admin 
I’ve seen numerous studies about how serif fonts (the ones with the little “tails”–like Georgia and Times New Roman) result in much better readership than sans serif (without the “tails”–like Arial and Verdana). And, I’ve seen studies state the exact opposite. This is important: when more people read your ad, it means more sales.
Your font choice can directly determine if a user reads or simply skims what you have to say. That’s a huge issue for just changing one line of code in your CSS. I hypothesize that your typography choices can be the difference between someone skimming and someone fully reading and comprehending.
Because competitors love to steal ads these days, I’m not going to give you my verbatim ad copy, but I’ll explain the goal, give the conditions, and be very specific with my conversion data.
The setup
I had a multi-step conversion path. The user tapped an ad on their smartphone and they saw landing page A or B of my split-test. The landing pages were text based (logo, headline, and body copy). They would read either A or B of the landing pages, and then would continue to the next page.
The call to action
The call to action was below the fold of most mobile phone browsers. It probably wasn’t on the giant Nexus Note, but that’s a small percentage of users. By having the call to action below the fold, I force more readership of the ad. If I just have a big orange button displayed prominently, much more will hit continue without reading my sales copy. Non-readers are less likely convert if your ad is written well and the call to action is below the fold, but if your ad is written horribly, it’s probably best to just use above the fold graphic gimmicks.
The fonts
I used Georgia (for “A”) and Arial (for “B”).
The results
| Clicks | CTA clicks | LP CTR | Leads | CR |
| 6285 | 1587 | 25.3{5d3ddbe771dfd7baccbc708ede1f8581564c9b62644010c5b52d123c48304749} | 251 | 3.99{5d3ddbe771dfd7baccbc708ede1f8581564c9b62644010c5b52d123c48304749} |
| 6303 | 1648 | 26.1{5d3ddbe771dfd7baccbc708ede1f8581564c9b62644010c5b52d123c48304749} | 311 | 4.93{5d3ddbe771dfd7baccbc708ede1f8581564c9b62644010c5b52d123c48304749} |
“B” increased conversions by 23.6{5d3ddbe771dfd7baccbc708ede1f8581564c9b62644010c5b52d123c48304749} with 99.5{5d3ddbe771dfd7baccbc708ede1f8581564c9b62644010c5b52d123c48304749} statistical significance.
Analysis
Whether or not reading comprehension was higher with “B,” a similar percentage of users still chose to continue from the landing page to the lead form. My guess is that Arial resulted in more people actually reading than skimming.
But why would Arial cause better readership? Because mobile users are used to reading sans serif fonts. Look at a partial screenshot of multiple smart phones:
Photo Credit: jfingas but meme-modified.
Guideline: The default font is a sans serif on mobile. There is a smoother transition if you keep the fonts similar.
The important point is to use others’ studies to help form a testing hypothesis. Then, prove what will bring the greatest profit by running a split test on your traffic.
In case you’re interested, this is the mobile detection script I’ve found to be very accurate (much better than the previous script that I’ve tested): MobileDetect.