Improving CTR on a landing page

Summary

I designed a change to a landing page aimed at removing barriers in the first decision step, with the goal of raising CTR. The A/B test showed a CTR improvement (+16 %), which also supported revenue (+14 %).

Under an NDA, I’ve left out confidential details in this case study.

The starting point

The chosen landing page was one of the highest-traffic ones. The problem wasn’t getting users to the page — it was convincing them to take the first committing step without being sure they’d get it right.

The consequence was a lower click-through rate than on other pages, and from a business standpoint we were losing potential revenue.

Hypotheses

H1 — Removing the input fields lowers cognitive load and raises CTR (at least +10 %).
H2 — Changing the hero and button copy convinces the user to fill in the input fields (at least +10 %).
H3 — Switching the layout from one column to two improves how the product’s value is communicated (at least +10 %).

Only H1 was tested, and it was confirmed. The other hypotheses weren’t prioritized given the ratio of expected impact to technical cost — but they remain an opportunity for follow-up testing.

The solution

We removed 2 input fields that were required to click the conversion button. The idea was to defer the commitment — a user could express interest with a single click, without having to hand over data right away.

The result

The A/B test ran for 3 weeks on a sample of over 4,000 users, so the results were statistically relevant (min. 98 % significance).

Removing the input fields lowered cognitive load and cleared up the ambiguity — the need to fill in fields only appeared after clicking the button. As a result, the conversion rate went from 24.87 % to 29.07 % (+16.77 %) and revenue rose (+14.13 %).

Insight

The problem wasn’t that users didn’t want to continue — it’s that they didn’t want to start “the wrong way”. Input fields in the first step signaled commitment before the user was ready to make it, which led to an early exit.