For some businesses it would take too much time and too many resources away from their business to designate someone in-house to learn all there is to know about Internet marketing as a whole. This is part of the many reasons why people turn to an Internet marketing agency for help. Not only is it expected of the Internet marketing agency to have the know-how (SEO, PPC, social media management, etc.) to get the job done, but it is also expected that they have developed efficient AND effective processes to reduce waste of time and other resources. Though inefficiencies are an unfortunate fact of business life, there are ways that you can better organize a usually time-consuming process: keyword research. Don’t listen to Col. Jessop over there; you can do it, and here are some pointers to help you get started.
What is Keyword Research?
If you haven’t already, I implore you to read SEO Moz’ Beginners Guide to SEO (particularly the section about keyword research). Not only can you find some helpful background on keyword research in this guide; you can also find some other helpful tips on optimizing your site. In a nutshell, keyword research is the process in which you research what keywords are most appropriate to your site (and its individual pages) that can be used to help your website (or web pages) stand out in search engine results pages (SERPs). Depending on how many pages you’re optimizing, this could take a few minutes to several hours. Not only does this task require your individual attention for a good chunk of your time, but it also often requires a lot of data analysis. For the sake of (attempted) brevity, I’m going to skim through the introduction to keyword research so we can get down to the keyword research process.
Disclaimer: Gone are the days that you can try to optimize your entire site (and all its pages) for one keyword and expect to rank #1 in Google for years (and bring in tons of visits). Keyword research (in the following steps) is not intended to help you “get rich quick” or spam the search engines. Some of the old keyword targeting efforts (buying links, cloaking content, keyword stuffing, etc.) that have been used by black-hat SEOs are antiquated practices that often do not result in any improvement of ranking and thus aren’t worth the effort or expense. If you’re expecting this post to give any insight on how to cheat the system, you will be disappointed so don’t bother reading further. However, if you’re interested in playing fair and look at SEO as a long-term commitment, then hopefully these steps will help you in your journey.
1. Priming Your Site First
As Miguel De Cervantes said: “To be prepared is half the victory.” As such, it is best to make sure that your site is in order before you start your keyword research. If you spend hours doing your keyword research only to find that you end up having to redo it all because there were issues with your site that went unnoticed, you will just end up having to do double work. When it comes to doing keyword research, it’s about working smarter, not harder. You wouldn’t keep pouring water into a bucket that has a big ol’ hole in the bottom of it, so why optimize your web pages if they’re not working? Before you start your keyword research, first check to make sure that the web page (and site as a whole) is running as smoothly as possible. Patch up any holes and make sure that your keyword research doesn’t become a futile effort by verifying that your site (or at least the pages you want to optimize) can be accessed and indexed by search engines.
Several years ago, the 3 big search engines (i.e., Google, Bing and Yahoo!) agreed on two protocols: robots.txt and XML sitemaps. These protocols help standardize how to provide access to your site’s pages for search engine spiders to crawl. Even if you have a robots.txt file on your root directory, it doesn’t necessarily mean that your work here is done. Do you have your XML sitemap listed in your robots.txt file? Are any of your important rules commented out? Are you actually disallowing your root directory? Simple things like these can put a damper on your keyword targeting efforts so resolve this before moving forward.
Also, it’s a good idea to check your XML sitemap to ensure that your targeted pages are listed in your sitemap. If your site is so big that you can’t fit all of your pages (and their corresponding attributes) into a single sitemap file that is less than 50MB in size, you might want to consider setting up a sitemap index.
Just like in the “bucket” example above, you’d be wasting your water-collecting efforts if you poured water directly onto the ground with nothing to receive it. If you’re building links to URLs that result in 404 errors you’re wasting your time and effort just the same. Although Google Webmaster Tools can also provide you with a list of 404 errors that they can pick up, you can also use other tools like Xenu Link Sleuth or Screaming Frog to detect these broken links. From there, you can implement 301 redirects for your broken links to point to the most relevant pages on your site. Just don’t go too hog-wild with 301 redirects because according to Google, you lose a small amount of link juice in the process.
Though you might assume that canonicalization issues only affect e-commerce sites, it can still affect non-e-commerce sites. If your server renders the non-www pages of your site and the www versions without redirecting one version to the other (e.g,. having all your non-www URLs 301 redirect to their www counterparts), Google may see these two URLs as two separate pages and will split the authority between the two. When it comes to canonicalization, you actually want to put all your proverbial eggs in one basket (so to speak). Here are some classic canonicalization issues that you should fix before taking any further steps at your keyword research:
- Index Page URL (/index.php, /index.html, /default.asp, /default.aspx, etc.)
- File Extensions (.php, .html, .htm, .asp, .aspx, .cfm, etc.)
- Non-www Version vs. www-Versions (http://www.website.com and http://website.com)
- Capitalized URLs (http://www.website.com/NewPage/ and http://www.website.com/newpage/)
- URL Parameters (http://www.website.com/category.html and http://www.website.com/category.html?p=1)
Malware, yuck! If you’re site has been hacked, chances are you might have malware on your server and it can hinder your Internet marketing efforts. Thankfully Google Webmaster Tools provides malware detection and can also provide some tips on how to fix a compromised site. When Google detects that your site has been compromised, they not only send out a notification via Webmaster Tools (if you have an account of course) but they also warn search engine users of the issue which can cause potential visitors to turn away before ever reaching your site. If your site has been infected, request a malware review with Google Webmaster Tools as soon as the threat has been eliminated. If you move forward with your keyword research but you end up being held back because of any of these issues, you’ll probably feel like one face-palm just isn’t enough.
Wait! You’re not done yet…there’s still just a little more priming that you should get out of the way before you kick off your keyword research. If you intend to do your keyword research for a portion of your site’s pages, it’s a good idea to take a look at those pages and gauge whether or not those pages (in their current state) are ideal for targeting. If your site looks like it was made in the late 90s, you might want to consider a website redesign. If people are coming to your site but are leaving as soon as they see it, it’s best to fix that problem before trying to bring in more visitors. This is particularly important when trying to bring in more mobile visitors. If your site isn’t very mobile-friendly, your potential visitors may leave immediately. Bringing visitors to your site is only part of the battle, keeping them on the site (and driving conversions) takes your Internet marketing to the next level.
One recurring issue I’ve found with doing keyword research is making sure that the pages that are being optimized are easily accessible. For organic SEO, this means ensuring that the pages you target can be accessed easily through intuitive, easy-to-find navigation. For many sites, this means having a main navigation (with drop-down menus) to guide your visitors through your goal funnel(s). That doesn’t mean you need to stuff your navigation with a bunch of drop-down menus until they disappear off the page. This can simply be achieved through having coherent, user-friendly information architecture. This seems pretty obvious, but not all websites take advantage of the potential up-selling opportunities that come from an improved navigation structure.
Why hold up your keyword research for something like this? Imagine you start making changes to your navigation (add breadcrumb navigation to your pages, move around your category pages, condense pages, etc.). All these changes not only affect which pages are available to do your keyword research for, but it also can help you identify how you can begin to “theme” your keyword research (more about that later). This can also help you to find more ways to better implement structured data markup to enhance your search engine result listings. Structured data markup (rich snippets) can be used for:
- Breadcrumb navigation
…and so much more!
There are many tools available to help you gather data and many of them are free. Though there are paid keyword research tools available, I recommend using the free resources first. Even though I’m a big fan of saving time, when it comes to digging around for data for keyword research, I often find that taking the time to fully immerse yourself in the data can result in “happy accidents.” Though you don’t want to get “stuck” in the data, you might find some useful tidbits of information in your data-gathering quest that you might not have found otherwise. Here are just a few free tools that I use (though there’s a bunch more):
- Google’s AdWords Keyword Tool – Provides data for what is gathered from AdWords (paid advertising data) but can give you a good comparison of search volume and keyword bidding competition, which can give you a good idea (even if it’s not 100% exact) of what keywords people are searching for in Google.
- Google Analytics – Gathers data from your site’s existing visitors (not their personal information, but some helpful insights like what keywords they queried in Google to get to your site, what search engine(s) they were referred from, whether they’re using a desktop computer or a mobile device, etc.).
- Google Autocomplete/Suggested Search – Provides a suggested keyword based on popular queries.
- Soovle – Ever wanted to see what comes up in the search suggestions for multiple search engines and sites? With Soovle, you can conduct one search query and receive real-time suggestions from Google, Bing, Yahoo, YouTube, Wikipedia, Amazon, eBay, and more.
- Google Search Results Pages – Why reinvent the wheel? By querying a keyword (or keyword phrase), you can see what sort of competition you’re up against. Also, you can see what other keywords your visitors may be targeting in their listing as well.
- Google Trends – Ever wanted to see what’s trending in Google searches? Google Trends not only shows you relative search interest of certain queries, it also allows you to compare multiple queries side by side.
- Ubersuggest – This tool can give you a lot of data all at once, and that can be a good thing. If you want hundreds of keyword ideas from real search engine user queries based off of one keyword/phrase, try this tool.
3. Data Gathering
With tools at your disposal and your website in order, it’s time to roll up those sleeves and get crack-a-lacking on your data gathering. One of the first things you’ll probably want to identify is what your ideal targets are. Although this may change on account of what you discover while doing keyword research, it’s a good starting point. If you’re building a new site from scratch, consider using a mock sitemap as your guide.
Pages Vs. Keywords
In some cases, doing keyword research may seem like an exercise in paradoxical thinking but that’s actually a good thing. Do you select your keyword targets first, or do you select your page targets first? What if you need to shift your keyword focus because of your pages’ limitations? What if you need to adjust your page targets to accommodate your keyword focus? As you gather more data while doing keyword research, you may find that you need to shift your focus, which requires you to gather more data. This is a good thing…in moderation. It’s really easy (especially if you’re like me) to get caught up in the data, over-analyze things, and want to make change after change after change. I’m telling you this now as a warning: you will never get it 100% perfect, so make sure you keep yourself on track to prevent spending an exorbitant amount of time spinning your wheels.
You won’t be able to target 100 keywords per page any more than you’ll be able to “own” a single keyword in the SERPs, so give yourself some reasonable limits to work around. When considering keyword targets, cut yourself off at 10 POTENTIAL keyword targets per page. Most likely, you won’t be able to target all 10 of these keywords on one page, so if you feel very attached to specific keywords, you might want to consider:
- Building out a separate page to target that keyword for; or
- Learning to let some keywords go. This is where your site architecture may need to be adjusted based on your research.
Let’s say you sell apples and oranges. You can create a single page focused on “fruits” and hope that your efforts of ranking for “apples” doesn’t come at the cost of having that page also rank for “oranges,” or you can create a fruit page that links to your “apples” page and your “oranges” page. If someone’s looking for fruit, you’d optimize the fruit page for the general “fruit” terms. If a search engine user is looking for very specific variety of apple, you could optimize your “apple” page for that. Notice how you are also creating linking opportunities for the general “fruits” page, the “apples” page and the “oranges” page. This not only helps your site’s users navigate through your site; it can also help with targeting specific keywords that are used as anchor text for these internal links. For example:
Your page title will likely be truncated past 70 characters in search engine results, and your meta descriptions will likely be truncated around 160. That doesn’t give you much room to target more than two or three keywords (depending on how many characters are in those keywords of course), so identify your primary (and perhaps secondary) keyword targets per page and let the other keywords act as supporting keywords. As search engines become better at identifying semantically similar words, it would be prudent to adapt your keyword research around this.
Thematic Keyword Research
To prevent keyword cannibalization and to improve your site’s information architecture, it’s a good idea to map out your keyword research in themes. Cyrus Shepard calls it a keyword theme, Bruce Clay calls it siloing, but whatever you call it, it is to better organize your keyword research (and mapping) so that your keywords are focused around a common concept (theme). As you start theming your keyword research, you will likely find areas where you can expand to longer-tail keywords which might warrant being targeted on a separate page.
Intrinsically-Focused vs. Extrinsically-Focused Keyword Research
One thing that I’ve had to come to terms with while doing keyword research for various sites has been reconciling keyword targets based on the keywords that a client is already well-optimized for, or stepping away from what the site already has by looking “outside the box.” I’ve come to label this as is “intrinsically-focused” versus “extrinsically-focused” keyword research. There are pros and cons of each:
|Already found in page content (less work)||Not found in page content, will need content to be revised (more work)|
|An “easy win” if already well-established||Might be more work to start from scratch|
|Limited growth opportunities||Potential growth opportunities|
|“Stick with what works”||“Thinking outside the box”|
With “intrinsically-focused keyword research”, you may focus on keywords that are already found in your pages’ content and/or anchor text for links pointing to a specific page. This represents some “easy wins” since you’ve probably already established authority on those keywords already. As the old saying goes, “if it ain’t broke, don’t fix it,” but what if you wanted to expand your keyword focus? By looking at what keywords your competitors are targeting that you are NOT targeting, you may find some insight that could help bring in more (converting) traffic to your site. Finding keywords that are not already included in a page’s content (but would still be relevant) is what I call “extrinsic keyword research.” Are “intrinsic” and “extrinsic” the right terms? Maybe not, but that’s how it makes sense to me, so I hope it makes sense to you too.
Which do you choose? It’s entirely up to you. Each of these philosophical approaches has its own merits. However, I tend to prefer hybridizing this approach similarly to how you might diversify your portfolio (only with keywords as the investment). That’s not to say you couldn’t “hedge” your keywords either, but that greatly depends on your goals and your market’s conditions.
4. Assessing Market Conditions
There are many variables to consider when it comes to selecting keyword targets and assigning them to specific pages. Doing keyword research without at least a little bit of knowledge of the “playing field” is a lot like shooting in the dark. Sure, you might get lucky and hit the target, but luck doesn’t always last forever and you shouldn’t run a business on it. There are a few things you should consider when selecting your keywords:
- Demand/search volume
- Seasonality or trends
You can use some of the aforementioned tools to compare your keywords against each other based on these values.
Remember the thought experiment “if a tree falls in a forest with and nobody hears it, does it still make a sound?” If you target a keyword but nobody’s looking for it, will it do you any good? Of course not. This is where the AdWords Keyword Tool can help. Although the search volume doesn’t equate to an exact number of monthly searches per keyword, it can give you some helpful data that you can use to compare with data from other keywords to help with your selection process.
Seasonality or Trends
Although the AdWords Keyword Tool can give you a good average monthly search volume, what if you want to view the search volume of a given keyword over time? A common example of seasonal keyword trends are Halloween-related keywords. Even though Halloween is consistently on October 31st of every year, there’s a huge spike in halloween-related search queries around this time of the year. Google Trends shows a good example of this in effect:
Other search trends may not be so simple, like with running shoes:
Notice that there’s a drop in interest in December but spikes in April and August. If you sell shoes through an e-commerce site, this sort of information can help you plan out the schedule of your Internet marketing efforts. Knowing when to target keywords is important in developing a good keyword targeting strategy (along with knowing “who” your demographic is, “what” keywords to focus on, “where” your page targets are, “why” you’re even investing the time to do it, and “how” to do keyword research).
The AdWords Keyword Tool provides competition data based on the amount of AdWords customers (advertisers) that are competing over a given keyword. This is provided in a percentage where the higher the number (percentage), the greater the competition (for keyword bidding for AdWords customers). Ideally, targeting a keyword with lower competition is better, and conversely, higher competition will often represent more difficulty which may turn you away from targeting it. On the other hand, if you go for the keywords with the lowest competition, you often risk that there’s a low amount of search volume associated with that keyword as well.
High Search Volume + Low Competition <> Guaranteed Conversions
Going after broad keywords that have low competition and high search volume does not guarantee conversions, and they don’t always make the best keyword targets. These types of queries may end up being for informational purposes only. Someone looking to learn about SEO may look for the keyword “SEO” but it doesn’t mean that they’re looking to hire someone to optimize their site. If a search engine user was in the market to hire someone to optimize their site, they may type in “SEO company” or “SEO agency,” which is more specific. We call these modifiers. They’re good to use since they are more specific, but keep in mind that they might have more competition (according to AdWords). The people who are willing to pay to bring in visitors for those keywords are probably getting some sort of ROI; otherwise they’d likely stop bidding for those keywords, so sometimes it’s worth “going with the flow.”
No Magical Keyword Research Formulas
If there were a magical formula for keyword research that you could apply to your keywords that would guarantee success, chances are someone would have already developed it (and they’d be making tons of money) by now. Don’t expect there to be a one-size-fits-all keyword research solution. Long-tail keywords are often considered “low-hanging fruit” because there’s often less competition associated with long-tail keywords and would potentially require less effort to start seeing results from optimizing your pages for them.
However, with long-tail keywords, you still run into the “tree falling in the forest” debate. If a page is well optimized for a keyword that nobody is searching for, it probably does you no good. I call these “vanity keywords.” Sure, you can rank for “Mea i ma’ama’ahia i na pa’ani ho’oikaika kino” and have a sense of pride knowing that you rank #1 in Google, Bing, and Yahoo for it, but if few people are looking for it what good does that do for your website (or business)?
Just remember, if your competitors are only thinking of numbers instead of people, you’re missing out on an opportunity to provide value to your customer base beyond what your competitors are providing. Your shoppers are humans, so it’s important to think like one when you’re doing your keyword research.
If you’ve themed your keyword research, the mapping should be a snap. Sometimes the issue may be that your pages aren’t well optimized for your intended keywords and you may find yourself having to make adjustments to your pages’ content. Depending upon your available resources, you may be averse to rewriting content. Do not be tempted to find shortcuts that linger in the black-hat SEO zone. If you’ve got a good keyword research strategy, you can use that as motivation to optimize your pages the right way.
Length < Depth
There’s no magical word count either. As the saying goes, “quality over quantity” and when it comes to your content, the word count alone shouldn’t dictate how successful your on-page optimization is. Although there have been several studies that show a positive correlation between word count and rankings, that doesn’t necessarily mean that you need to add a bunch of “fluff content.” Delving deeper into the “concept” that each pages’ keyword theme was built around not only gives you more ample opportunity at keyword targeting, it also shows your expertise in that subject which provides value to users. Users who like your page’s content might be more likely to link to it, which presents an opportunity for passive link building. “If you build it, they will come,” but “if you specialize in it, they will link to it.”
Now that you’ve identified your keyword targets for your pages, the question is: where do you put them? Obviously, you’d want to include the keywords within the body content. But there are other locations in which you can include the targeted keywords:
- Page Title (Title Tag) – used for display and ranking, is limited to 70 characters before being truncated in search engine results
- Meta Description Tag – not used for ranking in Google (though keywords that match queries will be bolded), is limited to 160 characters before being truncated in search engine results
- Heading (H1 and H2) Tags – some experiments have shown that there is a benefit to adding a targeted keyword to your page’s H1 tag
- Anchor Text (Links) – not as important for ranking as it once was but not entirely dead…yet
- Off-Site Web Properties (Social Account Profiles, Online Directory Listings, etc.) – as search engines become smarter at finding correlation between keywords through co-citation (co-occurence), it may be worth the effort of embedding your website’s targeted keywords in your social media profiles, descriptions, Yellow Pages listings, etc.
Do you use a different process or have some other tools you’d like to share? Feel free to mention it in our comments section below.