Mobile_optimization_followup_targetI thought I would follow-up the Adobe Sum­mit, The Dig­i­tal Mar­ket­ing Con­fer­ence with a post to answer some ques­tions about mobile opti­miza­tion that we heard regard­ing mobile opti­miza­tion. As a quick recap, we encour­aged every­one to chal­lenge their mobile assump­tions and test in mobile web and mobile apps. To help every­one do this, we shared the fol­low­ing four steps:

  1. Deter­mine your suc­cess met­ric – think about what suc­cess means in mobile
  2. Design a test to drive mobile strat­egy – find out where you should be invest­ing in mobile by testing
  3. Keep it sim­ple – learn to crawl before you run with mobile
  4. Mobi­lize your mboxes – use the SDK for native appli­ca­tion test­ing and nor­mal mboxes for your mobile web experience

Ques­tion #1: How do I know where to start testing?

You may have a mobile web site with hun­dreds of pages and no idea where to start. You may have a port­fo­lio of ten dif­fer­ent appli­ca­tions and won­der where to begin. One place to find out where you can start test­ing is with your ana­lyt­ics. Find out what pages have a lot of vis­its or what appli­ca­tion has a lot of users. Once you’ve deter­mined this, think about the test you would like to design for that start­ing point. This will give you the most effi­cient out­come with the largest impact.

Another way to find where to start is by seg­ment­ing your cur­rent Test&Target cam­paign traf­fic into mobile seg­ments today. This will show you what per­cent­age of your traf­fic can be fun­neled into mobile test­ing. Here are some seg­ments you can put into your cam­paigns now.



1. All Mobile Devices (smart­phones and tablets):

  • Mobile -> Hard­ware -> Mobile Device– > equals (case insen­si­tive) -> 1

2. Non Mobile Devices

  • Mobile -> Hard­ware -> Mobile Device– > does not con­tain -> 1

3. Smart­phones:

  • Mobile -> Hard­ware -> Is Mobile Phone -> equals (case insen­si­tive) -> 1

4. Non Smartphones:

  • Mobile -> Hard­ware -> Is Mobile Phone -> does not con­tain -> 1

5. Tablets:

  • Mobile -> Hard­ware -> Is Tablet -> equals (case insen­si­tive) -> 1

6. Non Tablets

  • Mobile -> Hard­ware -> Is Tablet -> does not con­tain -> 1


Ques­tion #2: Does Apple have to approve all my con­tent that is in the test?

No. With Test&Target you can deliver con­tent to a native appli­ca­tion after it has been released to the app store. The con­tent returned by Test&Target is deliv­ered to the app when the mbox is fired.  What is reviewed by Apple is the code in the app that inter­prets what comes back from Test&Target. Con­tent returned by Test&Target can be mod­i­fied after Apple’s approval with­out hav­ing to go through the approval process again.


Ques­tion #3: What are typ­i­cal suc­cess met­rics in mobile?

There are many dif­fer­ent things a per­son can do on a mobile phone. Call your busi­ness, share a check-in to your store with their friends, or opt in for push noti­fi­ca­tions from your app. What is crit­i­cal is defin­ing what is suc­cess in your mobile envi­ron­ment to your busi­ness as a whole.  You will most likely find that despite all the cool new things users can do with phones, your suc­cess met­rics will be the same as they relate to increas­ing the prof­itabil­ity of your business.


Here is a list of some met­rics I’ve seen in mobile:

  • RPV (Rev­enue Per Vis­i­tor) — Either via in-app pur­chase, or in a web view. This is usu­ally an easy trans­la­tion from desk­top to mobile for eCom­merce sites with mobile check­out abil­ity. If con­sumers can pur­chase from you on the smart­phone, I highly rec­om­mend using this met­ric to mea­sure success.
  • Engage­ment – These met­rics are com­monly used by media and enter­tain­ment. They are also very sim­i­lar to what is used on the desktop.
    • Ad con­sump­tion
    • Recency — loyalty
    • Lead sub­mis­sion – This met­ric is often eas­ier to track because click­ing on a but­ton from the phone itself makes calls.


Ques­tion #4: How do I make my mboxes most effi­cient on mobile websites?

After putting in mboxes on your mobile site, you will want to make sure they are light­weight.  Try gzip­ping your js file to reduce the size of the file by up to 73%. In addi­tion, com­bine this javascript with other script on your page to reduce the num­ber of requests made on the page.  Also keep the num­ber of mboxes on a page to no more than 1–2 per page.  Use https://​devel​op​ers​.google​.com/​s​p​e​e​d​/​p​a​g​e​s​p​e​e​d​/​i​n​s​i​g​hts to see what per­for­mance gains can be made on the page. Your mbox.js will need to be in the head of your doc­u­ment prior to any mbox calls being made.

I hope these answers help you in your quest for mobile opti­miza­tion. As you chal­lenge your assump­tions by test­ing mobile, you will dis­cover a whole new world of oppor­tu­ni­ties. If you have other ques­tions that I haven’t addressed, please feel free to put them in the com­ments below.

Oh, and there is one more thing…

If you didn’t catch the Sum­mit pre­sen­ta­tion, or you’d like to review it, click here to access it. :)