We had a directional cue pointing toward our subscription button, and it was also a place to provide and additional piece of supporting copy.
And we were immensely surprised when it not only lifted conversions, it lifted them by 45%!
We loved this landing page that was very well designed, very aesthetically pleasing, but after this result, we're learning that there's so much more that we can test, and that pretty isn't always to our benefit.
If I had to guess at why removing the visual cue worked better without seeing it, I would throw out a hypothesis that across the traffic coming to this page, perhaps the general preference was to not be instructed so overtly to the call to action. Perhaps the instruction/cue didn't give users a chance to feel like they were making a choice, but rather being directed to take an action. Of course this is conjecture, but I do think there are audiences and visitor types that don't want to be told what to do, and want to feel as though they are steering the interaction with a site. It would be interesting to know whether there was any significant difference in the experiment results when looking at different visitor types as well. Any data around performance by visitor type/channel (for instance paid search vs. referral vs. organic)?
Jodie, that is interesting. I have definitely experienced that phenomenon with developer-centric niches. In one usability test I actually saw a developer curse at a website because the CTA was overtly large and intrusive.
