How We Increased Our Website Conversion Rate by 46%

A while ago, I was asked at an interview to explain a situation where, as the product manager, I had to translate the output from an analytical tool into user problem statements whilst maintaining a balance between user experience and company goals.

I’ve had quite a number of such issues but I picked a particular story in 2019 when we experienced an unusual decline in conversion rate after redesigning the company’s website. The fact that there wasn’t a decline in website traffic (according to Google Analytics) contributed to the puzzle, so we needed a solution and everyone looked up to me.

The following are the steps I adopted.

1. Gather Data

In conjunction with the marketing team, we signed up for a free version of Hotjar.com. We really needed to understand what and where the issue was. Since google analytics already showed us where the majority of the drop-offs were happening, we monitored the page through Hotjar to watch the heatmap of users in real time. We recorded every user interaction for the next 1 week to be sure we gathered enough sample size.

2. Identify Problem(s)

After gathering enough events through hotjar, we discovered two main problems:

• The main CTA button was below the fold. This means that the button was hidden on a specific set of mobile devices. This was the major cause of the drop-off, especially for returning users who were used to the old design.
• In a bid to reduce the number of steps required to make a purchase, we eliminated a page by merging its content with another page. This caused another confusion as we watched the users navigate.

3. Prioritize user problems

Considering we have very few but important problems, we used the?Impact-effort matrix?to determine which of the problems to tackle first. We decided to sort out the CTA issue first because it required less engineering effort and will potentially bring the business more returns (in terms of conversion) compared to the collapsed page issue (which was sent to the backlog of issues to be treated when engineers have space to take them).

4. Create problem statements

We proceeded to create problem statements that describe each of the problems in detail. We also used data from Hotjar to support each problem statement and provided context for why it’s important to address.

5. Brainstorm solutions

We didn’t sweat much on this because the issue was apparent. All we spent time on was trying to make sure no device was left behind. In conjunction with the design and front-end team, we figured out a way to make the button visible on any mobile device.

6. Test and implement solutions

During the brainstorming section, we agreed on running an A/B testing for the CTA button colours to see which is more preferred. We ran this for 1 week and settled for the one with the most clicks.

7. Monitor the outcome

By the end of the month, we discovered that the conversion rate increased by about 46%.

In conclusion, using data-driven problem-solving techniques can help businesses identify and solve issues that impact their bottom line. In the case of our website redesign, we were able to increase our conversion rate by 46% by using Hotjar and Google analytics. We also used the Impact-effort matrix to identify and address prioritization issues. By taking a data-driven approach to problem-solving, businesses can not only improve their website performance but also build better relationships with their customers by delivering an optimal user experience.

You may be tempted to ask if I got the job. The answer is No, I didn’t. Maybe someone better did.

The End.

I talk a lot about product management, startups, and music on Twitter. Follow me?@phillips_tunde