meta_pixel
Tapesearch Logo
Log in
The Tai Lopez Show

How To Improve Your Website Performance by 100% with Split Testing

The Tai Lopez Show

Tai Lopez

Business

4.86.9K Ratings

🗓️ 23 January 2018

⏱️ 7 minutes

🧾️ Download transcript

Summary

Think of split testing as dating. By comparing performance, you can make choices on which path you’d like to go down. This is the reality of online marketing, that we are able to put users through funnels of testing and then analyze the metrics to optimize site performance.Learn more about your ad choices. Visit megaphone.fm/adchoices

See omnystudio.com/listener for privacy information.

Transcript

Click on a timestamp to play from that location

0:00.0

What we're looking for is like the Malcolm Gladwell concept.

0:03.6

Outliers, baby.

0:04.8

I don't like to give a shit about outliers.

0:06.4

When you hire people, find outliers.

0:08.0

There's two trillion average people you can hire.

0:10.8

Wait for the outliers.

0:12.5

Here's how it works.

0:16.8

You need, don't ask me why this number.

0:19.3

One of my friends with a PhD in Maths said,

0:21.0

in general, if you're doing two tests, 50-50, right?

0:24.8

50% to this one.

0:26.8

You need 52 results minimum.

0:29.4

So sometimes I see people publishing split tests.

0:32.5

I got 10 cells here and seven cells here.

0:34.3

See, this one's better than the 10 cell.

0:35.8

It's not statistically significant.

0:38.1

There are statistical significance calculators,

0:40.2

but even they are wrong more often than right.

0:42.7

So you need two things.

0:44.1

An online statistical significance calculator.

0:47.0

You want it, in my opinion, to be over 98% sure.

0:50.0

And the second litmus test is that you have over 52 on each.

...

Please login to see the full transcript.

Disclaimer: The podcast and artwork embedded on this page are from Tai Lopez, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of Tai Lopez and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.

Copyright © Tapesearch 2025.