callyway's score

Process - there's more to it than a document

Just writing about it is not enough ...
Just creating the document that defines it is not enough ...
You have to take it further:
    Don't just define it, create it.
    Don't just write it, build it.
    Don't just think it, do it.

I have been working on documents - wondering if something will come after the words.
That's why I decided to write these words.
All the words and plans and thoughts won't matter if you send those words out into the organization - and then nothing happens.

Action. Take it.

How do you get the buy-in needed to make it happen?

There's the exec sponsor route, of course - the old standard: do as I say because I'm the boss.
There's also the involvement route: make it a team effort and maybe it will work.

And then there's this weird other way --- I think of it as the middle way --- it involves everything and nothing.  It is the east meets west (and north and south) way.  And that is a rather mystical way ...

Yes, indeed.  Seriously, truly, actually.

I am talking about working within your system - whatever that system is - understanding it enough to know if you need an exec sponsor or maybe all you need is a nod.  Certainly you will need to work with a variety of individuals: so that means understanding their needs, their objectives and how this process will help/hinder those.  Does the process cross departmental lines?  If yes, then the middle way probably includes the involvement route.  And so, the weird other way (the middle way) is formless in that it is flexible enough to meet the needs of the environment: nothing more, nothing less: just what is needed to make it work.

The middle way for building a process that is used reflects the organizational constraints imposed and molds those constraints into the required framework to meet the objectives.  

Wow.  Did I really just use all of those words in one sentence???

Yep. I like it.  Seriously, truly, actually.  Why?  Because, though the sentence is filled with biz-lingo, the point is to build a process that is used ...... and that sentence describes the how-to for doing so. 

Now, to the Web Analytics of this post ..... when trying to define an analytics process .... there are a quite a few presentations about it and probably at least one or 32 blog posts about it .... so why say anything more?

Because it seems that tracking is still forgotten; that folks are still working quickly to tag after the fact due to, you guessed it, lack of process.  Now those folks may not even realize that is what caused their "hurry, gotta, we missed it" run to tag effort ... but that's it exactly.  If you find yourself chasing rather than leading: you've definitely lost or never found your process.

Recommendation: get yourself an analytics process.  Write, build it and figure out how to implement it within your organization (using the middle way, of course).  Quick.  Do it soon.  The only way to leverage the velocity of the Internet is to leverage the velocity that a well-used process delivers. 

Seriously, truly, actually. 

What if ...

you are given only 2 hours to analyze the checkout process of a website ...
where would you begin?

Would you jump right to the cart page and review it?
Would you add products to the cart and actually make a purchase - reviewing each page along the way?
If the latter, which products would you choose?

How much time would you allocate for building the presentation?
Would you review data or just review the process without trying to understand the available data?

What if ...

2 hours is not much time - particularly if the data is available - because you must use it if at all possible.
Does it make sense to push back and request additional time?
You bet.

Does it make sense to request a meeting to discuss what has been tried (if anything) -- to gather some history, some perspective, and potentially understand what plans have been made for updates?
You better.

Discovery is never a bad thing.
Discovery provides you, the analyst, with context and understanding.  It also means that you may not head down the same road ... no reason to recommend something that is already on the "to be completed" list.
How would you know that unless you ask?

Discovery - it's the meeting where you learn what you don't know, where you ask questions that may seem basic - but are very important to capture.  Take the time.  Push back.  Get the meeting before the expectation of a completed analysis even enters their minds.

Details.  Details.  Details.
Process.  Process.  Process.
The analytic process must always begin with discovery -- even when you work for the company rather than in an agency/client situation: discovery is merely a means to gather needed information prior to starting the analysis.

one small shift

Time gets away - feels as if I blinked and now it is July!
My last post was in May ---- crazy busy? Yes.

Since that last post, I have experienced a small shift.
I didn't move along the line -- the line shifted.
Do you remember that type of shift in basic economic analysis:
the entire curve shifted .......

The shift started even before the trip to Rose Mountain; but I didn't know it.
While there, it was official: one small jump of the curve.

And now, though I am back, I am not.
And now, though I am multi-tasking during the day, I am focused.
And now, that one small shift feels like a leap ... beyond faith.

I imagine that I can keep it close - through practice.
Maybe. Just maybe.

Time to load pictures ... wonder how much time is spent on average doing that ...

Speaking of average:
I wonder how many web analysts are trying to explain changes in average time on site...
or average order size (AOS) ...
or average order value (AOV) ...
at any one time on any Monday in America ...
Of course, I would need to make the decision to consider only one of those.
Then I would need to understand, at least roughly, how many web analysts there are in America.
Then I would need to identify how many are working in the vertical (if I chose AOV or AOS) ....
Hmm ... maybe understanding that is not all that important to me...

Maybe, just maybe, that's something you might want to do before you start running a calculation:
consider how much effort you'll have to spend to gather the data and make the calculation compared to the value of knowing the value of the data collected. 

To know for the sake of knowing ... nah. 
To know so that you can make a decision: maybe.
Depends upon your decision and the cost of gathering.
a small shift: consider the cost / benefit before jumping.

Why should I care?

At least 3 blogs (Mike Keye's, Marshall Sponder's and Michael Helbling's) have commented on the Eric / Avinash deal (for lack of a better descriptor)...

My thought:: why should I care?
The questions that surround the "deal" seem more toward the "what's going on with Britney" fascination than anything else.  Oh my -- let's see --
        Is it a clash?
        Is it a ripple in a relationship?
        Was there ever a relationship?
        What was the relationship?
        Are they friends?
        Are they friends joking?  Are they....

I don't understand.  But then I don't understand the celeb-stuff either.  Is it just something to talk about when it's perceived there's nothing else to talk about?  Or is that some folks love to speculate aloud? 

I think the most important question is:

        Why does Eric get "top billing" when Avinash's name begins with an 'A'?

The only thing that this may prove is that the question is always as important as the answer.

Moving on ...

Going mobile ...

I finally have a blackberry -- so I'm officially mobile now.
Reminds me of the old Who tune: going mobile ...
I used to ride home from college singing that song at the top of my lungs with the windows down: no air in the old car ... summer in Texas ... no radio either ... so my singing was all there was!  I thought the 4 hour drive was only 20 minutes or so thanks to that plus a few other tunes.

Now there's something else to it ... there's a lot more to the idea of "going mobile" these days.
In fact - it is what every web manager needs to think about if they haven't already been doing so.
We (the U.S.) are finally getting our mobile-ness in order and ready ... our devices may still be dated, but we are texting and searching and even ordering on our phones: so get ready for the rest.

What rest?
I don't pretend to know that (that's why I read Steve Smith) -- but I know this year there will be a BIG push (i.e., The BIG) toward mobile ... so we, in the analytics world, need to understand and know what to expect.  We need to think and talk and share.  We need to prepare the questions so that our clients are ready ... so that our employers understand ... so that we can eliminate the problems that typically tag along with any type of technology shift.

The Who can sing it and we can live it (and then measure it and report on it and re-tag it and loop back).

Note: I found this other post about mobile while searching for relevant mobile links.

Word of the day: diaspora

As I was reading Steve Smith's mobile article today, I came across the word: diaspora.  What a great word!

Of course, I had to go to Wikipedia to look it up even though I thought I knew what it meant -- I wasn't sure given the context in the article.   

Wiki is an excellent tool!                                                    Where were wikis when I was in school?                                                Oh, right, the dictionary and library were my wiki.                                          Back to the word though ...

The meaning of the word diaspora coupled with the use of the word in the article made me think about how many of the Web roles (Web analyst, Web marketer, Web manager) are migrating into the off-line agencies or into the off-line marketing departments ... which then made me think of "resistance is futile" -- assimilation is inevitable.  But, what does that have to do with Web Analytics?

Web analytics really is only a portion of the business intelligence that is needed for business decisions -- so it makes sense for the migration into either the marketing group or the strategy group -- rather than a stand-alone Web group.  Seeing the Web data within the total business context is important.  Understanding the relationship between channels is obviously important; reading the details of when and who is even more important (and complex).  Those will help the business reach the point of understanding why.  And that ---- the elusive why is the key.

Web analytics versus ...

What else is there?      What else would I be doing if not this?
I wouldn't be looking for the one single answer to the truth ... since it doesn't exist.
I wouldn't be tossing coins into a fountain waiting to see whether I get a heads or a tails and then calculating the rate of the loss in value of the coins while each flew through the air.

In fact, I might be developing relationships quietly while walking closer to another library.
Or maybe I would be in a library while walking quietly to closer relationships.

Irony is all around us --
    and the web analytics field is filled with it ...
    as is the political field (New Hampshire primary).
And now I can only think about tea...
Because I have a friend that sent me a special gift of tea (Adagio YUM!) and I wish I had brought that with me on this week of getting up and running with the new company.

So what then does all of this really have to do with web analytics?
I am working in the field and loving it though I am sure there are other things that I would enjoy ...
I'm going to get to "do this" AND work in my favorite library so it's not web analytics VERSUS ...
It's web analytics AND ...

Web analytics this year ...

So what will be the deal this year?  What will be the thing that we buzz about?
Besides tracking 2.0 and shifting further along the event line ... what will be The BIG?

Several folks say it must be mobile (see Eric's post for one) ... I think that's obvious (as does he).
The question about The BIG, for me, is will it significantly change our methods or the measures?
Yes we need to know the channel and yes we need to understand if it's the same visitor (available now if s/he opts in and logs in) -- what else will we need to know?  Or maybe the question is what else will we want to know?

With all of the recent Facebook privacy concerns: need and want are two very important distinct words now that all of use must understand.  Just because you can doesn't mean you should ... or that you have the authority to do so.

More thoughts on The BIG later ...

Preferences ...

It's easy to forget just about everything these days ... thanks to our handy tools such as contact lists that hold the memory so that our minds can be used for other important ideas such as remembering that there is a preference feature in iTunes that actually makes sense.

While loading some songs last night and today, I realized that my preferences must be wack-o --- since, in order to have something other than "Track XX" display, I was typing in the song name.  Why?  Because my advanced import preferences didn't have a check where a check was needed: to find the names of the songs on the Internet.  Oh thank heaven for the Internet!

After correcting my iTunes preferences so that the songs load quickly, I turned my thoughts to other preferences -- such as those found in marketing.  I know that many marketers, when building a form for a Web site, ask for every type of preference.  This extreme version of a preference form is meant to capture as much data as possible: so that they can fill their little database and make it a big database.

Sometimes these forms ask for too much; in fact, Jakob Nielsen recognized "cumbersome forms" as a top design mistake in 2005. 
When is a preference only a preference and not an invasion of privacy?
One clear answer: when it's in iTunes ... at least I hope so (I could be wrong). 

But then, I tend toward conspiracy theory when I start to think about the databases that I'm already in -- and it becomes near obsession when I think about the future databases that I'll be in...
or maybe those already exist in D.C. ??

This post has rambled from memory to preferences to conspiracy ... so where's the Web Analytics in that?
Web Analytics will help you determine if you've allowed the marketers to step over the line to near privacy invasion.  How so?
By monitoring your forms or changes in your forms: use the famous bounce rate as a measure (you'll notice that Grok's description is not about monitoring forms -- but a form is a page and thus it makes sense to use this measure).
 
If you make a major change or build a brand new form -- you may even want to monitor each field within the form: where specifically does your visitor drop out?  Before moving forward with field-level analysis, though, you'll want to weigh the cost of data-capture against the benefit: many reporting tools today charge for each event (rather than by the page view).

Holidays and more ...

The holidays started for me on 12/20 -- when I learned that my favorite Marine was home for the holidays!
That meant that I didn't have to drive to North Carolina to take Christmas to him -- what a great surprise.
The next day was my last day with Fry ... even though I had "so long snacks" on Wednesday.
I start my new job on 1/2 ---- I'm moving from the agency side to the vendor side of the Web Analytics world --- how cool is that?  Extremely.  Going to work for Coremetrics -- thoroughly and completely excited.

Sitting on the sofa right now with Christmas all around me -- it's difficult to think about Web Analytics or even a new job.  I'll get back to thinking about both on Wednesday 12/26 when I am setting up my office in my home (working remote will be great).

So for now, I'll drink some of my first Christmas gift (from a friend) ... some Adagio Tea. Yum!
And enjoy the warmth that it brings -- just like the holiday!