How to Get Started with Keyword Research

Keyword research is fundamental to any SEO or SEM campaign. But how do you go about selecting keywords?

Keyword selection is all about trying to find the words and phrases your potential customers are using as search queries, then leveraging those words organically or targeting them in pay-per-click campaigns. Keywords can be defined by two categories: broad and narrow. These are sometimes referred to as short-tail and long-tail, respectively. Generally speaking, it is often harder to achieve higher organic rankings for broad terms than for narrow terms. While narrow terms typically receive a lesser volume of searches per month, being able to attract and convert visitors on a combination of many narrow search phrases will result in better quality traffic and ultimately should generate more sales than the broad search traffic.

Keyword Research: First Steps

There are a number of tools with which you can get started. First, start with a broad term. In this example, I'll use the word "cookie." One method of searching for additional narrow phrases is to use search engines. For example, when entering the word "cookie" as a search query, I get the following on Google: The suggested phrases that appear are actual search queries used by searchers. Typically these are more popular phrases. While this method can be time consuming and tedious for research, often you can discover ideas for new content, then perform deeper keyword research on the suggested phrase you discovered. Another method for researching narrow terms is to use a keyword research tool, like the tools at SEMrush. To get started, simply type the word or phrase into the search box and SEMrush does the rest. The SEMrush keyword tool produces two lists. The first, on the left, shows a list of 33,521 different phrases that contain the word "cookie." In this view, we only see four columns. Clicking on the Full Report link shows seven columns. In order, the columns contain the keyword or keyword phrase discovered, the average number (volume) of searches per month, the Cost-Per-Click to bid on that phrase for a pay-per-click (PPC) campaign, a ratio representing competition of PPC advertisers for the phrase, the number of results when searched upon in Google, and a trend ratio of the volume of searches over the last 12 months. All this data can be exported to a CSV or Excel spreadsheet for analyzing. The box on the right shows similar data and can also be exported. However, the 1,947 keyword terms returned in the right box are related keywords. Related terms may — but not necessarily — contain the initial phrase searched upon. Often, related keywords are phrases that were typed in by searchers while refining their searches. Using a tool like SEMrush, you can also click one of the terms from the list to perform keyword research using that term as the root phrase.

Analyzing Your Keywords for Organic Search

As you can see, keyword phrase lists can get pretty long, pretty quickly. Once you have your words, what do you do with them? How do you know which words are best to use? For starters, use your common sense and read through the lists. If you're running a cookie recipe site, you likely don't want terms like "cookie monster" or "girl scout cookies." At least, you wouldn't them for the recipe portion of your site. However, those phrases might be neat to target for content marketing. You can also use SEMrush filtering tools to remove words from your entire keyword list. Remove all the phrases you're not going to target. When you're done, remove the phrases you currently rank well for. Remember, search engines rank pages — not sites. If you have a page that ranks well for a certain term, don't compete against yourself by using that term on another page. Target new phrases on each page. There are many ways you can target phrases. Popularity is one of the more common. After all, the higher volume of searches performed in a month, the more popular that phrase is, right? But you should also take into account how many results are returned on the phrases you're targeting. If you don't rank at all, you'll be the new fish in the pond of results. Will you be one of millions, tens of millions or even more? Many organizations refer to a metric known as KEI, the keyword effectiveness index. Simply put, the KEI is a ratio of the number of results for a given term with relation to the number of searches performed in a month. The formula for calculating KEI has been debated by SEOs. But the generally accepted formula has been the number of searches squared divided by the number of results.

(# of monthly searches)² # of results returned

This formula rewards terms that have a high search volume and a low number of results. The higher the resultant metric, the more effective the keyword should be. This formula, however, is far from perfect. It does not take into account spammy or unoptimized result pages. It also doesn't take into account any of the personalizing of search. Rich snippets are being called by some SEOs the "magic bullet" for search. Eric Schmidt already went on record saying that sites with Authorship markup will rank higher than those without. However, this formula still gives you an indication on where to start sorting your keywords.

Choose Your Words Wisely

Once you have your word lists filtered and sorted, get started on that copy. Remember to write your copy naturally. It should be easy for your visitors to read. Never write toward keywords and rankings. Always write for your users. Let keywords be your guide. Let variations be your companion. Use your newly discovered keywords and phrases wisely. Remember you need not target specific phrases only for higher on-page performance. Use your new keyword lists to glean new ideas for blog posts, supplemental pages and other content marketing initiatives. While some of the discovered phrases might not be an exact fit, people are actively searching for these words.

Author bio:

Thom Craver is an international speaker, digital strategist, author and adjunct professor. He specializes in SEO and Web analytics. His last article for SEMrush was "Redirects: The Why and How."