CTR is high because version B is hard to read. So more people have to visit the website to read more. It is bad because we pay for each click in CPC model. Fail

CTR is high because version B is hard to read. So more people have to visit the website to read more. It is bad because we pay for each click in CPC model. Failing to weed out uninterested people leads to increased cost of aquisition. So measuring CTR does not tell anything about if version B is better. It may be worse. To compare we'd need to check CPA with A/B testing.