Showing posts with label keywords. Show all posts
Showing posts with label keywords. Show all posts

Thursday, April 3, 2014

SEO: Getting More from Google’s Keyword Planner

via Practical Ecommerce http://ift.tt/1lauvzF

Keyword research is the foundation of search engine optimization. Keyword data allows us to peek into consumers’ desires and analyze what they really want.
Google AdWords Keyword Planner opens that window for us. It is free, open to everyone with a Google account, and it’s tied to the largest direct source of search data — Google. There are certainly other keyword research tools: Wordtracker, WordStream, Searchmetrics, and many vendors that provide all-in-one SEO management and reporting solutions. I prefer Keyword Planner because Google drives the most organic search traffic to all of the sites I work with, hands down.

Why Keywords Matter to SEO

Before we can optimize, though, we have to know keywords and phrases what we’re optimizing for. What do we need to rank for to drive organic search traffic and sales? It may not be what you think it is.
If your industry uses a lot of jargon or your company is mired in internal marketing-speak, the carefully crafted messages applied to your site may not match the words real searchers use to find what they want at all. Keyword research can help you shake off the shackles of internal language and give you ammunition to show management why some of the language on your site needs to change.
When you see how few people search for fancy jargon and how many search for simple everyday language, you’ll have a much better idea how to optimize your site to drive more customers.
It’s all about relevance. Using the words real searchers use to find products like yours increases the relevance on the pages of your site, which in turn increases the likelihood that those pages will rank and drive customers.

Listing Keyword Stems

The quality of the information you input into Keyword Planner determines the quality and depth of the keyword research you can extract from it. If you just type in three words and hit go, your results will not be of any use to your SEO efforts.
The first step to effective keyword research is brainstorming. Stop and think about the products you sell. What brands do you offer? How do you arrange your products into categories and subcategories? What filters do you apply to help customers pick their way through your product catalog? Each of these is a keyword stem, a piece of a larger keyword phrase.
For example, if you sell shoes, your site might be organized like this.
  • Three categories: women, men, kids.
  • Three subcategories within the casual category: athletic, casual, dress.
  • Three filters you can apply: black, white, red.
Each of these nine words represented on your site is a keyword stem, or a prompt to suggest a set of keyword stems. They’re not keywords by themselves because, if you sell shoes, there’s no way your site will be relevant for the subcategory keyword “dress.” It is relevant, however, for “dress shoes.”
Your goal at this stage is to collect as many keyword stems as you can. Dig through your site. Dig through your competitors’ sites. Search Google and see which words the sites that rank are using. Think about the problems your products solve. Can you translate those into keywords as well?

Concatenating Keyword Phrases

Stems are half the battle. We need to turn those stems into keyword phrases that we can input into Keyword Planner to extract quality keyword research.
Paste all those keyword stems into a single column in an Excel spreadsheet. Use the Remove Duplicates tool in Excel (Data Menu > Remove Duplicates) to make sure you have each stem listed only once or you’ll waste time researching keywords more than once.
You’ll probably end up with a list of hundreds of keyword stems, unless your product offering is very small. Looking through them, most of the stems are likely modifiers to your main offering. For example, the category word “dress” is really just a modifier for “shoes.” You need to add “shoes” to make it a relevant keyword phrase. That’s where concatenation comes in to play.
Concatenation refers to joining two or more things together. In Excel, the concatenation formula allows you to string together as many cells or words as you like. Because this is a formula, it can be copied instantly down a column of hundreds of cells to create a list of keyword phrases from your stems.
For example, concatenating our nine stems with the product offering of “shoes” using this formula =CONCATENATE($A2,” “,B$2) creates the result below, in the “Concatenation Result” column.
Concatenation result in Excel.
Concatenation result in Excel.
Go crazy with concatenation. It’s as easy as creating a formula and pasting it down the column next to your stems. With our nine-stem example, we’d definitely want to concatenate additional phrases that included both subcategory and filter to create keyword phrases like “men’s black shoes” and “red women’s shoes.”
Plural versus singular versions of words can trigger different results in the keyword tool, as can putting the same words in a phrase in a different order. Add as many of these variations as you can to get the maximum number of keywords from the keyword tool at the end.
Some of the phrases you concatenate won’t make any sense, such as “athletic kids’ red shoes.” It’s faster to concatenate some useless phrases and paste them all into the Keyword Planner than it is to try to sift through the many phrases to weed out the ones that don’t make sense.

Inputting Keywords into the Tool

Now we have a spreadsheet full of keyword phrases, likely hundreds or even thousands of them. But don’t worry. You can paste up to 200 phrases into the Keyword Planner at once. As daunting as this seems at the beginning, when you get into a rhythm with the tool you can probably export around 10,000 keywords an hour.
Log in to Keyword Planner and click on “Search for new keywords and ad group ideas” (labeled as 1 below). In the “Your product or service” field (labeled as 2 below), paste in up to 200 of the keyword phrases you concatenated. Then click the blue “Get ideas” button (labeled as 3 below) to generate a list of keywords.
Google Keyword Planner
Google Keyword Planner
After clicking “Get ideas,” select the “Keyword Ideas” tab (labeled as 4 below). This tab displays the keyword phrases from actual users when they search for the words you entered into the tool. The “Search terms” column (labeled as 5 below) shows the search phrases you entered into the tool. The “Avg. Monthly Searches” column (labeled as 6 below) shows how many times on average per month that that keyword was searched for in Google, averaged over a period of 12 months. The “Competition,” “Suggested Bid,” and “Ad impression share” columns are relevant to paid search. It’s helpful to know how much the competition is willing to pay for ad placement, though, so you know how stiff the competition is likely to be for organic rankings as well.
"Get ideas" page in Google Keyword Planner.
“Get ideas” page in Google Keyword Planner.
The “Keyword” column (labeled as 7 above) lists the keywords that Keyword Planner determines are at least vaguely relevant to the list of keywords you entered into the tool. In this example, the tool provided 800 keyword suggestions in addition to the nine already entered at the beginning.
Keyword Planner will return a maximum of 800 keyword suggestions, which is why it’s so important to make sure the keyword phrases you concatenate are hitting all of the niches that are important to you. With each variation in phrasing, you’ll get another set of keywords that are unique. If you just input a couple of phrases, you’ll miss the unique variances in the way that your customers search, possibly missing out on an important trend you could capitalize on.
Glance through the keywords and determine for yourself if you feel that they’re relevant. You should be able to tell within the first page whether they’re on target for your site or not.
If they’re relevant enough, click “Download” (labeled as 8 above) to export the keywords in CSV format, which can be opened in Excel.
If the keywords suggested feel too broad — such as “nike outlet” and “payless shoes” — Keyword Planner offers a couple of different ways to filter the list and narrow the scope. In the “Keyword options” filter (labeled as 9 below), turn the “Only show ideas closely related to my search terms” option on. This option is similar to phrase match in paid search — all of the keywords listed will include the words you pasted into the tool at the beginning of this process.
The image below shows the difference in results with the “Only show ideas closely related to my search terms” option selected. See how the words may be in a different order or the phrases may include additional words? That’s the beauty of keyword research: discovering keywords that you hadn’t considered.
Keyword Planner results with the “Only show ideas closely related to my search terms” option selected.
Keyword Planner results with the “Only show ideas closely related to my search terms” option selected.
If only a few keywords were irrelevant to your product offering in the broad results, use the Include/Exclude feature (labeled as 10 above) to list words that must be included or excluded from the keyword list. For example, if you don’t sell Nike shoes, you might add the word Nike to the exclude list. If you want the keywords to at least contain the word “shoe,” you’d add “shoe” to the include list.
Don’t bother reading through every keyword the tool lists. As you continue the process and paste more and more keywords into Keyword Planner, you’ll start to see repeats where the tool returns some of the same keyword suggestions for many of the phrases you enter. If you try to screen out all the duds at this point, you’ll waste time repeating work that can be done more quickly at the end in Excel. Just download everything once you’ve got the relevance filtered to where you want it and repeat the process of inputting keyword into the tool.

Managing Keyword Data

Depending on the size of your concatenated keyword phrase list, you could have many CSV files sitting in your computer’s download folder. Merge these into a single file, either by manually copying and pasting or using this handy Windows command line method for merging CSV files.
Your CSV file undoubtedly contains many duplicate keyword phrases. Use the Remove Duplicates tool in Excel again, and begin the process of reviewing the keywords for relevance. After cleaning up the formatting and removing some unnecessary data columns, the keyword research looks like the image below, only much longer.
Sort keywords by relevance in Excel.
Sort keywords by relevance in Excel.
When you’re staring at a list of 50,000 keywords it seems impossible to manage. Try these tips for making the process easier.
  • Use the “Filter” tool in Excel (Data Menu > Filter) to turn the top row of your table of keyword data into a series of filters. As you filter or sort the data, the cells in each row will stay in sync so you don’t have to worry about the data in each column getting mismatched.
  • Sort the data by average monthly searches, from largest to smallest. The larger the number of searches, the broader the search phrase is likely to be. This is a good way to remove keywords that are too broad for you to target quickly.
  • Sort the data by keyword from A to Z. Keyword phrases that start with the same irrelevant word will be grouped together for fast removal.
  • Use a text filter in the “Keyword” column for cells containing a word that’s irrelevant. For example, filter for “flip flops” if your online shoe store doesn’t carry that kind of shoe. All of the phrases containing flip flops will appear and can be deleted at once.

Ongoing Process

Keyword research is an ongoing process. You’ll need to do it regularly, perhaps quarterly, to identify new trends and determine if some groups of keywords are increasing in demand and some are decreasing.
As you return to keyword research, consider how consumers think, to help you answer questions and to organize the content on your site. As you do, you’ll likely think of new pockets of keywords that you hadn’t explored before. Keep adding new data to your research over time and it will become a valuable tool not just for SEO, but also for areas like content planning, market research, and production development. I’ve addressed those additional uses previously, at “Using Keyword Research beyond SEO.”

Monday, January 13, 2014

Single sourcing and duplicate content (search engine optimization)

via I'd Rather Be Writing http://idratherbewriting.com/2013/12/16/seo-duplicate-content-and-single-sourcing/

Single sourcing and duplicate content (search engine optimization)

One of the challenges technical writers face in search engine optimizing their documentation is deliberating between single sourcing and duplicate content. Google tries to return a variety of unique results, rather the same versions. If you have multiple versions of the same content online, Google will likely just rank the version it perceives as best and bury the others.
I’m referring here to single sourcing the same content to multiple online versions, not necessarily a version for different mediums. Suppose you have 9 online guides for “ACME Software”: Version 1.0, Version 2.0, Version 3.0. And for each version, you have a Beginner Help, Administrator Help, and a Role-based Help (for managers). Further, you have printed versions (also online) of each of the 9 guides. This means you have 18 guides total.
In every version of the guide, you have a topic called “Setting Preferences.” What happens when a user searches for the keywords “setting preferences for ACME help” on Google? Will all versions of the “Setting Preferences” topic appear, given the high degree to which this topic is duplicated?
No, only 2 versions will probably appear — one web version and one print version. Here’s Google’s explanation:
Google tries hard to index and show pages with distinct information. This filtering means, for instance, that if your site has a “regular” and “printer” version of each article, and neither of these is blocked with a noindex meta tag, we’ll choose one of them to list. In the rare cases in which Google perceives that duplicate content may be shown with intent to manipulate our rankings and deceive our users, we’ll also make appropriate adjustments in the indexing and ranking of the sites involved. (See Duplicate content.)
In other words, Google will try its best to only show search results that are unique to the user. If two web pages have pretty much the same content, only one page gets shown. Google does recognize a difference between printed material and online material, so it’s likely that both the PDF version of the guide and the specific page will appear.
In sum, if a user wants version 2.0 of the guide for administrators, he or she has a low chance (maybe 12%) of actually getting that result in a Google search.

Why duplicate content gets dropped: Spammy sites

Even though the single sourcing scenario I described isn’t addressed in many discussions of duplicate content among SEO experts, the topic of duplicate content is widely discussed. This is because a great many websites scrape content from other sites. For example, I sometimes get pingbacks from spammy sites that have scraped (copied) the content from my RSS feed and posted it on their site in order to rank for specific keywords.
Google doesn’t want to penalize the victims of site scraping by demoting both versions of the content. Therefore Google demotes the spammy site while promoting the original site. Its algorithms are usually smart enough to tell the difference, but if it accidentally prioritizes the spammy site, you can report the scraper site through this form.

Canonical tags to the rescue

Duplicate content has also been a problem for mainstream web platforms such as WordPress. Not only is there a version of this post in the single page view, there’s also a version on the homepage, date-based archives, category archives, tag archives, and possibly series archives.
To let Google know the right page to prioritize in the search results, you can add the following link in the head of all pages that are the same:
1
<link rel="canonical" href="http://www.example.com/the-real-page"/>
where the-real-page is the canonical page.
The canonical link tag let’s Google know that this version of the page is in the canon — the other versions that duplicate this page are to be ignored.
If you’re publishing multiple versions of help material online, consider adding the canonical link tag to the version that you want surfaced in the search results. Adding a canonical link tag will ensure that one page gets priority over other pages. (See rel=”canonical” on Google Webmaster Tools Help Center for more information.)
Let’s apply the canonical tag to the previous scenario: Which “Setting Preferences” topic should we prioritize as the canonical topic?
One solution might be to add canonical links to the latest version only, and to choose the administrator’s guide. In the administrator’s guide, we could add a sidebar with links that allow users to access previous versions or other guides.
The problem with this method is that it limits Google to one type of search result only. If you’re a manager using version 2.0 instead of 3.0, how can you search for that version of content on Google? You can’t. You would be limited to the search results provided within your HTML help output.

Linking instead of duplicating

In an effort to avoid duplicate content, let’s explore a different strategy altogether. Instead of duplicating content, try putting one version of the content online and link to the various places where you need it.
For example, you might have different tables of contents that show different arrangements of the material (a Beginner TOC, Administrator TOC, and Manager TOC), but all TOCs that contain the “Setting Preferences” topic would point to the same topic.
With this strategy, you’ll quickly realize that you can’t have a Beginner TOC appear on the left of Setting Preferences, as well as have the Administrator TOC appear on the left of Setting Preferences, as well as a Manager TOC on the left of Setting Preferences.
And this is the problem with TOCs: they lock you into a fixed navigation based on a pre-defined idea of content order and containers. Further, they make it nearly impossible to deal with duplicate content.

Moving from print to web models

Mark Baker expands on the TOC problem in a comment in a previous post. He encourages us to move from print to web models:
Tom, I think the tech comm’s obsession with single sourcing and reuse are an indication that it is still stuck in paper-world thinking. Reuse makes perfect sense in the paper world because a person reading one book has to go to a lot of effort to refer to another book — in some cases it could be days or weeks of effort and considerable expense. Better than to repeat material in each book where it might be required than to reference another book that the reader might have a hard time locating.
On the Web, on the other hand, following a link is instantaneous and costs nothing. It makes far more sense to link to ancillary material than to reuse it inline in multiple places. (Which, of course, relates directly to the discussion in your last post. )
Tech comm should be worrying far less about reuse and far more about effective linking.
Of course, there are still many pubs groups that have to produce some form of off-line media, whether it be paper or captive help systems. As long as they have to keep producing those media, reuse does continue to make sense for them.
The problem with current practice, and with current tools, is that they are optimized for this kind of output and still create output for the Web by taking the individual separate documents generated on the paper/captive help model, with all of their duplicated (reused) content, and simply converting them to HTML. (Which is exactly why, as you observe, many of these tools create output that is not search friendly. In addition, such material is usually link-impoverished, which makes it not browse friendly either.)
What we need are a new generation of Web friendly, Web first, hypertext-oriented tools that can also do reuse for paper/captive help outputs, but which use links rather than repetition in the Web output. In other words, they need to be able to compose entire documentation sets entirely differently for the Web and for print/captive help.
In other words, the table of contents is a very paper-centric model for organizing content. Mark isn’t necessarily recommending that we delete the TOC altogether, just remove it as a fixed element to the side of each page.
You could still have a page that lists arrangements of links, such as what the WordPress Codex authors do here. But the table of contents doesn’t carry through to every single page, except at a high level.
wordpress table of contents and navigation model
Instead of TOCs, Wikipedia uses a model called portal pages, and you can see an example here for Ancient Rome. The portal page is like a trail map to the content.
Wikipedia portal page
The same topic could have numerous different portal pages that show you different entry points through the content. The portal pages might have different focuses based on different business scenarios.
Here are a couple of other minimal TOC approaches. Here’s an article from Google Webmaster Tools Help that shows a minimal TOC on the left.
Google TOC
There’s a lot more content in their Google Webmaster Tools Help Center. But they aren’t trying to force all content into one massive TOC. The same topic might be linked to in the Get Started, Learn More, and Fix a Problem sections.
Google's master list of topics
DataSift’s help gives another example of a TOC-less organization. The only TOC on the page is a list of pages highly related to the current page you’re viewing.
DataSift's TOC
The documentation homepage contains a larger variety of links. These links are not forced through as a TOC sidebar to every page.
DataSift help home

Assume readers begin at search, not the TOC

Wouldn’t many readers be lost without the fixed TOC on the left side of each page to guide and navigate them through the topics? Mark would say, assume that a user begins at search and searches for content matching specific keywords. Once the user lands on search results, if there are other articles related to the existing article, those articles should be linked within the article. Thus a reader navigates by two main methods:
  • Search results
  • Internal links within topics
Without question, this kind of organization puts more burden on the page to stand as aEvery Page Is Page Onecomplete whole. As a reader, you don’t want to jump back and forth to a portal page each time you want to read another paragraph or list of steps. In this model, pages are longer, more all-encompassing in helping users achieve their goals.
For more details, see Mark Baker’s book, Every Page Is Page One. (I highly recommend this book.)
The TOC-less model is most common with wiki architectures such as with Mediawiki sites. The main strength is to free content from one fixed structure. Removing the TOC allows you to create a trail guide (or portal page, or main contents list — whatever you want to call it) that provides a variety of paths through the same content. This way you can have 9 versions of guides while avoiding duplication of content.

Researching the right keywords (search engine optimization)

via I'd Rather Be Writing http://idratherbewriting.com/2013/12/23/researching-the-right-keywords-search-engine-optimization/


Researching the right keywords (search engine optimization)


Probably the most important decision in search engine optimizing your content is to target the right keywords. If a user searches for X, but your page is optimized for the keyword Y, your pages won’t usually appear in the results.
Most SEO experts recommend that you research the most likely keywords your users will search for, and then optimize for those keywords on your page. You can use Google Adwords’ keyword analysis tool to see popular keywords. Once you log into adwords.google.com, go toTools and Analysis > Keyword Planner.
For example, let’s say I wanted to research popular keywords related to “technical writing” for an upcoming blog post. According to Google Adwords, here are 20 terms related to technical writing that have the greatest number of average monthly searches.
KeywordAvg. monthly searchesCompetition
courses in technical writing49,5000.94
technical writing workshops40,5000.94
technical writing training certification22,2000.87
technical writing consultant18,1000.95
technical writing agency18,1000.87
technical writing manuals18,1000.99
technical writing school14,8000.96
technical writing rates14,8000.58
technical business writing14,8000.86
contract technical writing14,8000.95
training for technical writing14,8000.95
software technical writing14,8000.87
technical writing editing12,1000.96
technical writing society12,1000.12
technical writing projects12,1000.65
learning technical writing12,1000.94
technical writing seattle9,9000.52
technical writing contract9,9000.82
twin technical writing9,9000.09
uta technical writing9,9000.21
The majority of searches focus on learning technical writing through courses, workshops, certifications, or other online training. Another large number of searches focus on finding jobs in technical writing, including contracts, projects, and area-specific searches.
The competition column indicates how competitive the keyword is based on the number of sites ranking for the keyword. I’d have much better luck optimizing for the term “seattle technical writing” than for “contract technical writing.”
But unless I wanted to always write information about technical writer courses, jobs, degrees, or other beginner or career type material, I’m not going to find this keyword analysis tool very helpful.
What if instead I want to know which keywords to use in a post on “terminology selection with help documentation”? Well, the searches aren’t popular enough to form a dominating trend, so the keyword analysis tool fails to recommend much (in this example, “term emulator” is the only result).

Keywords from your own search analytics

When Google’s analytics fail to help you find the right keyword, the next place is to look is your own search analytics. What keywords are people using already in relation to your site? Here’s a list of the top 25 keywords that bring people to my site during the past year (from November 22, 2012 to December 22, 2013).
keywords264,131
% of Total: 71.83% (367,711)
264,131
% of Total: 71.83% (367,711)
(not provided)174,47066.05%
grasshopper7820.3%
i’d rather be writing7670.29%
technical writing jobs5760.22%
how to become a technical writer4850.18%
drawing thinking tool children4710.18%
technical writing skills4620.17%
idratherbewriting425 0.16%
life of a technical writer2950.11%
quick reference guide template2880.11%
id rather be writing2690.1%
technical writer interview questions2690.1%
technical writing tools2550.1%
technical writing certificate programs2280.09%
technical writing careers2140.08%
quick reference guide2040.08%
“jeff coatsworth”1980.07%
how to create video tutorials1950.07%
madcap flare1910.07%
moving to california1780.07%
becoming a technical writer1490.06%
technical writer skills1450.05%
create video help files1380.05%
technical writing interview questions1340.05%
sharepoint customization1330.05%
What trends are there? Not surprisingly, a lot of people who find my site via search are looking for technical writer jobs, skills, and certification programs. They’re looking for ways to become a technical writer and for a better understanding of a technical writer’s life.
Beyond these beginner tips and career questions, there are a few keywords for “quick reference guides”, “video tutorials,” and “flare.”
A few anomalies in the keywords exist. I’m obviously not hoping to draw searches for “grasshopper,” “jeff coatsworth,” or “moving to california,” but those keywords also draw people to my site. (The grasshopper reference relates to a brief post containing a picture of a grasshopper I once took.)
In looking at the keywords that users searched for to reach my content, you can see that many of the keywords are pretty general. The keywords people are searching for aren’t creative, catchy titles. They are “how to become a technical writer,” or “technical writing careers.”
When formulating titles, the plainest speech will probably connect with the most hits. Of course, linkbacks also factor into the findability algorithm, so a catchy title might indirectly gather a lot of hits if people link to it.
Overall, though, the keywords aren’t that helpful because 66% of them are “not provided.” A couple of years ago, to increase privacy for users, Google restricted the analytics available about the searches people make. For users logged into Google, their search results are not shown in analytics. (You can read more about the keywords “not provided” privacy policy here.)

Sources of traffic

One more fact to analyze is where the searchers are coming from. Although social has become a huge trend, it still accounts for a very little percentage of traffic. This chart shows where the 367,711 of my visitors came from during the past year.
Traffic sources367,711 total visits
google / organic251,964
(direct) / (none)58,392
bing / organic6,704
feedburner / feed5,545
t.co / referral3,536
yahoo / organic3,140
feedburner / email2,916
google.com / referral2394
linkedin.com / referral1,816
dlvr.it / twitter852
Twitter (“t.co / referral” and “divr.it / twitter”) is the most popular social network, but it doesn’t even contribute to 5,000 visits. This chart shows that Google still dominates as the #1 traffic source.
What do we do with all of this keyword information about technical writing if we’re not trying to convert online searches into signups for an online technical writing course, a technical writing degree program, or a technical writing contract? What if we’re writing about something much more specialized than keywords that would form a general Google trend?
It’s hard to say. Certainly, if I were to specialize in writing about how to become a technical writer, including training and other preparation, and also post about technical writing jobs, it would make traffic to my blog soar.
But helping people break into technical writing isn’t my passion, so it always remains on the periphery. I wrote a series about how to get a job in technical writing, and that’s pretty much all I have to say about the topic.
Fortunately, SEO is about more than targeting a few specific keywords related to your industry. There’s also a trend known as the long tail, where you try to maximize visibility by targeting diverse micro-keywords.

The Long Tail and SEO

The Long Tail refers to the overwhelming power of micro-keywords to outperform major keywords. For example, although lots of people arrived at my site through searches for “technical writing jobs,” that 0.22% was overwhelmed by the thousands of non-trend-dominant keywords that also brought people to my site.
At a previous Search Engine Strategies (SES) conference, SEO expert Avinash Kaushik explained:
One of the key strategies to win at search, both organic and paid, is not playing the head game really well but playing the tail game really well.
He said the diversity of small key terms creates a long-tail effect for search engine results. As an example, he highlighted the “marvelous job” Mormons have done in optimizing for the word “church.” If you google “church,” lds.org is one of the top results.
Avinash says LDS Church has so much content online about such a diverse number of topics, the little hits on all those topics, though small, add up to a search engine visibility that trumps the search engine visibility that comes from the major brand words.
Avinash says understanding the long tail isn’t possible when you just look at the top 10-20 rows of your keyword metrics. For example, if you log into Omniture or Google Analytics and examine your top 20 keywords, those keywords won’t give you the full picture of what’s bringing people to your site. You need to use other visualization techniques, such as keyword tag clouds (see tagcrowd.com), to see the long tail.
The biggest words in the tag cloud indicate the predominance of those terms on your site. Analyzing the search engine metrics, you might be inclined to evaluate a site’s effectiveness in ranking for those major terms. But Avinash argues that it’s more important to look at the microwords because collectively these small search words create more search engine visibility than the large keywords.
For my site, generating a list of keywords from tagcrowd.com yields the following tag cloud:
keywords from tagcrowd
As you can see, there’s not a whole lot here that’s useful. Overall, for the technical writer creating product documentation, there’s probably not a tremendous amount of useful information to glean about keywords from keyword planners or analytics tools.
For the most part, people arrive at a site through a variety of keywords, and it would be a mistake to optimize only for a narrow keyword target. Perhaps the more diverse number of keywords on your site, the better.