Over eight years running an agency, Paul ran more than a thousand split tests for clients. Most of the winners were not the ones anyone predicted.

He'll show the real tests and the real numbers. Whether a quiz in front of your opt-in beats a straight opt-in page. What changing one word in the first quiz question does to customer value. The headline test that moved lifetime value more than any redesign did. And the worst test he ever ran.

Some were page tests. The bigger ones were not. On one account the winning move was not on the page at all, and lifetime value doubled in thirty days as a result.

The businesses in that data set were not small. One went from £14 million a year to £23 million in twelve months. A US client doubled from $6 million to $12 million. Another went from $4 million to $20 million in three years and became the second fastest growing education company in America on the Inc. 5000 list.

All of it was done by hand. Paul now runs EverConvert, which automates the same process. It takes under four minutes to set up and is running at an 84% win rate and a 46% average lift across accounts.

In this talk, Paul will cover:

Which tests won, which lost, and the numbers behind both, so you can judge what applies to your funnels
How to spot the moments in a customer journey that predict value, and how to test whether they cause it
How to read the data and make creative solutions to convert more customers from the assets you already have.