One of the questions you should deal with in CRO is the impact a particular page (the homepage for example) or class of pages (product or category pages) on the overall website conversion rate. The answer to that question determines the impact of optimizing that page on your bottom line.

Most of the conversion optimization work that I see nowadays is an over-architected usability fix deployed as AB tests.

Coming from a development background, I made sure that we hire some of the best developers to work on FigPii – the platform has been the internal tool we use to conduct our CRO projects.

But I had an issue with one of our senior developers. As he starts working on a new feature, he will review all the previous code and come up with better ways to write it. He will refactor the code and clean it up. And he will do this again and again.

With every new feature he worked on, we ended up with a better-written …

I have been a fan of the work from Basecamp for a long time. Started using some of their products back in 2006, read all the books they published and tried to implement some of their suggestions at Invesp.

Ayat mentioned that Jason posted recently about a new position at Basecamp – a head of marketing. So, before reading the job details and what it entails, I thought to myself, if I were responsible for hiring a “head of marketing” at Basecamp, what would I ask him to do?

This would also make for an interesting case study away from many of the …

I still recall the early days of running conversion optimization projects. Clients would come in with horrible websites, outdated designs, and never had paid any attention to usability.

You make a few fixes and test them against the control. And BAM! You have an amazing increase in conversions.

For most websites, long gone are these days.

Sites are well designed. Most good designers understand fundamental usability principals. Most websites have the basics covered.

The good news is that we can now focus on more pressing conversion issues. The bad news is that it takes significantly longer to …

Any experiment that involves later statistical inference requires a sample size calculation done BEFORE such an experiment starts. A/B testing is no exception. Calculating the minimum number of visitors required for an AB test prior to starting prevents us from running the test for a smaller sample size, thus having an “underpowered” test.

I will take you back to 2007 to one of the very first CRO projects we conducted for a large online retailer. The website received a massive amount of traffic (100k per day). Since it was one of our first projects, and the team was a bit trigger happy with AB testing, our first test included 30 different designs to run against the control. The headache of implementing all of these designs was overwhelming. None the less, we launched the test after about six weeks of implementation.

Seven of our challengers beat the original design within five days. The team …

Image source: Twoface

I have meant to write this post for a long time. There has been a lot written about one-tailed vs. two-tailed tests. However, most of the articles approach the topic from a purely statistical perspective providing many formulas but do not show how to do the calculations. Others articles approach the issue from a high-level business perspective. So, here is my attempt to delve deep into the topic. Some sections in this article are heavy on stats and formulas. If you are not interested in these, you can skip them! Special thanks to our data sciences team …