It's Too Soon To Pull the Plug on SQRs
+ This tool gives you answers to data inquiries in seconds
Welcome to Part I of Edition No. 66 of my weekly newsletter, providing practical analysis in the world of digital content strategy.
Contents
Tip: Get Answers in Seconds with GA4
Analysis: Why We Need Search Quality Raters
Tip: Get Answers in Seconds with GA4
I don’t hear a lot about the search bar at the top of the GA4 Reports dashboard getting used. (I’m guilty of neglecting it myself.)
This search function, though, is a great way to get quick, one-off answers to your questions.
Here’s how to use it:
1. Type your question in normal, conversational wording.
2. If it doesn’t show up directly in the search results preview, click on the relevant result.
3. So long as you didn’t ask something too complicated, GA4 should be able to provide the answer in seconds.
When would you use this? Instead of pulling up dashboards or opening complicated spreadsheets, you can answer inquiries for a manager, colleague or even yourself in a matter of seconds.
Did you find this tip useful? Share it to help spread the word.
Analysis: Why We Still Need Humans Rating Search Results
In Friday’s newsletter, I shared a link about how Google is ending its contract with thousands of search quality raters.
I’m going to tell you what a search quality rater does, and why getting rid of them could be bad for high-quality content creators (and good for sewer-dwellers).
What is a Search Quality Rater?
First, the short answer directly from Google:
In short, Google hires internet-savvy people – through third-party companies – who are fluent in their geographic region’s language.
In other words, these people not only need to know how the internet works but also how to apply it to their culture. They’re like a litmus test for people who live where they live and talk like they talk.
Their job, then, is to evaluate web pages in relation to search terms. If someone from Charlotte searches “panthers”, they’re likely looking for information on the football team. But what about someone in Pittsburgh? They may be looking for information on the local university’s sports teams.
And someone in another part of the world searching “panthers” may simply want to know more about the actual animal.
I’m making up that example, but it shows the kind of decisions SQRs have to make when evaluating online content vs. the intent of a search term.
While this is something bots or automatic software programs could certainly do more quickly – and cheaply – than humans, there are still cases where they may not be as accurate. And therein lies the problem of taking this job from thousands of people and giving it to a machine.
If you want the review 168-page SQR manual – last updated Nov. 2023 – you can do so here.
To save you some time, I have answered a few key questions:
How Does Google Use Search Quality Raters?
They assess search results based on relevance and quality to refine Google's algorithms. They don't directly impact rankings but provide feedback on how well search results align with user queries.
This is called “search intent.” As in the panthers “example above,” search intent is the type of information you’re looking for in relation to a keyword. It’s the difference between searching for “men’s shoes” because you want to buy a pair as opposed to learn more about how they are made.
By using humans to evaluate search intent, Google can take their feedback and tweak the algorithm so that search results more closely align with what a user wants.
Here are just some of the specific tasks they’re expected to perform:
Identify the main content of a page
Determine the reputation of content creators
Identify (un)trustworthy content
Identify low- and high-quality main content
How are SQRs selected?
Raters are selected for their comprehensive understanding of local cultures and internet use. They must be adept at research and interpretation of guidelines, ensuring a broad and representative user experience is simulated in their evaluations. This diversity in raters is key to capturing varied perspectives.
A Google search (ironically?) unveils some of the companies used to contract these roles. There appears to be a three-part test: 1) technical 2) page quality 3) page quality + needs met. Those who are hired by the respective companies seem satisfied, though you might find complaints about how Google treats them.
Specific Tasks Performed
SQRs perform detailed evaluations of web pages and their relevance to specific search queries. They assess factors like the authority, trustworthiness and quality of content. (More commonly referred to as E-E-A-T.) They also gauge how well a page meets the user's search intent, categorizing the usefulness of the information provided.
Why Do We Need Search Quality Raters?
Importance to SERPs Quality
The raters' assessments are integral to maintaining high-quality search results. They provide human insights that algorithms alone cannot, ensuring that the search results are not just algorithmically relevant but also practically useful and trustworthy from a user's perspective.
But with billions of pages to evaluate, it takes a lot of people to make this practical.
Impact on Gaming SERP Rankings
These individuals help prevent manipulative SEO practices from skewing search results. Their evaluations counteract tactics aimed at exploiting algorithmic vulnerabilities, ensuring that high-quality, relevant content is prioritized in search results over content that is optimized solely for ranking purposes.
As I mentioned above, robots can do this task – rating search quality – faster and cheaper than humans. That’s because they can process larger swaths of information much more efficiently. They develop patterns. Patterns that can be gamed.
Gaming SERP results is nothing new. It’s something people always have done and always will do in the name of misinformation, money or other twisted motives.
The problem with “gaming” people – i.e. SQRs – is that eventually, they realize the wool’s been pulled over their eyes. Bots, however, need someone to tell them they’re being fooled – whenever it’s a new black-hat tactic on which they have yet to be trained to identify – so they can course-correct.
Despite the strides being made with artificial intelligence, it’s far from ready to be trusted with something as important as matching content to search intent.
I say “as important” because millions of people still rely on Google daily for answers to their questions.
Some of less consequence, like recipes for a pineapple upside-down cake or how to unclog a sink. But what about an undecided voter looking for more information on two candidates? Or someone looking for a home medical remedy (never a good idea, no matter who’s rating content) instead of going to a doctor?
There are bad actors the world over trying to figure out how to rank their content as high as possible so they can profit off ad impressions or manipulate the less-informed.
I’m sure there will come a time when AI programs can do this task better than humans with very little oversight. But if the current state of SERPs is an indication, we’re not even close.
In the meantime, there are some practical ways to “fight back.”
Continue producing the highest quality, most original content possible for your respective audiences
Don’t fall into the trap of generalizing and casting a wide net
Make your website experience as positive as possible
Spotlight related content
Make the UX positive (don’t overwhelm with ads and popups)
Send great newsletters
If someone has given you access to their inbox, make the most of it. Aside from spam/promotion folders, this is another area where you completely control what content a user sees and how it’s presented to them
Keep reading with a 7-day free trial
Subscribe to Gerick News(letter) to keep reading this post and get 7 days of free access to the full post archives.