Alex Yarosh Get Free Snapshot

Source record

@build_in_public TikTok profile avatar

@build_in_public

2025-10-24

“What is the biggest jump in website performance you've ever seen from a single A/B test? “ “Lifetime achievement, 1280% increase. I dumbed down the language...

Source Text

i

What is the biggest jump in website performance you've ever seen from a single AB test? Lifetime achievement, 1,280% increase.

I dumbed down the language, did a ton of customer, non customer interviews, found that the ideal customer profile was wrong, needed to target someone earlier, less experienced.

The language on the page was completely mis targeted, shifted. All that was like a lightning strike. Magical to watch.

Here's a similar clip from my podcast make it Homer Simpson. Simple, crazy. How many times people have their own made up word for their accessories or their own made up word for things that nobody else would understand except for them?

We know that people don't click on things that they don't know where it's gonna take them. A lot of times they're hurting themselves by doing that.

The header, you want it to be crystal clear what they're getting, not this long whole bunch of extra words stuffed in there. com.

Source Intelligence

i

In a reported A/B-test case, the speaker attributes a large lift to customer and non-customer interviews, correcting the target audience, and simplifying misaligned page language; the 1,280% figure is an unverified anecdote.

answer-ready-service-page-checklist · asserts

  • Interview customers and non-customers, compare their language with the page header and target audience, then A/B-test a clearer version and treat the observed result as local rather than universal.

Questions this source answers

i

What is this source mainly about?

In a reported A/B-test case, the speaker attributes a large lift to customer and non-customer interviews, correcting the target audience, and simplifying misaligned page language; the 1,280% figure is an unverified anecdote.

What should an operator take from it?

Interview customers and non-customers, compare their language with the page header and target audience, then A/B-test a clearer version and treat the observed result as local rather than universal.

Which topics does it connect to?

This source is connected to answer-ready-service-page-checklist.

What public evidence supports the record?

I dumbed down the language, did a ton of customer, non customer interviews, found that the ideal customer profile was wrong, needed to target someone earlier, less experienced. The language on the page was completely mis targeted, shifted.