Sunday 20 March 2016

vocabulary - Is there a way to measure how "emotive" a verb is?

Split Testing



(Honestly, that's all I can think of).



If you put out two versions of something with slight differences, and one of them performs better, than I suppose you would rank that one ahead of the other in terms of it's emotional impact. This would give you some sort of relative ranking, if not an objective one.



So, if you had a website with two versions, and "Ever feel like your business" resulted in a 10% conversion rate, while "Ever suspect your business..." only resulted in a 5% conversion rate, you would rank the one other the other. You could do this with any number of words, phrases, and layout elements, individually and in combination.



I think this could get a bit complicated though, as the impact of a word or phrase may be tied in with it's environment - other words, phrases, colors, fonts, images, etc., that change along with it, and you may find that two weak changes can be combined into a strong change, and how do you rank that?



I suspect a relative measure is the best we can get. I also suspect that you wouldn't be able to take that measurement and generalize to other uses of the word or phrase, as the context and 'environment' of a word or phrase will also affect it's impact.



I believe there is some split testing software that will let you try many different tweaks and find the best combination, and I imagine some of them have meaningful measures of the results. I'm not particularly familiar with them though, so you'd have to consult the Google.



(Obviously, this mainly applies to sales, or other copy with some sort of measurable conversion)

No comments:

Post a Comment