Scroll-stop teardown
TerraCycleThe homepage
What it's trying to do
The homepage has to do two jobs that pull against each other. Convince a person their coffee pods and toothbrushes have somewhere to go, and convince a company that a whole waste stream can be handed over. The page picks the consumer job first, which is right, and then stacks the rest on top of it.
What works
The tab row sitting directly under the hero is the correct first move. Most popular, Recycle at home, Recycle at school, Recycle in-store, New programs. That sorts by where a person is standing rather than by what TerraCycle sells, and it puts a program in reach before any scrolling happens.
The search field asks "What would you like to recycle?" instead of saying Search. That's the question a real person arrives with, in their words.
Three fixes
-
the hero shows the trash, not the place
The H1 is "Recycle the unrecyclable with TerraCycle®." Beside it sits one photograph: a coffee pod, a razor, a face mask, a toothbrush, a candy wrapper, a mascara wand, floating on white.
That's the before. Every visitor already owns that picture. It's in their bathroom. The word carrying the headline is "unrecyclable," and the photo doesn't argue with it.
The place where it stops being unrecyclable is the one thing nobody else has, and it appears nowhere on the page. One still from inside a facility, in that slot, does work the studio cutout can't.
the hero proves the headline instead of repeating it.
-
the best sentence is in section five
"We operate in over 20 countries, engage over 200 million people, and recycle billions of pieces of trash through our various platforms annually."
That sentence lives inside the "Learn about us" card, under "Get to know TerraCycle," which is the fifth section down, behind a Learn more link. Almost nobody reaches it.
Move it under the H1. It's the only line on the page a competitor couldn't also write.
the fact that makes the claim believable gets read by everyone instead of by the few who scroll past the Foundation.
-
the machines answering for you can't read the page
The homepage has no canonical tag and no structured data. Zero schema blocks in the source. terracycle.com/sitemap.xml returns a 404, and so does /llms.txt. robots.txt is one line and points at no sitemap. The og:title and og:description tags are in the head and both are empty, so a link pasted into Slack renders with no title on it.
People ask chatbots what to do with a used razor. The answer gets assembled from whatever is machine-readable, and right now that's a meta description.
the company that actually owns the answer gets named in it.
What I'd make instead
"TerraCycle Made. Shop products made from the trash we've recycled together."
That card is the whole company in one sentence, and the page never shows it happening. The loop is claimed in a card and closed offscreen.
So film the loop. One object, start to finish. The coffee pod on the left of the hero photo, into a box, onto the line, out as the thing on the TerraCycle Made shelf. No program name, no partner brand, no voiceover explaining what the camera already showed.
Nobody else can shoot it, because nobody else runs the floor.
If one of the three is worth doing, I'll send the rest of what I wrote down. If not, ignore me.
Jason Portillo
Boring Stories · boringstories.io