Practical Ecommerce

SEO 101, Part 6: Going Deep on Keyword Research

This is the sixth installment of my “SEO 101″ series, following “Part 1: What Is SEO?,” “Part 2: Benefits of SEO,” “Part 3: Keyword Research Planning,” “Part 4: Keyword Research Tool Tips,” and “Part 5: Google Keyword Planner.”

Deep keyword research is critical for laying the foundation for a strong search engine optimization strategy. The data’s importance extends far beyond use in writing copy. Smart marketers use keyword research to build their site’s architecture, as a proxy for market research, or to mine for additional related products customers want to buy.

In “Part 5: Google Keyword Planner,” my primer on mining for SEO keyword data, I used as an example an ecommerce site that sells pens and ink refills. With just a handful of seed words, I finished with 619 relevant keywords, which begs the questions: “Do you really need to go this deep into keyword research? Where could you possibly use all 619 keywords on pens and ink refills?”

The answer is “yes.” You absolutely do need to dig deeply into keyword research, for several reasons. First, keyword research tools return only the data you ask for plus a small set of suggested keywords. The suggested keywords are the treasure, because those are the keywords you didn’t know existed. But to fully explore these unexpected suggested keywords, the resulting set of keyword data will be quite large. Failing to dig deep allows your competitors to capitalize on that treasure while your site focuses solely on the obvious surface keywords.

Analyze in Aggregate

In addition, keyword data is even more valuable in aggregate as it is by individual keyword. One keyword may seem relevant, but taken in aggregate you may see patterns that change your keyword strategy, or even your marketing strategy.

For example, in the keyword research on pens and ink shown below, 59.1 percent of the keywords contained the word “ink” and only 36.8 percent contained the word “pen.” Given that ink refills are disposable and need to be purchased again and again while high-end pens are purchased less frequently, this data makes sense.

Combined searches for all “ink” keywords and all “pen” keywords

Unfortunately, the data below also shows that customers are overwhelmingly looking for “cheap” products. That doesn’t bode as well for an ecommerce site that builds its reputation on selling fine writing implements and accessories. In addition to wanting cheap products, customers also want the “best” product. This is an interesting juxtaposition, possibly indicating that they want their writing implement to demonstrate their personal success, but they don’t want to spend a lot on that image.

Combined searches for “pen” keywords that contain these descriptive words

However, in the process of removing irrelevant keywords from the data set we noticed a lot of “printer” and “toner” keywords combined with “ink” keywords. So how many of the broader “ink” phrases are verifiably related to pens that you sell?

It’s true that as an ecommerce store that sells luxury pens and ink refills, you absolutely sell “ink” and “ink cartridges.” And it’s also true that keyword research shows 110,000 searches for the keyword “ink” and 40,500 searches for “ink cartridges” on average per month. But how many of those searches can your pens-and-ink ecommerce site win?

Google it. If Google’s search results return a different product type than you sell, it’s highly unlikely that your ecommerce store will be found relevant enough to rank on the first page for that keyword. Googling “ink” and “ink cartridges” returns only printer ink and a few tattoo sites. No pen ink refills are displayed anywhere in the paid search results or the organic search results. They just aren’t relevant according to Google’s algorithmic interpretation of what people want when they search for “ink” and “ink cartridges,” as shown below.

Google search for “ink cartridges.”

When all of the “ink” keywords have been removed that can’t be absolutely linked to pens and ink refills, the keyword research will contain only the absolutely relevant keywords. This cleansing process also removed 234,480 searches from the keyword data set, but those were searches the site couldn’t compete for anyway.

Now the picture looks very different, as shown below. A mere 18,100 searches (9.6 percent) are for “ink” keywords as opposed to the 249,960 searches we were looking at previously.

Combined searches for all “ink” keywords and all “pen” keywords, with “printer” related “ink” keywords removed.

But even more interesting, removing the printer-related keywords dramatically reduced the number of “cheap” keywords in the data set while leaving the “best” keywords untouched at 10,220. This massive swing in data, based on removing a large chunk of semi-relevant but likely unwinnable “ink” keywords, paints a much different picture.

Finding the Right Searchers

Searchers seek the pens online that our pen and ink ecommerce site sells. They want the “best” pen, and are less concerned with price. Searchers still look for “cheap” and “discount,” but they also want “expensive” and “luxury.” The much lower incidence of “ink” searches could indicate that customers don’t feel the need to search for pen refills, perhaps because they are considered an easy commodity to acquire at brick-and-mortar stores that customers already frequent.

Deep keyword research will also reveal opportunities for content creation. The data below shows several areas of interest around fountain pens that could be fulfilled with videos, illustrated articles, or other interesting content types.

Keywords that could spur content creation ideas

Last, but certainly not least, deep keyword research identifies a wealth of keywords from which to choose to optimize individual pages. Even a small ecommerce site that sells 10 products across three categories will have at least 14 pages to optimize: the home page, three category pages, and 10 product detail pages.

With so few products to research, the keyword data set will probably be limited to a hundred or so keywords — small but still far more than needed for 14 pages. Each page will need to be assigned a primary keyword to target, and one to four closely relevant secondary keywords to add variation to the content.

That’s potentially 70 keywords assigned to 14 pages. The remaining 30 keywords from a the data set of 100 keywords will probably either represent too few searches to bother with, will be too similar to other keywords to optimize for, or won’t really be as strictly relevant to your content as you had originally thought. In any case, it’s far better to have extra keywords than to miss opportunities to win new customers via organic search.

Amassing a deep set of keyword research gives you optimization choices, customer insight and data by which to make countless business decisions. It’s absolutely worth a couple more hours to dig deep into keyword research to discover this wealth of data that’s freely available.

For the next installment of this “SEO 101″ series, see “Part 7: Mapping Keywords to Pages.”

Jill Kocher

Jill Kocher

Bio   •   RSS Feed


email-news-env

Sign up for our email newsletter

Comments ( 3 )

  1. LOOM September 25, 2014 Reply

    Great series, thank you. Just a few questions:

    1) when I have my list of keywords for my product and it about 30, do I use all of them on my product page as long as they flow naturally with the product description?

    2) how do you find keywords for a specific product name and not a general name e.g. not for fountain pens, but for a
    Delta Serena Fountain Pen?

    • Jill Brown September 25, 2014 Reply

      Thanks, LOOM. Great questions.

      1) Each page needs to be assigned a single primary keyword — the keyword that is most relevant to that page and has the most searches. If the keywords are all different phrasings of the same primary keyword (fountain pens, pens fountain, fountain type pens, fountain pen, buy fountain pens, fountain pens online) then yes, absolutely work all of these very close variations into the same page of content. If the 30 keyword phrases are not extremely close variations of each other, if you have 1 product page with 30 relevant keywords all of high value, it’s possible that there’s too much information all on the same page. it will be very hard for 1 page to target and rank for more than one highly valuable primary keyword.

    • Jill Kocher September 25, 2014 Reply

      2) Keyword research for branded phrases is the same as general product phrases. If there’s no data for the branded phrase, it means that not enough people search for that phrase consistently enough for Google Keyword Planner to return data. For example, an average of 30 people a month search Google for “Delta Serena Fountain Pen,” but there is no data for “Delta Serena Collection.” So not enough people search for that phrase consistently enough for Google Keyword Planner to return data for “Delta Serena Collection.” When there’s no data for a phrase, try researching similar keyword phrases to find one that does have search data associated with it. Try the following:
      – shortening the phrase
      – using different synonyms
      – looking at your analytics to see which phrases already drive organic search to that page
      – examining the phrases competitors use on similar pages.