The basic goals of keyword research are pretty simple, to find the words and phrases people use when searching on engines like Google or Bing for your content, product or service.
While most available keyword data keyword data is for Google and Bing, research can apply to any search engine including video search like YouTube, specialty search engines like travel sites and app stores like a mobile device app store or the Chrome Web Store.
In a nutshell, research focuses on finding,
The words and phrases people search for the most (volume)
How many sites are optimized for keywords and phrases (competition)
How hard it will be to rank for a keyword term (difficulty)
The rarely found ideal is to find high search volume keywords with low competition or difficulty. More often we look for words and phrases with a difficulty that the website we’re working on can overcome or aspire to. We also seek lesser used words and phrases that contain more popular keywords.
Macro Search Patterns
Last year Comscore estimated that Google served 2,161,530,000,000 searches. That’s 5,922,000,000 per day and a 15% increase over 2012.
Google serves 181 countries ad 146 languages.
Each day 15% of the searches Google sees were never searched for before.
In 2012 Google reported it has seen 30 trillion URLs and crawls over 20 billion URLs per day.
Most searches are for pop culture and entertainment, trending news and events, sports and retail brands. You can view today’s trending queries on Google’s Hot Searches.
If you graph the search volume for all keywords, or the keywords targeted to a specific industry or market, the result will be a long tail chart.
Search Demand Curve
A small number of search terms get the bulk of the searches; this is the high tail. A sizable group of keywords and phrases receive a high to medium number of searches; this is the chunky middle. The vast majority of keywords receive only a few queries; this is the long tail.
High tail queries tend to be shorter, one to three words, and more general. Used cars is a good example. The chunk middle or mid tails consist of slightly more specific searches such as red used SUV. The long tail consists of longer and specific searches like red 2010 4 door Toyota Camry. From a marketing funnel perspective, high and mid tail keywords are good for awareness and brand building while long tail keywords provide the most convertible traffic.
When selecting keywords it’s important to take into account the goal of the person making a query. The keywords you use in your content should match what you want a visitor to do with the intent behind the keyword term.
Many searches are made to find and visit a particular website. If someone searches for Yahoo or IBM or Reddit, chances are high they want to visit those websites.
Another popular type of search is for information. This might looking for what movies are showing, the weather forecast or directions on a map. Another type of information query is typing in the name of a show or actor to find out more on sites like IMDB or Wikipedia.
Often people search for brands, makes, models, authors and a host f other keywords to research something that might, or might not, lead t a future purchase. A search for Cannon SLR could lead a person to several sites in search of features, price comparisons and other knowledge. The same person may also search for Nikon SLR and Olympus SLR so they can compare different cameras’ features and benefits.
When someone searches with the intent to buy something, sign-up for a subscription or make a specific conversion like downloading an eBook, we call that a transactional query. Transaction searches frequently come after completing research and have a specific item and vendor in mind.
Entities and Synonyms
Over the years search engines have become more concerned with the meaning behind search queries and less so about the specific words or phrases. This is because the intelligence within search engines has evolved and become more sophisticated. We are now in a transitional period where creating separate content for different search queries that all mean the same thing is becoming less effective because Google and Bing want to display the best results regardless of the actual words somebody types.
The word Google uses for something that lots of different searches probably refer to is an entity. Think of entities as people, places or things.
We aren’t fully there it’s a slow transition. That said, it’s well on its way.
Google Search Timeline
The evolution of Google search.
Google Keyword Planer & Bing Keyword Tool
Google and Bing both provide free keyword research tools. Google’s tool is a finite set of popular AdWords queries, searches for which people click on PPC ads. Bing’s keyword tool uses actual organic search data. The limiting factor for any keyword research tool is the finite space available in the database. Size and processing speed constraints make it impossible to include the vast majority of all search queries.
For many businesses a keyword researcher must look beyond keyword tool results. Intuition and judgment have to come into lay. Look at previous keyword traffic and other websites. Brainstorm phrases that may generate traffic even though they don’t appear in the tools.
An important example of long tail keyword research is for local search intent. Green Lake Thai Restaurant may not appear in a keyword tool, but Thai Restaurant definitely does. It’s a sound call to assume people are searching for Thai food in Green Lake.
When selecting keywords for local search, think about the different names everyone uses for geographic areas. I live in Seattle, specifically the overlap between Phinney Ridge and Greenwood. Were I creating a website targeted at my community I’d include Seattle, Green Lake, Phinney Ridge, and Greenwood as well as any other common names for my area and the surrounding neighborhoods.
Another example of specialized search is mobile. People searching from their phones often seek an immediate, nearby solution. Search engines customize mobile search results to reflect this. If you’re a local business having menus and inventories online can be a good way to capture mobile search traffic. Be sure to add additional SEO signals like your address and a map link.
As you conduct your keyword research you’ll come across several terms. Here is a glossary to help you understand what you’ll see.
Also known as the query, search phrase, search term and by other names, this is simply the query being researched.
The total number of queries made on Google.
Google reports average monthly searches and lets you select query matching by broad, phrase and exact. For SEO we prefer to use exact match search counts to keep expectations realistic.
The exact search query in word order. No more and no less.
The exact search query in word order plus any additional words. This number includes exact match searches.
Results that include any of the words in the query. If your query is Search For Results, the results include matches for just search and just Results. Pages that include only for will be excluded because for is a stop word, a commonly used word that gives no relevant content by itself.
Phrase and broad match counts can give researches insight into the overall size of the search market, but there will be noise, especially with broad match search numbers.
SEO Traffic (SEOT)
SEOT is the volume of traffic you may expect if you rank no. 1 for a given keyword. It is 42% of all searches, a useful approximation. Traditionally, we expect 60% to 70% of clicks to be on organic results and just above 40% of all clicks on the #1 result.
This is an aging generalization. Its accuracy keeps falling as Google changes how it presents search result placements and Google users become increasingly savvy. Today the reality is far more complex, but we don’t have a better formula so we keep using the old one.
The percentage of phrase-match searches to broad-match searches. This is useful for determining if the search query is used as a part of long tail searches.
A trend chart is a bar chart of the one year’s worth of monthly traffic volume Google provides in it’s Keyword Planner tool. This is a nice way to see obvious trends and seasonality.
AdWords Impressions (AWI)
Estimated number of times a PPC advertisement will be shown as the #1 rank for a particular keyword in Google. This is a generalization and, like SEO Traffic, should be taken with a grain of salt.
Majestic Title Competition (MSEOTC)
The number of pages in Majestic’s index that mention all of the words in a keyword term in the title of the page.
Majestic URL Competition (MSEOUC)
The number of ages in Majestic’s index which mention all of the words in a keyword term in the URL of the page.
Majestic Anchor Text Competition (MSEOAC)
The number of pages in Majestic’s index that mention the keyword term in the anchor text of an external, inbound link.
Majestic Domain Competition (MSEODC)
The number of pages in Majestic’s index that mention al the words in a keyword term in the domain name of the page
Majestic Title & Anchor Text Competition (TAComp)
The number of pages in Majestic’s index that contain the keyword term in both the page title and in the anchor text of an external, inbound link.
SEO Competition (SEOC)
The total number of webpages globally that mention a specific keyword term, in the same (phrase) word order, in Bing’s index.
Title Competition (SEOTC)
The total number of webpages globally that mention all of the words in a keyword term in the title of the page.
URL Competition (SEOUC)
The total number of webpages globally that mention a specific keyword term, in the same (phrase) word order in their URL, in Bing’s index.
This is the ratio of SEOTC to SEOC. A low score can indicate weak competition in a market.
SEO Value (SEOV)
The total monthly value of traffic that a website ranked #1 in Google for a keyword could potentially achieve. (SEOV = SEOT * AWCPC)
Average AdWords CPC (AWCPC)
An indication of what you might expect to pay to have your ad appear for a particular keyword in Google. CPC is cost per click.
AdWords CTR (AWCTR)
The percentage of people who click on the #1 ranked advertiser in Google (AWT) as a percentage of the total traffic for a keyword. CTR is click thru rate. AWT is AdWords traffic.
Keyword Effectiveness Index (KEI)
Total Searches / SEO Competition
This is a simple measure of keyword difficulty. It is only a quantity calculation and doesn’t take into account the quality of the SEO Competition.
Keyword Difficulty is a term you’ll come across quite often. Moz includes this in its toolset. They retrieve the top 10 rankings for any keyword, then assign that keyword a Difficulty Score based on the pages that currently rank for that word. More often researchers don’t have an actual formula. They eyeball the different metrics to give them an idea of which keywords will be the easiest or hardest to rank for.