Once Upon A Time in Web2
What if Reddit and Medium had an affair that resulted in an indigo baby? A rebellious child who didn’t blindly follow any rules, who had a strong sense of justice, and wanted to change the world.
That child became known as Hive.
Raised in secret by their cool aunt, Substack.
The Family Secret Revealed
But every family has a big secret that they only reveal when the time is right.
For 6 years, Hive was so busy decentralizing everything that they forgot a major part of their DNA. A family superpower that was always there, but they never tapped into.
The whole family knew how to grab the headlines, from search engines to the new AI agents crawling the web.
So I decided to have a look. How can we learn from them?
I Feel The Need For Speed
Here are the baseline speed tests for our Web2 relatives:
Medium PageSpeed: Performance 69 SEO 92
Reddit PageSpeed: Performance 48 SEO 69
Substack PageSpeed: Performance 47 SEO 92
I also checked our own frontends earlier this week on the 24th of May.
Peakd PageSpeed 1: Performance 45 SEO 67
Ecency PageSpeed 1: Performance 62
Then I decided to check again yesterday as it took a bit longer to write this post. I'm so glad that I did because there was a massive improvement in just 5 days!
Someone was listening.
Peakd PageSpeed 2: Performance 63 SEO 83
Ecency PageSpeed 2: Performance 84 SEO 83
I also checked our video frontend 3speak for comparison:
3speak PageSpeed 2: Performance 52 SEO 92
It has the same SEO score as Medium.
It is absolutely amazing to see that Hive frontends are suddenly faster and performing better than their famous Web2 family.
Platform Performance Comparison
| Platform / Frontend | May 24 Performance | May 24 SEO | May 29 Performance | May 29 SEO |
|---|---|---|---|---|
| Medium | 69 | 92 | — | — |
| 48 | 69 | — | — | |
| Substack | 47 | 92 | — | — |
| PeakD (Hive) | 45 | 67 | 63 | 83 |
| Ecency (Hive) | 62 | — | 84 | 83 |
| 3Speak (Hive Video) | — | — | 52 | 92 |
We have successfully unlocked one part of the puzzle. The engine is officially running fast.
United we stand, divided we fall (canonicals)
Let's talk about canonicals. If speed is the engine, canonicals are the GPS.
I purposefully avoided this topic in my first SEO audit. It's bigger than a technical fix — it's a structural crisis that needs consensus from all frontends.
To illustrate, here are screenshots of how PeakD and Ecency handle canonicals, using my own SEO audit post as the guinea pig — which I originally published on PeakD.
Peakd has a self-canonical.
Ecency also has a self canonical.
Both are claiming to be the original.
PeakD is right. Ecency isn't.
@ecency recently acknowledged changing their canonical approach, arguing Google treats canonical tags as hints, not rules anyway.
They argue their server-side rendering means Google was choosing their version anyway, making the old convention redundant. From a single frontend perspective, that localized logic holds.
That's technically true. But it doesn't justify a unilateral decision that affects the whole ecosystem.
I then checked a post I originally published on Ecency (How a Bitcoin ATM Led me to Hive Open Days Alicante) to see what PeakD does with it.
PeakD correctly points back to Ecency as the original source.
So we have one frontend playing by the rules and one rewriting them.
When frontends fight over ownership, Google doesn't pick a winner — it splits the authority between them, or gives up entirely and shows neither.
Our engine is world class now. But the GPS is sending us in circles.
It’s Complicated (Schema)
Imagine slaving away writing a technical project update or a one-week travelogue of the best beaches in Thailand. But they're all filed under “Miscellaneous”.
That’s the schema problem. Schema.org was launched in 2011 to establish a unified standard for structured data on the web.
For years, Hive frontends gave Google either mixed signals or none at all. It’s like saying “It’s complicated” to the bots.
The tides are turning.
@asgarth posted that PeakD recently shipped JSON-LD structured data — a significant step toward giving search engines a proper blueprint of our content.
Here is the current schema of PeakD. Note that when I checked on the 25th, they only had the author schema and as of the 29th May, they added this new schema.
Peakd's new SocialMediaPosting schema:
There's still room to grow — SocialMediaPosting schema works, but long-form content eventually needs Article or BlogPosting to truly compete with Medium and Substack. That's a conversation for another day.
So, I also checked Ecency’s schema and found that they had quietly implemented the BlogPosting and Breadcrumbs schema.
There's still a conversation to be had about schema conventions across frontends — and what author identity means for creator authority in the age of AI search.
But that's a discussion worth having as a community.
A Call for Synergy
The discussion has started and we seem to be moving in the right direction. Questions are being asked and improvements are shipping fast.
If we want to grow faster, we will need to do it together.
Then Hive, the black sheep of the family, will finally grow up to be an evolved version of its Web2 family.
Missed the first SEO audit of hive.io? Start here. Infographic credit: Gemini Flash 3.5
