Testing Big Ideas vs Incremental Changes

Why test big ideas with massively different variables, rather than the traditional, incremental approach?

Here are some reasons why testing big ideas is better:

  • It usually takes a new, single-minded design to repair the disconnect between benefits and needs. The goal of any conversion page is to connect a product’s benefit to a target audience’s need in the most efficient manner possible. Any disconnect that exists is often because these two important things were never identified, or connected, when the webpage was designed.

  • Major insights and opportunities are missed because one element required another element to work. Usually, conversion barriers are not due to a single button, headline, or image. People experience a webpage as a whole, not a collection of different assets. Like zooming in a microscope, it’s easy to miss the forest for the trees, and unrelated elements for patterns.

  • Focusing on tactical changes is like finding your strategy from the wrong end. Know your strategy; get to the destination quicker.

  • Testing individual elements takes a lot of time.

  • You’ll clearly know what to test with a proper plan. If you don’t have one, you’ll soon hit a plateau and ask yourself, “What do I test next?” And you won’t know.


What if it doesn’t work the first time? What do we do then?

Remember you’re trying to reconnect the product’s benefits with the customer’s needs. If the new variant page fails against the old one, you have one of two problems:


  1. You misinterpreted, or didn’t gather, the right data about a conversion barrier

  2. Your implementation didn’t break down the barrier


The best way to work through this scenario is conduct additional user research to find out which elements are, and aren’t, working. It can often take a couple of passes through this process to get to the destination. 

Beware of going big all the time - your job is keep a balanced portfolio of tests running.

In addition, a potential downfall of this approach is that you need to have access to accurate user research data - sometimes a problem in itself for startups.

Massive conversion rate increases are achievable by going big - if you do it right. Solve real problems that your users have and your numbers will follow their happiness.