When most of us talk about or plan a Site­Cat­a­lyst imple­men­ta­tion, we tend to focus on data col­lec­tion.  If you’re doing it right, you’ll start by defin­ing your busi­ness objec­tives and cre­at­ing a web mea­sure­ment strat­egy.  You’ll prob­a­bly dis­cuss how to build your JavaScript bea­con.  You may explore the option of using the Data Inser­tion API or inte­grat­ing an email or ad serv­ing ven­dor.  These are all vital com­po­nents, but there is another area that usu­ally doesn’t receive the atten­tion it merits.

Data Orga­ni­za­tion is Impor­tant Too

You cap­ture the data on your web­site or from your appli­ca­tion, then it goes to Omni­ture for pro­cess­ing, but where exactly does it go?

All the data that Omni­ture receives and processes is stored in a Report Suite.  One of the bits of data that Omni­ture col­lects on every request is the report suite ID, which ulti­mately deter­mines a data point’s final rest­ing place after it is processed along with the rest of the click stream.  The size of the report suite deter­mines the scope of all the data within in it.

Buck­ets and Buck­ets of Data

It may help to think of a report suite as a bucket that holds the data you’re inter­ested in.  You might have a bucket for your US site, one for your French site, one for your Ger­man site etc.  The coun­try man­ager respon­si­ble for the busi­ness in France is crit­i­cally inter­ested in every lit­tle detail of what’s hap­pen­ing on the French site (yes, I did the illus­tra­tions myself, how did you know?).

She’ll want to know how much rev­enue or how many leads are gen­er­ated.  Which cam­paigns, key­words, and refer­ring domains are gen­er­at­ing those events.  Is the new microsite dri­ving traf­fic to the mul­ti­me­dia?  Which of the home­page pro­mo­tions are most pop­u­lar?  Did they watch any videos or take the sur­vey that we invited them to par­tic­i­pate in?

She wants answers to each of these ques­tions as it relates to the French site.  The US and Ger­many are periph­eral because those sites fall out­side the scope of her respon­si­bil­ity.  Data col­lected from these other coun­tries is only an obsta­cle in her way.  We want to orga­nize our data col­lec­tion and pro­cess­ing so that she gets all of the data she needs and none of the data she doesn’t.  We need to orga­nize our data in a way that France can be eas­ily sep­a­rated from the other countries.

Big­ger Ques­tions Need Big­ger Buck­ets

In con­trast to a coun­try man­ager who wants to know every detail about an indi­vid­ual coun­try, a CMO is inter­ested in the big pic­ture items across the whole globe and prob­a­bly isn’t too con­cerned about an indi­vid­ual cam­paign run­ning in Japan.  Try­ing to aggre­gate data from each indi­vid­ual coun­try site is painful, time con­sum­ing, and intro­duces prob­lems with data accu­racy.  A CMO wants to look in a sin­gle loca­tion for data that spans the entire orga­ni­za­tion.  We need another bucket of data that has every­thing in it.

Archi­tec­ture

When we talk about a Report Suite Archi­tec­ture, we’re talk­ing about the hier­ar­chi­cal lay­out of the report suites (or buck­ets) within a com­pany.  It is the blue­print that tells Omni­ture how to put all your data together when it is received.  It deter­mines how big each report suite is, where it fits in rela­tion to the oth­ers, and which data goes into it.  You can orga­nize a hier­ar­chy around many dimen­sions, but the right one has to do with your orga­ni­za­tion and how your stake­hold­ers think about their business.

Orga­ni­za­tion Tech­niques

In my expe­ri­ence, there are 3 pri­mary lines along which an ana­lyt­ics pro­gram man­ager  will choose to orga­nize report suites:

  1. Geog­ra­phy
  2. Site/Section
  3. Prod­uct Line

If you think about how the chain of com­mand runs in your orga­ni­za­tion, it’s likely that one of these three dimen­sions is a good fit already, and a log­i­cal choice.

You can use JavaScript to direct your data into 1 or many report suites by comma-separating report suite IDs in the s_account vari­able.  Another com­mon tech­nique is to let VISTA do the work by look­ing for par­tic­u­lar val­ues in the page URL or a coun­try vari­able that you’re coding.

The Omni­ture Suite

Within the Omni­ture suite, we’ve laid out our report suites by Prod­uct Line.  Each indi­vid­ual prod­uct report suite rolls up into the Omni­ture Suite report suite (I’ve not pic­tured them all to save on space).

Prod­uct man­agers and engi­neers can get data about prod­uct and fea­ture usage, prod­uct adop­tion, etc., the indi­vid­ual report suites, while the global level pro­vides a holis­tic view of cus­tomers across the whole online mar­ket­ing suite.  If you’ve seen the prod­uct wheel when you login into Site­Cat­a­lyst, that hap­pens to be a pretty fair rep­re­sen­ta­tion of how we’ve orga­nized our report suites (raise your hand if you noticed that Insight was recently added to the wheel).

Archi­tect­ing a hier­ar­chy for your report suite helps you deliver the right infor­ma­tion to the right peo­ple, with­out drown­ing them with the unnec­es­sary pieces.  In my next post, we’ll take a deeper dive into the report suites them­selves to dis­cuss mak­ing the most of the indi­vid­ual data points.

2 comments
Ben Robison
Ben Robison

Reuben, If there are business questions that require a hierarchy like that, then it's a great idea. I've seen hierarchies that go up to 4 & 5 levels deep. If the regional GMs need the detailed metrics at the deepest levels of granularity, then three or four levels of a hierarchy may be the best solution. If they only need aggregated high-level metrics, then you might be able to answer the questions at hand with a prop or an evar, which is less complex than managing multiple levels of a hierarchy.

Reuben Poon
Reuben Poon

Great post, Ben! What are your thoughts about more advanced segmenting of these "buckets"? For example, what if a company wants to do multiple levels? Let's say that the organizational technique is geography. But above individual country, there are regional GM's that want to see collections of countries. So you might have EMEA that wants to see all of Europe combined. Then APAC, Americas, etc. This would be a three tiered report suite structure. What are the pros and cons of doing it this way? What are the alternatives?