👀 Weekend Update: Links Matter, Google Stays Lying About it

GM nerds,

Welcome back to the newsletter that’s here to congratulate AI Overviews for slightly improving. Amazing to see a Google AI product getting close to clearing

Here’s what’s good this week:

→ Common A/B Test Pitfalls to Avoid 📉

→ Ultra-Hyped Perplexity.AI Catching Heat 🤖

→ What Else Is New: Links Matter, Google Stays Lying About it 👀

→ more, obviously 💬

But first…

​This Week's Video 🗣️🎙️​

In this week's video, I shared some my reasons for not offering hourly coaching and consulting, including the low average rate, negative impact on personal brand, and the potential for addiction to these types of services. Additionally, I discussed the importance of staying in touch with the industry and the dangers of becoming out of touch while filling up your calendar with hourly consultations.

​YouTube​​

What’s Going Down @ Google 📊

→ Google has been saying for some time that links don’t matter. Pretty sure nobody with an even somewhat functioning brain ever believed that, and a recent study confirms this skepticism: “96% of sites in Google’s top 10 positions have 1,000+ links from unique domains,” and “only 0.3% of websites ranking in the top 10 had less than 100 backlinks and none had 50 links or less.”

Linked above is Danny Goodwin’s breakdown of the study, but you can also go straight to the source @ Internet Marketing Ninjas to read the “Backlinks Google Study” in more detail.

→ Google’s “Accelerate” looks like it will be a sort of centralized space with various resources designed to teach marketers to master AI tools and products. Resources include video tutorials and live streams, skill assessments, and tools for campaign automation, ad creation, and predictive insights.

AI Insight 🤖

→ Like every other majorly hyped AI product that’s hit the scene over the last few years, Perplexity is already catching heat. The Pages feature has some (self-proclaimed) “rough edges.” Basically, people have noticed the AI search engine has been republishing content from Forbes and others, using basically identical words, without proper attribution.

→ Many major AI organizations and tech firms are challenging a possible new AI safety law in California. They claim this legislation risks driving innovation out of state. The bill includes a “kill switch” component which is taking a lot of the heat.

→ This piece, which warns marketers of the “dangers of overreliance on AI-generated content,” isn't just your average AI-hater take. It's actually got some solid insight into specific things to keep in mind when using ChatGPT: tone of voice and brand consistency & repetition and information gain. Not taking these into account when using ChatGPT can negatively impact your content— click through to learn why.

Social Media Round-up 📲

→ Reddit just announced its first ads API partnership with Sprinklr, allowing users to create Reddit campaigns that integrate Reddit’s Data API.

→ Instagram is testing a new Reels feature that lets you share content from your Drafts with a small, controlled group before publicly posting.

→ X is testing a new analytics feature for Premium users, allowing users to organize audience data into specific date ranges and metrics, as well as analyze impressions for individual posts by audience gender, age, and country.

→ In the EU, advertisers on LinkedIn have lost the ability to target users based on data collected from LinkedIn Groups.

Hot Tip of the Week: A/B Testing Fundamentals

I simply won’t believe that any of you nerds don’t know the value of A/B testing. But even still, you might not be doing it perfectly, and this Semrush guide has some really clear insight that could remind you of some effective fundamental principles.

Click through for analysis re: how to run A/B tests for landing pages, how to decide what to test, the "Types of A/B and Split Tests for Landing Pages" (Variant A/B Testing, A/B/C Testing, Multivariate A/B Testing, and Split URL Testing), and more.

Finally, the piece features some common pitfalls of A/B Testing and how to avoid them. Here are three worth sharing:

- Choosing the Wrong Metric
- Stopping the Test Too Soon
- Testing Too Many Variables.

Click through for the rest.

From the Vault: The Secrets to Asana’s $5B Saas Success

This old Ross Simmonds thread dove deep into “Asana’s $5M SEO moat” and how they leverage elite top-of-the-funnel (TOFU) content to end up with "130K paying customers making them a $5B SaaS behemoth."

Looking at Asana’s SEO metrics, social media game, multi-dimensional and multi-lingual TOFU content creation and distribution strategies, and more.

Much to learn.

Jacky Chou​
Founder of Indexsy
hello@indexsy.com
Let's Connect on Twitter👇

​​

1238 Seymour Street, Vancouver, BC V6B 6J3
​Unsubscribe · Preferences​