Hal­loween is around the cor­ner with ghosts, gob­lins and other scary crea­tures abound­ing. One thing you don’t need to be scared about as you embark upon an opti­miza­tion pro­gram is a Test&Target mbox. In fact, they are quite the oppo­site of scary, giv­ing mar­keters the flex­i­bil­ity and full con­trol to dis­play var­i­ous alter­na­tives of con­tent for test­ing or tar­get­ing to see sta­tis­ti­cally sig­nif­i­cant results on what com­pels dif­fer­ent cus­tomer groups to con­vert more.

Wait, what’s an mbox?

I’ve actu­ally heard it referred to as a “mag­i­cal box” which, in my book, is a pretty good descrip­tion.  In essence, it’s a mar­ket­ing box or mar­ket­ing con­tainer  that you typ­i­cally “place” around spe­cific areas of your web­page where you can swap out vari­a­tions of con­tent for A/B..N, mul­ti­vari­ate tests, and tar­get­ing. Mboxes can also be used to track vis­i­tor behav­ior and suc­cess met­rics on a page (clicks, shop­ping fun­nel, orders, etc.).

Mboxes auto­mat­i­cally cap­ture anony­mous pro­file infor­ma­tion about your vis­i­tors (such as refer­rer, key­word searched, geo-location, device, if you are a first-time or repeat vis­i­tor, among oth­ers). With mboxes, the pro­files are also pro­gres­sive, mean­ing, as your vis­i­tors nav­i­gate through the site and inter­act with con­tent, mboxes cap­ture their pref­er­ences and intent and then mar­keters can use that data to pro­vide in-session behav­ioral tar­get­ing based on their inter­ests and affini­ties. Lastly, you can pass in your own pro­file data, which you can then use for per­son­al­ized targeting.

This allows mar­keters to define cus­tomer seg­ments based on hun­dreds of very dis­tinct data vari­ables, such as cat­e­gory affin­ity, device type, entry point, and geo-location. They can then tar­get them with a pro­gres­sive dia­logue and expe­ri­ence at each stage of their path through their con­tent towards increased con­ver­sion, adapt­ing to the chan­nel and time by which they access.

  • Flex­i­bil­ity: You can have an mbox at the top of the page, for a sin­gle line of code imple­men­ta­tion, and/or mboxes around spe­cific con­tent areas, it’s really flex­i­ble. Dif­fer­ent imple­men­ta­tions may be war­ranted based on resources avail­able, and the matu­rity of your opti­miza­tion pro­gram. In fact, some of our clients use a hybrid approach to allow for full flex­i­bil­ity. For exam­ple, with an mbox at the top of the page (which allows you to cre­ate mboxes any­where else on the page and mod­ify css, etc), you have a lot of con­trol, but would need help from a tech­ni­cal resource or to uti­lize your Adobe Dig­i­tal Test&Target con­sul­tant. For an in-house opti­miza­tion pro­gram with full mar­keter con­trol, once mboxes are on the site, there’s lit­tle to no addi­tional need for IT, unless you are run­ning a com­pli­cated test that requires func­tional changes.  
  • Breadth: You can have mul­ti­ple mboxes on a page and across mul­ti­ple pages to main­tain a con­sis­tent and rel­e­vant cross-page expe­ri­ence. What some may not know is that you can also use an mbox off­site within emails, tar­geted ban­ner ads, social media, and on your mobile sites & apps. Wait, I can test my Face­book pages? Yes, you can!  
  • Con­trol: Mboxes can eas­ily be turned on and off within the Test&Target admin inter­face (with­out hav­ing to change code on your site) based on test­ing and tar­get­ing needs at any time by the mar­keter and do not dis­rupt the reg­u­lar func­tion of a site. So, what this means is, you don’t have to worry about overages!
  • Foun­da­tional: Mboxes are also the frame­work used for Adobe’s auto­mated behav­ioral tar­get­ing with Test&Target 1:1 and Adobe Rec­om­men­da­tions. They inte­grate eas­ily with the other prod­ucts in Adobe’s Dig­i­tal Mar­ket­ing Suite – you can drag/drop mboxes from CQ (our Web Expe­ri­ence Man­age­ment solu­tion), test your Search& Pro­mote busi­ness rules, cre­ate dynamic ban­ners with Scene7, and share seg­ments for advanced tar­get­ing with our Ana­lyt­ics solutions.
  • Safe: To safe­guard against ser­vice inter­rup­tions and con­nec­tiv­ity issues, all Test&Target mboxes are equipped to include default con­tent (defined by the client), which is sur­faced in the event that the user’s browser can­not con­nect to Test&Target. Default con­tent is trig­gered if the user’s browser can­not con­nect to Test&Target within a defined time­out period, which is configurable.

Um, really, what’s an mbox? (Technical)

Test&Target is a content-targeting appli­ca­tion that is inte­grated with web­sites by means of a JavaScript library. Spe­cific areas of high-value web­pages on a customer’s site are des­ig­nated with “mboxes” and replaced with a JavaScript func­tion. An mbox is a div tag that wraps around the exist­ing con­tent on a Web page. The div tag is then fol­lowed by a sin­gle line of JavaScript to cre­ate the mbox instance. We also have an Ajax method of invok­ing an mbox, if needed.

Other opti­miza­tion tools that imple­ment via “one sin­gle line of code” say that no fur­ther code is needed for test­ing, and this is not true. They require you to use their con­sul­tants to imple­ment tests and make any changes (there’s a lot of behind the scenes stuff that goes into this setup), which can be costly and takes away your con­trol. What isn’t said is the poten­tial risk in let­ting that DOM manip­u­la­tion or 3rd party ven­dor inject code onto your site.

Adobe does offer a full-service option with Adobe Dig­i­tal, if you pre­fer an agency model and want a 3rd party ven­dor to run your test­ing pro­gram. This team runs the gamut from strat­egy through analy­sis, includ­ing cre­at­ing designs, setup, cod­ing, and QA. Or you could take a hybrid approach, where your opti­miza­tion team does some test­ing on their own and uses Adobe Dig­i­tal in con­junc­tion as they grow their program.

Will mboxes slow down my site?

No, no opti­miza­tion prod­uct runs quicker than Test&Target due to our major invest­ment in an unri­valled global EDGE net­work. We deliver 25 bil­lion server calls per month with ease.  We have 750+ cus­tomers and some of the most respected enter­prise level cus­tomers in the world, includ­ing 4 out of the 5 top finan­cial insti­tu­tions and 8 out of the 10 top retail­ers.  Our impact to page load time is less than min­i­mal, tracked at only a few milliseconds.

Load time isn’t an issue, even for the largest of our enter­prise clients who receive bil­lions of vis­i­tors per year. We rec­om­mend keep­ing the num­ber of mboxes to under 7 per page, which appears to be more than enough for solid test­ing on any given page. For assis­tance or ques­tions, I highly rec­om­mend you lean on your Test&Target con­sul­tants to guide you;  they are experts in the tool and in the opti­miza­tion field. We’ve seen the great­est suc­cess from clients that learn and fol­low our best prac­tices on imple­men­ta­tion, test­ing, and build­ing an opti­miza­tion pro­gram overall.

No skele­tons in this closet

So, why be scared? The answer is you shouldn’t be, there are no skele­tons in this closet. Test&Target gives mar­keters full flex­i­bil­ity, power, and con­trol in a robust test­ing and tar­get­ing solu­tion.