives you "fresh beer," "fresh brew," "cold beer," "cold brew," etc.

3 - Find how much traffic each search term gets, and create a short list of those terms that get at least some minimum amount of traffic that is worth any time or effort to rank for

4 - For each search term on the short list, look at its search results (make sure your own Google profile and search history doesn't influence the results) to get a sense of what the intent of the searchers is, and if the searchers are actually searching to solve the problem you solve; if so, keep the search term on the short list; if not, eliminate it from the short list--e.g., if searches for "brew" bring up mostly links to the macOS software called "brew" and not beer-related sites, then don't include "brew" in your short list

5 - For each search term still on the short list, starting with the most promising search terms based on traffic and relevance, figure out how competitive the current content is. In other words, figure out which first-page SERP ranking you think you can achieve. Usually you'll get beaten by major media sites and other highly authoritative sites, so go down the page until you find a domain and content you think you actually have a shot at beating. Then note which SERP position that is and what the competing content is. If you don't think you can get to #10 or better, prioritize that search term for later, so you don't waste your time chasing windmills.

6 - For any search term you think you can compete with for first-page rankings, figure out HOW your content will be better than the existing content you're trying to beat--e.g., various SEO mistakes they've made that you can do better on, usefulness of the content, timeliness of the content, more backlinks to the content, etc....

7 - Pick about a handful of the absolute most valuable search terms you could possibly want to get--based on high-quality, high-volume traffic--that are also probably too competitive for you to rank for initially, and mark those as "strategic" search terms

8 - Put together a plan for internal linking, backlinking programs, blog organization, and other sitewide SEO improvements that will help you eventually rank for these highly competitive search terms that you can't easily rank for with just a single page or blog post

9 - Once you have your top 10 search terms that are NOT "strategic" search terms, and have identified what your competing content is, and how to beat that content to achieve a certain first-page SERP ranking, then start going down that list and picking which term you're going to work on next; write the content, make sure it's on-page optimized, publish it, promote it, measure whether you achieved the target SERP ranking, figure out why you didn't, fix the problems, and rinse and repeat until you achieve the SERP ranking goal; then continue to the next search term on the list, and the next, and the next....