How Google Dishes Out Content by Search Intent

What type of content do super successful sites create for each level of search intent? It’s not what you’d expect.

In the STAT whitepaper, Using search intent to connect with consumers, we looked at how SERP features change with a searcher’s intent — informational, commercial, transactional, or local. It was so chock-full of research that it sparked oodles of other content inspiration — from the basics of building an intent-based keyword list to setting up your own search intent project, to Scott Taft’s guide to building your own search intent dashboard.

But while doing the research for the whitepaper, we found ourselves pondering another question: is there a similar relationship between search intent and the kind of page content that Google sources results from?

We know from our study that as searchers head down the intent funnel, the SERP feature landscape shifts accordingly. For example, Google serves up progressively more shopping boxes, which help close the deal, as a searcher moves from awareness to purchase.

So, as consumers hunt for that perfect product, does the content that Google serves up shift from, say, category pages to product pages? To get to the bottom of this mystery, we mounted a three-pronged attack.

Prong 1: Uncover the top SERP players

Since Google delivers the content they deem most helpful, figuring out who their SERP favs are ensured that we were analyzing the best performing content.

To do this, we used the same 6,422 retail keywords from our original research, segmented them by search intent, and then gathered the top 12 results (give or take a few) that appeared on each SERP.

This gave us:

  • 6,338 informational intent results,
  • 35,210 commercial intent results,
  • 24,633 transactional intent results,
  • and 10,573 local intent results

…to analyze the stink out of. (That’s 76,754 results all told.)

From there, we dug into root domains (e.g. eBay.com and Amazon.com) to uncover the four most frequently occurring businesses for each search intent category.

We made an executive decision to exclude Google, who claimed top billing across the board, from our analysis for two reasons. One, because we attribute shopping boxes and images to them, which show up a lot for retail keywords, and two, because they aren’t exactly a competitor you can learn from.

Prong 2: Identify content page managers

After compiling the winningest sites to snoop on, it was time to see what kind of content they were offering up to the Google gods — which should’ve been easy, right? Wrong. Unfortunately, examining URL structures for frequently occurring page markers is a somewhat painful process.

Some sites, like Homedepot.com (who we wish had made the list for this very reason), have clean, easy to decipher URL structures: all product and category pages are identified with a “/p/” and “/b/” that always show up in the same spot in the URL.

And then you have the Amazon.coms of the world that use a mix of seemingly random markers, like “/s?rh=” and “/dp” that appear all over the place.

In the end — thanks to Stack Overflow, SequelPro, and a lot of patience — we were able to classify our URLs, bringing us to our third and final prong.

Prong 3: Mash everything together and analyze

Once we got all of our ducks in a row, it was time to get our super-sleuth on.

Informational intent (6,338 results)

This is the very top of the intent funnel. The searcher has identified a need and is looking for information on the best solution — is a [laptop] or [desktop computer] the right choice for their home office; what’s the difference between a [blender] and a [food processor] when making smoothies?

Thanks to the retail nature of our keywords, three product powerhouses — Amazon, Walmart, and Best Buy — rose to the top, along with Wikipedia, whose sole purpose in life is to provide the kind of information that searchers usually want to see at this stage of intent.

Although Wikipedia doesn’t have page markers, we chose to categorize their search results as product pages. This is because each Wikipedia entry typically focuses on a single person, place, or thing. Also, because they weren’t important to our analysis: while Wikipedia is a search competitor, they’re not a product competitor. (We still love you though, Wikipedia!)

Diving into the type of content that Amazon, Walmart, and Best Buy served up (the stuff we were really after), category pages surfaced as the preferred choice.

Given the wide net that a searcher is casting with their informational query, it made sense to see more category pages at this stage — they help searchers narrow down their hunt by providing a wide range of options to choose from.

What did have us raising our eyebrows a little was the number of product pages that appeared. Product pages showcase one specific item and are typically optimized for conversion, so we expected to see these in large quantities further down the funnel — when a searcher has a better idea of what they want.