Is it just us, or was Q1 2023 kinda crazy?
Google and Microsoft started duking it out in the AI wars. The source code of Russia’s biggest search engine got leaked. And did we hear something about a marketing community being acquired?
Read on to catch up on the SEO news and insights that might have passed you by the first time round.
The most popular stories from our Rich Snippets newsletter
Get up to speed with the ten most popular links we shared in our weekly Rich Snippets newsletter during Q1 2023:
1. The Content Cyborg: How to Use AI Writing Tools in Content Marketing – Animalz
Let’s be honest:
The AI think pieces aren’t going anywhere soon…
Luckily, this one from the folks at Animalz is actually well worth a read.
Angela Robbins and Ryan Law have spent the last two years experimenting with generative writing tools.
Their verdict? “AI is a force multiplier for great writers.”
Check out the pair’s article to hear the case for becoming a “content cyborg”.
Best of all, you’ll get five actionable tips for using AI to create better content based on a few years of in-the-trenches experience, not just hot takes.
2. E-E-A-T and major updates to Google’s quality rater guidelines – Search Engine Land
Just in the nick of time for the holidays, Google delivered the hors d’oeuvres no one asked for – significant revisions to the search quality raters guidelines.
The revised document has some substantial changes to E-A-T, including nine pages worth of new content and an amuse-bouche of extra ‘E’.
E-E-A-T (or Double-E-A-T) expands the acronym to add Experience and reframe Trust as the result of Experience, Expertise, and Authoritativeness.
As always, E-A-T fairy Lily Ray is here to save you from cramming the 176-page document like it’s finals week. In this article, she breaks down the most significant changes to the Search Quality Guidelines, section by section.
3. The State of Mobile User Experience – Nielsen Norman Group
It’s been 15 years and half a dozen mobilegeddons since Apple launched the first iPhone. Today, about 58% of web traffic is from mobile devices.
Collectively, we’ve finally reached a period of maturity in mobile UX – despite the bloated 594% increase in page weight in the last ten years.
Look at us – showing up proper for that ten year reunion!
Nielsen Norman Group has been around much longer than that. The UX/UI consulting firm started its first State of Mobile report way back in 1998.
That’s the same year Google was founded.
What has NNG learned in the past 14 years?:
- Consistency between device types is key.
- Everyone hates interstitials.
- The simpler the login, the more likely we are to use it.
- No, no one scrolls for the sake of scrolling.
- If your app needs another app to function, then I need you to figure out in-app browsing or GTFO.
Yes, there’s more in the full report. And yes, your C-Suite will likely be swayed by the omniscient branding.
4. Google’s March 2023 Core Update: Winners, Losers & Analysis – Amsive Digital
When Google’s March core update finished rolling out the early reactions were all over the place.
It seems like this go-around was significantly more impactful than the September 2022 update.
But who won in volatile SERP-a-whirl? Which industries were most impacted?
Who could make sense of this chaos and cake?
Thank the stars for Lily Ray and the team at Amsive Digital.
This new piece uses Sistrix visibility changes across 7,000+ domains and includes aggregate analysis by Similarweb website category.
And just for the heck of it, it features charts of all 7,000+ sites, so you can search for yours.
Hot dang!
5. Where’s the (Search)Love?
SearchLove San Diego wrapped in March, representing the end of an SEO conference dynasty.
The final hurrah featured a dream line up, and many of the speakers have shared their decks. The must-sees are:
- Aleyda Solis – SEO Low Hanging Fruit (featuring a jaw-dropping flow diagram)
- Mike King announced a new topical relevance scoring tool
So long, SearchLove, and thanks for all the fish <3
6. Large language models broaden AI’s reach in industry and enterprises – VentureBeat
Alright, we’re assuming everyone’s been asked at least one ChatGPT question by a client or a stakeholder by now.
Hopefully none with budgetary implications (bless their hearts).
Whether your experience was horror or hilarity, it’s probably time to unmask the killer catchphrase.
ChatGPT is a large learning model (LLM). LLMs are learning algorithms that can recognize, summarize, translate, predict, and generate languages using very large text-based datasets.
Remember Furbies? These tiny little deals with the devil traded a toy designer’s soul for a set vocabulary and the ability to listen for keywords (and even learn a few new ones). They would respond with a mix of these keywords and “furbish“, a translatable pidgin language.
[Fun fact: Jamie’s heard so many jokes about it being possessed, it started saying murder. Truly the perfect childhood toy.]
ChatGPT is a much, much bigger version of 1998’s Black Friday trample trophy.
Furby knew 60 parameters (“love”,”friend”,”happy”, “kill”, etc.). ChatGPT knows 175 billion.
But if ChatGPT is posed to pull market share from Google, then that basically leaves SEOs optimizing… for very large Furbies.
“Generative AI” takes this new expanded vocabulary and replicates patterns. It’s not creative, but it’s good at moving the words around.
JinaAI does a great job succinctly describing these machinations.
When you enter a query:
- The LLM processes it with its encoder network, converting the input sequence into a high-dimensional representation.
- The decoder network then uses this representation – along with its pre-trained weights and attention mechanism – to identify the specific piece of factual information requested by the query and search the LLM’s internal representation of this knowledge (or its nearest neighbors).
- Once the relevant information has been retrieved, the decoder network uses its natural language generation capabilities to compose a response sequence stating this fact.
And that’s why we’re pointing you toward their guide to optimization for large language models.
LLMO is a new concept. This article coins the phrase and offers a rough sketch of its mechanics and shared concepts.
7. Yandex scrapes Google and other SEO learnings from the source code leak – Search Engine Land
January saw a Yandex source code repository roll onto torrent site BreachForums like a Russian tech potato.
Screenshots of the original post show it claimed to contain everything but the anti-spam bits.
Engineer Arseniy Shestakov spotted 14 Yandex products in his “friend’s” analysis:
- Yandex search engine and indexing bot
- Yandex Maps
- Alice (AI assistant)
- Yandex Taxi
- Yandex Direct (ads service)
- Yandex Mail
- Yandex Disk (cloud storage service)
- Yandex Market
- Yandex Travel (travel booking platform)
- Yandex360 (workspaces service)
- Yandex Cloud
- Yandex Pay (payment processing service)
- Yandex Metrika (internet analytics)
Are you salivating? The leaked source code for Russia’s largest search engine?
Are you also terrified because wtf is an aapi_file_list.txt?
Fair.
You know what helps with understanding crazy code? ChatGPT.
Mike King and Ben Wills fed Jamie’s LLM frenemey each statement and created a spreadsheet to translate the function.
Scoffing because Yandex only holds 0.85% of the world market share for search engines? Yeah, well it also allegedly scrapes Google.
Mike summarized his findings in this SEL piece. Notable nuggets include:
- The overlap between Google and Yandex
- Insight into Yandex’s architecture
- There are 17,854 ranking factors in the codebase
- There are upper limits to prevent over-optimization (you keyword stuffing monster!)
- Yandex uses very few machine learning models
- Clicks are 100% a factor
The article is dense. The full breakdown is a modern Ulysses with fewer fart jokes. The full spreadsheet is available in exchange for your joining the iPullRank mailing list.
8. SEO fails: Horror stories from some of the world’s smartest SEOs – Traffic Think Tank
We’ve all made mistakes that still haunt our nightmares years later.
And what better way to exorcize those demons than to see how some of the biggest names in SEO have messed up too?
So, to celebrate the 100th edition of Rich Snippets, we asked the smartest people we know to share their biggest SEO fails with us.
Hear from some SEO legends, keynote speakers, and seriously smart marketers who’ve…
- Deindexed entire websites
- Rank tracked local search queries in the South Pole
- Sent confidential documents to the wrong client by mistake
And much more!
Some of these horror stories will make you laugh. Some will make you want to die from second-hand embarrassment. But they’ll all make you feel a bit better about your own SEO fails.
9. Google Search’s guidance about AI-generated content – Google Search Central
Until recently, Google seemed fervently against displaying generative AI content in SERPs.
And then the shareholders demanded to know where Google’s AI chatbot was.
Enter stage left: Google Search’s guidance about AI-generated content.
We’re going to need both sides of the Google fanatics/haters to take a reflective pause here for a moment. No one in tech is immune from the devil’s bargain with dumb money.
The Search Relations team is being instructed to be part of a company-wide effort to push a product out the door by a CEO who doesn’t seem to understand that a “hackathon” surrounded by the still-warm chairs of your former colleagues is more accurately called a seance.
Here’s a tl;dr of Google’s guide to *correctly* ruining SERPs with AI:
- Automation != spam
- You still need E-E-A-T to rank
- Re-read that Helpful Content Update again one more time – and pay attention to between the lines
- SERPs have adapted to elevate original news reporting
- Please don’t give AI an author byline
- Please, please at least proofread it before you hit publish
10. Optimize Interaction to Next Paint – web.dev
In case you were feeling comfortable with your Core Web Vital scores, remember…
Interaction to Next Paint is lurking around the corner.
If you’re not familiar, INP is the first CWV contender poised to go beyond pageload and into user flow. It’s measured using the visual feedback that accompanies user input (think: tapping a thumbnail to see a product image or the color change of a pressed ‘Add to Cart’ button).
This game changer is looking to observe the latency of all click, tap, and keyboard interactions that occur throughout the lifespan of a user’s visit to a page.
What’s a good INP? 200 milliseconds or less in the 75th percentile of page loads.
INP has been part of the BigQuery Crux API dataset since April 2022. A fresh web.dev optimization article from Jeremy Wagner and Philip Walton is transparent about its diagnostic and troubleshooting shortcomings.
Key takeaways include:
- Find slow interactions in the field data using the web-vitals JavaScript library
- Workarounds for when you don’t have enough RUM Data
- How to identify input delay
- How to fix long input delays
- Optimizing event callbacks
Plus, it features a darn fine pep talk about how persistence is key.
The hottest topics in our Slack community
Get the inside scoop on the hottest topics from our exclusive Slack community during Q1 2023:
Dealing with AI fatigue
A TTT member kicked off an open and honest discussion within the community when they asked if anyone is keeping up with all the latest developments in AI at a time when there seems to be a new “next big thing” every other week.
Responses varied from “I don’t care at all” and “honestly, I am blocking out as much as I can” all the way to one agency owner who said a member of their content team has decided to change careers as they’re worried about AI taking over writing jobs.
But for the most part, TTTers are keeping their eye on the latest developments and dipping their toe into AI through small tests while trying not to fall victim to shiny object syndrome.
Lots of love for the TTT Toronto meetup
The community shared lots of love for the Toronto TTT meetup that Tory Gray and Jess Joyce ran back in February.
Bibbidi-Bobbidi-Booth
Folks loved how TTT OG James Norquay drew attention to the Prosperity Media booth at a recent conference: hiring a magician to tailor some “SEO tricks” to people walking past.
James says this “generated a whole bunch of buzz for the booth and turned out to be good fun”.
Check it out on YouTube so you can see for yourself why you might want to consider this for your next booth 👀
How to start a successful local meetup
After sharing the success he’s had by setting up an e-commerce meetup in his city, Jonathan Gorham gave an impromptu masterclass in how to promote and run a local meetup.
His top takeaway: when you’re hosting an event, people approach you, making it way easier to network (especially if you’re usually quite shy and standoffish at networking events).
Ink building
Back in March, Michael Brennan from Legacy Communications shared that its links for life for him – literally.
Michael has no ragrets… despite one of the biggest names in the industry pointing out his new ink looks as like a paperclip as it does a link 🫣
An email deliverability masterclass
When a TTT member asked the community for recommendations on how to improve cold email deliverability, Aleksandar Ljubinkovic stepped in with an absolute masterclass in making sure your outreach emails aren’t getting caught in folks’ spam folders.
BARD reputation
Our members weren’t exactly feeling Google’s BARD announcement when TTT MVP Jess Joyce shared it in our #stay-updated channel.
One member wondered if BARD stands for “Brutally Abrupt Revenue Decrease” for anyone formerly ranking for featured snippet queries, while another suggested Google must have laid off its branding team 🫣
So, it’s safe to say Google’s first forays into AI chat haven’t exactly set TTT alight…
Matt’s insights into navigating stock option packages
In March, Matt shared some thoughts for everyone that’s considering a move into a private company where stock options make up a portion of your compensation.
He took us on a deep dive into:
- The tax implications you should be aware of when it comes to stocks
- What happens when you exercise stock options
- How few opportunities you have to sell stock that’s not in a public company
- The biggest value-add clauses you should look to negotiate into your contract when it comes to stock options
How much do agency owners make compared to freelancers?
A freelancer kicked off a super interesting thread in our #agency channel when they posed this question to the freelancers who’ve turned agency owners:
Let’s say hypothetically you cleared $200,000 as a freelancer. How much revenue did you need to earn as an agency to pay your employees/overhead and maintain the same $200,000 pay?
This opened up a candid conversation where Joel Klettke of Case Study Buddy, Kane Jamison of Content Harmony, Greg Heilers of Jolly SEO, Blake Denman of RicketyRoo, and Tory Gray of The Gray Dot Company all shared their insights into the pros and cons of being freelance or an agency owner – in terms of both money and beyond.
A peek behind the curtain of a killer content strategy
Jenna Potter gave us a peek behind the curtain at how the team at Codeless took Monday.com to 300,000+ monthly blog visitors in 18 months.
She walked us through the process they’ve developed to produce nearly 1,000 pieces of content (and counting) for the SaaS unicorn. Then she answered members’ follow up questions on how the team working on Monday.com is set up, how they approach link building, and their content brief process.
That’s all, folks!
Want to stay clued up on all things SEO without having to wait until our next quarterly update? Sign up to our weekly Rich Snippets newsletter and join Traffic Think Tank to get real-time updates and join the discussion.
And if you want to catch up on what you missed earlier in the year, catch up on our quarterly highlights from Q4 2022 and Q3 2022.