Gaining better insights through segmentation

Let’s talk about an important component of developing actionable insights from your search data; namely, segmenting your keyphrases and assets. Gaining better insights through segmentation might sound simple but almost nobody is doing it.

The typical approach:

    • In organic search, people report on the rankings and performance of individual terms and phrases in relation to content and assets
    • In paid search, we look at the average position, CPC and quality score by term, campaign or ad group.
    • In analytics, we talk about traffic and engagement by page or keyword.

But there is a better and much more brand-oriented way to look at your data and discover actionable recommendations!

Most search marketers report to clients on site traffic and the visibility in the search results (a.k.a. rankings) while also trying to explain in-depth details about their search results to them. However, from working with a wide variety of Fortune 500 brands, I can tell you that most brand managers are not very interested in the search details. Generally, search is only a small percentage of their overall media spend. That means if the search is 5 percent of the brand’s budget, the search marketer is only getting 5 percent of the client’s attention—so every word and recommendation counts.

Almost all brand managers I have worked with do not really care about individual keyphrases or the difference in rankings between plural and singular search terms or abbreviations versus long versions. They care about the bottom line, the increase in visibility, and their sales/competitive advantage at the product and line level.

Now of course, most search marketers love all this stuff—drilling down into the analytics, comparing the breakdowns of search terms and keyphrases, and deciphering which ones or which combinations deliver the most traffic or best results.

But the problem is that: a) most brand managers do not have the time and interest to dig into the keyword-level detail data and b) being too deep in the data limits us to look at results purely through a search lens instead of through the lens of consumers or brand managers. Segmenting keyphrases can help.

Segmenting keyphrases

Let’s look at a tangible example of how segmenting keyphrases produces better insights into a brand’s search marketing campaign. To illustrate this, I am using a keyphrase strategy for an imaginary pet care brand.

Let’s assume the keyphrase strategy contains about 200 terms such as:

  • Cat litter
  • Wet cat food
  • Puppy chew toys
  • Feline shampoo
  • Doggy treats
  • Kitty brush

Search marketers generally would record the rankings and search volume for these terms; I have even seen the work of some agencies that have started grouping results further by adding another column, “category.”

In my opinion, this is just the start. Why stop at these broad results for the keywords or keyphrases (rank, volume, category) when further segmentation of analytics can give marketers so much more valuable and actionable information—and enable them to look at the findings from a product perspective?

In general, I would add a few more columns to the above strategy. These columns will differ on a brand-by-brand basis, and should align with the business goals and product lines. For our fictional pet food business, I would add:

  • Animal Type (cats, dogs, birds, etc.)
  • Animal Age (kitten/puppy, senior, etc.)
  • Sub-Product Types (wet food, dry food)
  • Search Term Intent (transactional, navigational, informational, etc.)

With this data segmentation, you can glean insights and tell stories based on what it tells you. On the highest level, you could report on the opportunity by Product Type:


Okay, that’s pretty nice—but still quite general. Now let’s apply the value of Animal Type to Product Type.


Immediately it becomes clear that there is a huge opportunity for dog health-related search terms, while the consumer interest for cat health terms is rather low in comparison.

What else can you do with this type of segmentation? How about segmenting the data by consumer intent:


You can start seeing a clear indication of the intent of the different queries. All these insights provide valuable pieces of the brand’s search story that can inform an overall content strategy. By segmenting the data for the keyphrases, we learn so much more about the story behind them and how they affect traffic (and ultimately, website content).

Look at what we learn about a brand’s visibility compared to its competitors when we segment the visibility data for search terms beyond just volume. For example, the illustration below shows a combination of paid and organic visibility by Product Line (colour-coded by the different competitors/sites) when applied to rankings and share of shelf.


This gives you the ability to tell your brands’ stories around their competitive product/category visibility in search, which is much more interesting than rankings of specific terms. Telling a brand manager that his product has 10 percent visibility compared the competitors who have 20 percent visibility will undoubtedly cause some reaction.

Another way to use segmented data is against website traffic. For example, the next graph shows a brand’s 2013 organic landing page traffic. On first look, it appears that traffic is declining (which is correct).


However, when we segment the landing pages by product line (there are two) you can see the actual trend more clearly and accurately—one product line within the brand is declining and the other is increasing. Therefore, looking only at surface results of keyphrases does not tell the real story of organic search.


Having these deeper, segmented insights gives a fuller picture of a site’s content: what’s working, delivering, attracting, engaging, and selling.

Start finding insights through segmentation now

It is really hard to fully understand the potential of segmenting until you start applying it to your own data. Even as I write this I am thinking about measuring the social engagement (shares, likes, etc.) against animal types (do dog or cat owners share more? Who has more followers? And so on.)

If you as a search marketer want to communicate efficiently and strategically with your brand, stop talking about the top three terms and rankings on the plural versus singular. Start segmenting your terms and content into categories that matter to your brand.

A good first step is to take your brand’s top landing pages and segment them. If you then evaluate the pages in your categories against KPIs such as conversion rate, time on site, and bounce rate you will be amazed at the stories that will surface (as an example, bounce rate and pages per visit dramatically vary by query intent or perhaps even by the type of animal the consumer owns. Do dog owners read more than cat owners? Do puppy owners have a higher bounce rate than owners of a senior dog?)

Go and find out!

Related Posts

Building a New Website

Building a New Website

Building a new website A step-by-step plan Building a new website Creating a new website can be an exciting but daunting…
Read More