Hey folks, this is Russ Jones, Adjunct Search Scientist here at Moz and Principal Search Scientist at System1. I want to talk today about a long-standing theory in search engine optimization, which generally goes like this: reviewers, aggregators, and non-manufacturing retailers will, over time, push makers and manufacturers out of the SERPs. The recent Google Product Reviews Update is just one further step down this long path leading away from makers and manufacturers. Let’s dive in.
Before we get started, we need at least a few definitions. What is the difference between a reviewer, aggregator, distributor, non-manufacturing retailer, and a “maker”?
Reviewer: A site like Tom’s Guide or PCMag uses its industry credibility and writers to produce comparison guides for products. They’re normally funded by advertising or affiliate agreements.
Aggregators: While I have no clear sitewide example, these are content providers that rely on the ratings of other sites to determine the content, whole cloth.
Non-manufacturing retailers: While there is some overlap here as many retailers have gotten into the manufacturing game, these are sites like Best Buy, Amazon, Walmart, and Overstock.
Makers: These are businesses that both make and sell their products. They can be big brands like Blue Buffalo and Apple, or smaller businesses like Hardcore Hammers or Eley Hose Reels.
Why should we care?
This is a fair question. Do we really care about the performance of maker/manufacturers on Google as some sort of moral or ethical measure? I think we should, so let me give you just a few brief reasons why before examining the evidence of the squeeze:
Bias filtering: Each class of site (reviewer, aggregator, retailer, and maker) have a different set of biases that can only be overcome by weighing each one against another. Reviewers and aggregators tend to be paid by ads or affiliate agreements, which can incentivize dishonesty. Retailers are paid by the sale of products on their shelves, thus they also have an incentive to be dishonest in rankings. And makers themselves have a self-bias. It’s the middle of the Venn Diagram of these data sources that makes good decisions possible.
Innovation: I wouldn’t be the first to point out the perverse system of startup funding, which has — at best — an unbalanced impact on who and what gets produced. If startups that produce a truly amazing product must pay their pound of flesh to the gatekeepers (reviewers, aggregators, and retailers) from the offset because direct, organic e-commerce is no longer an option, there’s an unnatural filter on what is produced and by who.
Finally, and this is the most important reason we should care: Google has a vested interest in pushing maker/manufacturers out of their organic results because they represent a massive advertising market. I want to be clear, here: I’m not accusing Google of intentionally doing this. Proving intent is one of the hardest things you can do without actual admission. However, if it is part of a growing pattern of Google pushing businesses out of organic, making them rely on ads, we should pay attention.
Now, I will be the first to admit that this isn’t the most exciting subject. Most websites aren’t maker/manufacturers, so you may be tempted to dismiss this research as simply irrelevant to your activities as an SEO or webmaster. However, I think that it’s important we understand directional trends that are driven economically and algorithmically in Google — it’s a way of thinking and planning, a strategy wrapped in a cautionary tale.
Is the trend real?
While access to data on Google and its algorithm has never been easy to obtain, there are good sources for historical SERPs. As a brief side note, I want to emphasize the importance of rank tracking, as it remains a primary source of information well beyond where your site currently ranks for a keyword. It allows us to investigate much larger trends, behaviors, and updates, all of which help us do our jobs better.
So, what does the data show? I took 50 singular, superlative product terms in the form of “best product type” and accessed the top 10 search results for that term in the month of January over the last nine years. I specifically chose singular terms like “best blender” so as not to intentionally bias the search results towards rankings where the intent was clearly for comparisons. (Asking “what is the best blender?” versus “what are the best blenders?” is an important distinction.)
Thus, we’re looking at 500 data points for each year from 2013 through 2021. Each website that ranked in these cohorts was labeled by hand to identify whether a site was a reviewer, aggregator, non-manufacturing retailer, or maker/manufacturer.
I compiled a number of statistics with relation to the categorization of a site and its likelihood to rank. The first, most straightforward question to ask is whether there is any “squeeze” at all, so to speak.
Are maker/manufacturers less likely to rank for singular product terms in 2021 than they were in 2013? The results are, frankly, stunning.
Between 2013 and 2016, roughly 50% of the SERPs collected in January included at least one maker/manufacturer. This is a very important data point, in my estimation, because it reveals that the search result was — at least in principle — capable of delivering the user to the answer of “what is the best product” rather than through an intermediary (non maker/manufacturers). Given the expectation of at least some diversity in results, a reasonable user should expect that on some occasions, Google could actually identify what is (based on some set of metrics) the best product of a certain type and include it in the top 10.
Something dramatic happened between January 2016 and 2017, but there are no clear updates that would target just this type of site and type of query, at least from a cursory review of update histories. Nevertheless, we see a stunning drop to just 15% (from 50%) in a single year. The trend continued such that by 2020 and 2021, out of the 50 terms and 500 rankings, only one was a maker/manufacturer. One.
In order to further confirm the trend, I followed the average (mean) ranking position for maker/manufacturers that were in the top 10 over that same time period. What’s interesting in this equation is that we see a much smoother line between 2017 and 2020 in rankings drop for maker/manufacturers. While many just dropped out of the top 10 in the first year (2017), the losses were steady over the next few years.
This additional information indicates that the trend is continuing, and that other maker/manufacturers who are holding on to rankings for these types of generic terms may not have much time left.
But there is another insult to injury in this trend, which was well articulated by Dr. Pete in 2015’s “The Incredible Shrinking SERP”. You see, once the mean position for maker/manufacturers passed the number 8 spot, they were at great risk of being removed from the first page altogether.
By 2019, for the same set of terms, the average SERP had nine or fewer results. This meant that the remaining maker/manufacturer pages were no longer on page 1. Effectively, 0% of maker/manufacturers benefited directly from organic traffic for these singular, superlative terms.
Maybe users prefer reviews, comparisons, & aggregators
When considering the many reasons why this might occur aside from the simplest explanation that Google decided it didn’t trust the maker/manufacturer websites, the thought came to mind that perhaps users just prefer comparison pages. This would be a charitable explanation and, as an information query, users would want to find a variety of sources that help them make an informed decision. However, I think there are several problems with this assessment.
Let’s take the example of “best mattress topper”, one of the 50 queries tested as part of this project. The first clue that there is more to the story than simply trying to include good review and comparison content is that Google chooses to include non-advertisement products in the search results! Instead of pages, they use a carousel.
There are several important points to make about this inclusion of a product carousel:
They prove that Google is aware, at least to some degree, that users would like to know exactly what the best topper is and be able to click immediately to that product.
Two of the first four in the carousel are not mentioned at all, anywhere, in the top 10 comparison reviews, while later items are.
They do not appear to be ordered with any relation to popularity, rating, or relevance to the query.
They indicate that Google has significant entity information on the products in question.
It’s a strange occurrence that Google knows users want the product (the answer to the question, not links to pages that answer the question), and that they have the product information but choose not to use it to either link directly to the maker/manufacturer, OR to rank the product carousel based on data extracted from the top reviews and comparisons that fill the organic rankings. But I think it gets worse.
Google’s product review update
Google announced an update that would target reviews and comparisons to ensure quality results. Among the many expectations listed in the update were knowledge about the product, what sets it apart from competitors, and providing quantitative measures. There is something incredibly important about this type of request of webmasters:
Either Google is in the position where it knows this information and will be able to validate it in an effort to determine which reviews and comparison sites are trustworthy,
Or Google is in the position where it does not know this information, and will only be able to compare this information from site to site in order to identify trustworthiness.
If Google is in scenario #1, then they have the capability to aggregate the results from the current review and comparison pages and determine the truthfulness of their statements (insofar as they are not merely opinion). If they’re in scenario #2, they have absolutely no business judging reviews and comparison sites until they’ve elevated their algorithmic capabilities, in order to use comparative data to determine truthfulness, thus warranting a move up to position #1!
In any scenario, Google should be capable of extracting the answer — or at least handful of answers — to these queries using their entity knowledge, product knowledge, link graph, and information extraction capabilities, which allow them to send traffic directly to the makers and manufacturers rather than intermediaries.
For example, we know that the two companies with the most listings in the top 10 reviews and comparisons are Tempur-Pedic (8 top 10 listings) and Viscosoft (7 top 10 listings). Tempur-Pedic does enjoy the second listing in the product carousel but that, of course, does not link to Tempur-Pedic’s product but rather to another Google listing of products filled with ads.
We have a word for this in our marketing lexicon: nothing more than a glorified interstitial.
Perhaps Viscosoft has a more egregious position. Despite nearly edging out the top position among the organic comparison sites, their products occur nowhere within the 24 products in the product carousel, despite having the highest ranking maker/manufacturer page for best mattress topper at #18!
Maybe they aren’t in this supposed “organic product carousel” because of this:
Why would Google ever choose to add a product to their carousel if they can’t ultimately make money off of it? The Viscosoft mattress topper search result page, as of this writing, has no ads.
The sad reality: Google hasn’t learned its lesson
Despite congressional inquiry and incredible research performed by Rand Fishkin and many others proving that Google is doing everything they can to keep you on Google, it appears that they are still intent on capturing potential customers into a giant click jail where the only way out is to click on an ad. But what is more egregious in this case is not that Google is merely keeping you on their site, but that they have a non-ad-labeled carousel called “Popular products” that clicks through to a special advertising experience search result (which we can trigger with specific search parameters, all of which is documented below.
Step 1: Popular products
Step 2: Specialized ad experience interstitial
Step 3: You can change the query, but retain the right bar ad experience.
I wish there was some good news for takeaways, but I just don’t really see much in the way of things getting better for maker/manufacturers. There are strategies, of course, but most of them will involve getting other sites to sell or market your product rather than your own.
The new review guidelines explicitly state that you should compare products to their competitors, which is a huge legal risk for most maker/manufacturers. This leaves them in a really difficult situation: either try to get your product reviewed by honest sites (which is an incredibly difficult task often requiring giving away free products that then must be acknowledged in the review), or spending money advertising or selling on major retailers and marketplaces like Amazon.
But if there is one thing we do know, it’s that there is no reason to believe that Google will actually list the best product or its site in the search results any time soon — there is too much money to be made by putting Google Ads between the user and the product.