This post was originally published on the STAT blog.
Whether you’re tracking thousands or millions of keywords, if you expect to extract deep insights and trends just by looking at your keywords from a high-level, you’re not getting the full story.
Smart segmentation is key to making sense of your data. And you’re probably already applying this outside of STAT. So now, we’re going to show you how to do it in STAT to uncover boatloads of insights that will help you make super data-driven decisions.
To show you what we mean, let’s take a look at a few ways we can set up a search intent project to uncover the kinds of insights we shared in our whitepaper, Using search intent to connect with consumers.
Before we jump in, there are a few things you should have down pat:
Search intent is the motivating force behind search and it can be:
We left navigational intent out of our study because it’s brand specific and didn’t want to bias our data.
Our keyword set was a big list of retail products — from kitty pooper-scoopers to pricey speakers. We needed a straightforward way to imply search intent, so we added keyword modifiers to characterize each type of intent.
As always, different strokes for different folks: The modifiers you choose and the intent categories you look at may differ, but it’s important to map that all out before you get started.
For our whitepaper research, we pretty much tracked every feature under the sun, but you certainly don’t have to.
You might already know which features you want to target, the ones you want to keep an eye on, or questions you want to answer. For example, are shopping boxes taking up enough space to warrant a PPC strategy?
In this blog post, we’re going to really focus-in on our most beloved SERP feature: featured snippets (called “answers” in STAT). And we’ll be using a sample project where we’re tracking 25,692 keywords against Amazon.com.
Setting up projects in STAT means making use of the segmentation tools. Here’s a quick rundown of what we used:
Learn more about tags and data views in the STAT Knowledge Base.
Now, on to the main event…
To kick things off, we’ll identify the SERP features that appear at each level of search intent by creating tags.
Our first step is to filter our keywords and create standard tags for our search intent keywords (read more abou tfiltering keywords). Second, we create dynamic tags to track the appearance of specific SERP features within each search intent group. And our final step, to keep everything organized, is to place our tags in tidy little data views, according to search intent.
Here’s a peek at what that looks like in STAT:
Our standard tags (the blue tags) show how many keywords are in each search intent bucket: 2,940 commercial keywords. And our dynamic tags (the sunny yellow stars) show how many of those keywords return a SERP feature: 547 commercial keywords with a snippet.
This means we can quickly spot how much opportunity exists for each SERP feature by simply glancing at the tags. Boom!
By quickly crunching some numbers, we can see that snippets appear on 5 percent of our informational SERPs (27 out of 521), 19 percent of our commercial SERPs (547 out of 2,940), and 12 percent of our transactional SERPs (253 out of 2,058).
From this, we might conclude that optimizing our commercial intent keywords for featured snippets is the way to go since they appear to present the biggest opportunity. To confirm, let’s click on the commercial intent featured snippet tag to view the tag dashboard…
Voilà! There are loads of opportunities to gain a featured snippet.
Though, we should note that most of our keywords rank below where Google typically pulls the answer from. So, what we can see right away is that we need to make some serious ranking gains in order to stand a chance at grabbing those snippets.
Now, let’s take a look at which SERP features appear most often for our different keyword modifiers.
To do this, we group our keywords by modifier and create a standard tag for each group. Then, we set up dynamic tags for our desired SERP features. Again, to keep track of all the things, we contained the tags in handy data views, grouped by search intent.
Because we saw that featured snippets appear most often for our commercial intent keywords, it’s time to drill on down and figure out precisely which modifiers within our commercial bucket are driving this trend.
Glancing quickly at the numbers in the tag titles in the image above, we can see that “best,” “reviews,” and “top” are responsible for the majority of the keywords that return a featured snippet:
This shows us where our efforts are best spent optimizing.
By clicking on the “best — featured snippets” tag, we’re magically transported into the dashboard. Here, we see that our average ranking could use some TLC.
There is a lot of opportunity to snag a snippet here, but we (actually, Amazon, who we’re tracking these keywords against) don’t seem to be capitalizing on that potential as much as we could. Let’s drill down further to see which snippets we already own.
We know we’ve got content that has won snippets, so we can use that as a guideline for the other keywords that we want to target.
In our blog post How Google dishes out content by search intent, we looked at what type of pages — category pages, product pages, reviews — appear most frequently at each stage of a searcher’s intent.
What we found was that Google loves category pages, which are the engine’s top choice for retail keywords across all levels of search intent. Product pages weren’t far behind.
By creating dynamic tags for URL markers, or portions of your URL that identify product pages versus category pages, and segmenting those by intent, you too can get all this glorious data. That’s exactly what we did for our retail keywords
Looking at the tags in the transactional page types data view, we can see that product pages are appearing far more frequently (526) than category pages (151).
When we glanced at the dashboard, we found that slightly more than half of the product pages were ranking on the first page (sah-weet!). That said, more than thirty percent appeared on page three and beyond. So despite the initial visual of “doing well”, there’s a lot of opportunity that Amazon could be capitalizing on.
We can also see this in the Daily Snapshot. In the image above, we compare category pages (left) to product pages (right), and we see that while there are less category pages ranking, the rank is significantly better. Amazon could take some of the lessons they’ve applied to their category pages to help their product pages out.
So what did we learn today?
Want to see it all in action? Get a tailored walkthrough of STAT, here.
Or get your mitts on even more intent-based insights in our full whitepaper: Using search intent to connect with consumers.
More in our search intent series:
Powered by WPeMatico