To every­one who found time to join my ses­sion at Adobe Sum­mit 2014 on Adobe Ana­lyt­ics tips and tricks: Thank you! I had a great time prepar­ing and pre­sent­ing, and hope­fully you found the con­tent engag­ing and instruc­tive. For those who did not get a chance to grab the hand­out after the ses­sion, it is now avail­able for you to down­load. (I apol­o­gize for the file size, which is about 65 MB, the cost of high res­o­lu­tion. I tried to get it down as low as I could.)

In the remain­der of this post, I am going to share the nine tips that made up the bulk of my ses­sion, sim­i­lar to what is pro­vided in the hand­out but with a bit more detail. Please share this post, in addi­tion to the hand­out, with any of your col­leagues who didn’t get a chance to attend my session.

Tip #10: Clas­si­fi­ca­tion Rule Builder

One of the top­ics that gen­er­ated more inter­ested than expected as part of this tip was reg­u­lar expres­sions. The best way to build flex­i­ble rules to add meta­data to your track­ing codes, prod­uct SKUs, arti­cle IDs, etc. is using regex. If you’re like me, you’ve tried a few times to mas­ter reg­u­lar expres­sions, and failed each time. I still haven’t really mas­tered them, but the Rule Builder makes it much eas­ier to piece together the matches you need by offer­ing a great set of exam­ples which you can copy and use with­out leav­ing the tool.

  • Key Take­away: Clas­si­fi­ca­tions make it eas­ier for you and your col­leagues to inter­pret and seg­ment your data by adding meta­data; Rule Builder makes it easy to add the meta­data auto­mat­i­cally, as data is col­lected. Use Excel for spread­sheets, not as much for build­ing clas­si­fi­ca­tion files!
  • Best Prac­tice: Use sub-classifications as addi­tional lookups to pop­u­late meta data another level deep. For each clas­si­fi­ca­tion that has sub-classifications, upload a smaller Excel file, which serves as VLOOKUP file for the clas­si­fi­ca­tion engine.
  • Doc­u­men­ta­tion Key­word: clas­si­fi­ca­tion rule builder
  • Addi­tional con­tent: Brian Au (blog post) and Matt Free­stone (part 1, part 2, part 3, and part 4)

Tip #9: Real-Time Reporting

Ses­sion atten­dees were treated to a photo of Wild Bill, the Utah State Uni­ver­sity bas­ket­ball super­fan, don­ning a hula out­fit in an attempt to dis­tract an oppos­ing free throw shooter. Yes, real-time ana­lyt­ics can be a dis­trac­tion if used inap­pro­pri­ately. How­ever, there absolutely are valid use cases for real-time data that apply across mul­ti­ple ver­ti­cals and busi­ness mod­els. What I shared in the ses­sion was just one of those that comes up again and again as I talk to cus­tomers all over the world: I need to know imme­di­ately when there is a prob­lem on my site.

  • Key Take­away: Real-time data can enhance your dig­i­tal mar­ket­ing oper­a­tions by pro­vid­ing insight into what con­tent or prod­ucts are res­onat­ing right now, but also what isn’t work­ing at all—what might be bro­ken or offline.
  • Best Prac­tice: To start, set up error report­ing in real-time, and make this data avail­able to teams that can take action on errors if/when they occur on your site. Adobe Dynamic Tag Man­age­ment or another TMS can make it easy to cap­ture this and other rel­e­vant data in a prop for use in the real-time report.
  • Doc­u­men­ta­tion Key­word: real-time report
  • Addi­tional con­tent: Me (blog post) and J.D. Nyland (blog post 1, blog post 2)

Tip #8: Tableau Integration

It was nice to see our friends at Tableau do a blog post (see link below) on this inte­gra­tion dur­ing Sum­mit. We know that many of you have been export­ing data from Adobe Ana­lyt­ics and import­ing it into Tableau, but this has always been painful because Tableau doesn’t natively under­stand which fields com­ing out of Adobe Ana­lyt­ics are dimen­sions, which are met­rics, and which are time ranges. The result is that you’ve had to do a lot of man­ual work to prep your data for import. This inte­gra­tion aims to solve that problem.

  • Key Take­away: Tableau is a pop­u­lar dash­board­ing and data visu­al­iza­tion tool, and Adobe Ana­lyt­ics is a key com­po­nent of your mar­ket­ing data. It is very easy to cre­ate Tableau exports out of Adobe Analytics.
  • Best Prac­tice: Use Data Ware­house to cre­ate Tableau exports for con­sump­tion by your Tableau instance. Click “Advanced Deliv­ery Options” in Data Ware­house to select Tableau Export. Then set up Tableau Server for auto­mated import using cmd­tab.
  • Doc­u­men­ta­tion Key­word: tableau
  • Addi­tional con­tent: Tableau Soft­ware (blog post).

Tip #7: Sim­pli­fied Left Nav

This one is all about help­ing your col­leagues self-serve with data more effec­tively. With a sin­gle click, you can turn the exist­ing left-hand nav­i­ga­tion menu—which is usu­ally 10–15 items long, depend­ing on which fea­tures you have enabled—into a stream­lined six-item nav­i­ga­tion; addi­tion­ally, the group­ings of reports are based on the way that users typ­i­cally ask ques­tions of data, rather than the tra­di­tional group­ing by the under­ly­ing vari­able type.

  • Key Take­away: Adobe Ana­lyt­ics gives you a ton of crit­i­cal data, which is great—but which can make nav­i­gat­ing the reports con­fus­ing for less-experienced users. The sim­pli­fied menus allow your users to find the data they need far more quickly and easily.
  • Best Prac­tice: Select the sim­pli­fied menu, then make edits to cus­tomize the sim­pli­fied, stream­lined menu to your needs. Go to Edit Set­tings > Gen­eral > Cus­tomize Menus, then click “Restore Sim­pli­fied.” Once the set­ting has been applied, you can make addi­tional edits cus­tomized to your busi­ness needs.
  • Doc­u­men­ta­tion Key­word: sim­pli­fied menu

Tip #6: Admin Con­sole improvements

Par­tic­i­pa­tion, Pathing, Mer­chan­dis­ing eVars, geoSeg­men­ta­tion, Event Seri­al­iza­tion, Hid­ing report suites from the drop-down menu, and more. This one is a win for every­body. It frees you up to spend less time on the phone with Client­Care or your account team. It frees Client­Care and your account team up to pro­vide higher-value interactions.

  • Key Take­away: Adobe Ana­lyt­ics offers unpar­al­leled flex­i­bil­ity and power around your on-site/in-app imple­men­ta­tion. With recent changes, you can now con­trol a dozen set­tings that were pre­vi­ously only avail­able by call­ing ClientCare.
  • Best Prac­tice: Expand your imple­men­ta­tion! The Traf­fic Vari­ables, Con­ver­sion Vari­ables, and Suc­cess Events pages in the Admin Con­sole allow you to turn on par­tic­i­pa­tion, pathing, seri­al­iza­tion, mer­chan­dis­ing, and more. You can also hide report suites from dis­play­ing to your users in the drop-down menu.
  • Doc­u­men­ta­tion Key­word: admin con­sole enable features
  • Addi­tional con­tent: Me (blog post)

Tip #5: Anom­aly Detection

You asked us to tell you not just when some­thing rel­e­vant changes, but when you should care about it. We took a big step in that direc­tion with Anom­aly Detec­tion. It’s worth call­ing out for empha­sis that this is also avail­able via the API. A sec­ond best prac­tice that I could have shared is to cre­ate a sim­ple script based on Anom­aly Detec­tion that runs the report each morn­ing and sends an email to inter­ested par­ties with its observations.

  • Key Take­away: No human can sense every­thing that is worth inves­ti­gat­ing, but Adobe Ana­lyt­ics can. Find the “unknown unknowns” by call­ing out sta­tis­ti­cally sig­nif­i­cant changes in trend.
  • Best Prac­tice: Build cus­tom “fil­tered” met­rics to focus in on anom­alies for spe­cific pages, sec­tions, prod­ucts, cam­paigns, or other dimen­sions for any met­ric that you choose. Click “Add Fil­tered Met­ric” when con­fig­ur­ing Anom­aly Detection.
  • Doc­u­men­ta­tion Key­word: anom­aly detection
  • Addi­tional con­tent: John Bates (blog post)

Tip #4: Adobe Ana­lyt­ics API

In my ses­sion at Sum­mit, I shared the Mar­ket­ing Tech­nol­ogy Land­scape Super­graphic com­piled by Scott Brinker over at chief​martec​.com. It includes 947 dif­fer­ent tech­nolo­gies across 6 major cat­e­gories and 42 sub-categories. The point is that no mat­ter how many Adobe Mar­ket­ing Cloud solu­tions you’re using, you are prob­a­bly being asked to facil­i­tate addi­tional inte­gra­tions with other tools. The Adobe Ana­lyt­ics APIs received 19 major enhance­ments over the past 12 months which make it much eas­ier for you to do this, includ­ing the abil­ity to bring Adobe Ana­lyt­ics data into R for cus­tomized sta­tis­ti­cal analysis.

  • Key Take­away: As you are asked to inte­grate dig­i­tal ana­lyt­ics data with other sources, APIs become key; they allow you to retrieve data pro­gram­mat­i­cally so that you can oper­ate or take action on them in your own appli­ca­tions or other systems.
  • Best Prac­tice: Use “RSite­Cat­a­lyst” pack­age with the R lan­guage to make easy-to-build Adobe Ana­lyt­ics API calls and per­form your own statistical/predictive analy­sis against your data. Start with basics—standard devi­a­tion, scat­ter plots, his­tograms, and build out from there. You can get RSite­Cat­a­lyst, doc­u­men­ta­tion, and the pack­ages that it requires at cran​.project​-​r​.org.
  • API Doc­u­men­ta­tion: devel​oper​.omni​ture​.com
  • RSite­Cat­a­lyst Doc­u­men­ta­tion: cran​.project​-​r​.org
  • Addi­tional Con­tent: Randy Zwitch (blog post)

Tip #3: Com­pare Segments

Com­par­i­son is every­thing. It pro­vides the con­text that you need in order to inter­pret your data cor­rectly and arrive at the right con­clu­sion. Reminder that you can always go to Ad Hoc Analy­sis to com­pare mul­ti­ple seg­ments and date ranges in a sin­gle view!

  • Key Take­away: Com­par­i­son is crit­i­cal in analy­sis, and Adobe Ana­lyt­ics has expanded its com­par­i­son capa­bil­i­ties to include seg­ment com­par­i­son in the web UI.
  • Best Prac­tice: Do basic com­par­i­son of seg­ments to deter­mine which prod­ucts, arti­cles, videos, etc. are most rel­e­vant to dif­fer­ent groups of users, then run tar­geted cam­paigns or offers to improve con­ver­sion for var­i­ous segments.
  • Doc­u­men­ta­tion: com­pare ana­lyt­ics seg­ments (this gives a spe­cific exam­ple of seg­ment com­par­i­son, but you’ll get the idea)
  • Addi­tional con­tent: Me (blog post)

Tip #2: Mobile App Life­cy­cle Metrics

Hav­ing designed a few mobile imple­men­ta­tions for prod­uct teams and mobile agen­cies in my day, I know how painful this can be. That is why life­cy­cle met­rics are so excit­ing: with lit­er­ally a sin­gle line of code dropped into your mobile app code base, get every­thing your app teams need in order to get started on report­ing and analy­sis. This also inte­grates with Adobe Mobile Ser­vices to allow you to do not just ana­lyt­ics, but also tar­get­ing and other mar­ket­ing activ­i­ties from a sin­gle tool.

  • Key Take­away: Adobe Ana­lyt­ics has made it pos­si­ble to imple­ment SDKs in your mobile apps and get rich, pow­er­ful data with min­i­mal effort from your devel­op­ers. One line of code and the inclu­sion of the SDK will pro­vide “life­cy­cle met­rics” (and dimen­sions) includ­ing installs, launches, crashes, device name, app ID, days since last upgrade, and more.
  • Best Prac­tice: Use the pro­vided Mobile Overview report as base­line report­ing for your mobile teams. Mobile cohort analy­sis should be your mobile prod­uct group’s best friend—and it is read­ily avail­able with mobile life­cy­cle metrics.
  • Doc­u­men­ta­tion Key­word: ana­lyt­ics SDK 4
  • Addi­tional con­tent: Ray Pun (blog post) and me (blog post)

What Hap­pened to Tip #1?

Tip #1 was some­thing com­ing soon to Adobe Ana­lyt­ics. I won’t rehash all of it here, but it’s some­thing we have heard as a request from many of you, which we will help your col­leagues use Adobe Ana­lyt­ics more con­fi­dently and with­out need­ing as much man­ual guid­ance from your ana­lyt­ics team.

There you have it! I’ll be post­ing addi­tional tips and tricks as time allows over the com­ing weeks. As always, I’d love to hear from you if you have tried any of these tips, whether you’ve got ques­tions, com­plaints, com­pli­ments, or any­thing else to say!


From someone who wasn't able to attend Summit this year, thank you for condensing the takeaways from the week long summit into an easily digestible post. Quick question: Tip #4 and Tip #8 - Isn't there an API for Tableau integration? Looking forward to tip #1 :)