It is rare in this world that we get to expe­ri­ence pub­lic vin­di­ca­tion of our ideas. So it was with excite­ment that I read on the SeekingAl­pha blog about Yahoo pres­i­dent Susan Decker’s com­ments dur­ing the company’s fourth quar­ter earn­ings call. I must have missed this when it first came out with all the post-Christmas excite­ment, but I revis­ited the earn­ings release just this past week in light of all the Microsoft/Yahoo chat­ter and what a gem I found!

Decker said that, mov­ing for­ward, Yahoo would use vis­its rather than unique vis­i­tors as the most rel­e­vant met­ric for track­ing the rel­a­tive suc­cess of Yahoo sites.

She said:

With con­sumers access­ing the web in so many ways, we’ve looked for a more uni­fy­ing global met­ric that’s more flex­i­ble across Yahoo’s and our part­ners’ prop­er­ties and use­ful across mul­ti­ple devices and geo­gra­phies. We expect to use vis­its to Yahoo’s global start­ing points and anchor sites to be the most rel­e­vant met­ric going forward.”

She points out that the met­rics that have been dis­cussed in the past, such as uniques and page views, “may not tell the story of what’s hap­pen­ing and the key, value-creating start­ing points for con­sumers and advertisers.”

Ha! Let me bask in the light of a swift moment of “I told you so.”

Almost exactly two years ago, I wrote a blog post in which I said that, as a met­ric, vis­its were far more use­ful than unique vis­i­tors when track­ing web per­for­mance. And, though I received plenty of kudos from oth­ers within the indus­try, it must be said that I got my share of “what could this wacky guy be think­ing?” type of responses. The guys at Future Now, for exam­ple, were par­tic­u­larly scathing in their grokdot­com blog.

My the­ory was based on a few sim­ple (or, as grokdot­com called them, “sim­plis­tic”) reasons:

1. Vis­its are more accu­rate than unique visitors.

2. Every visit rep­re­sents an oppor­tu­nity to per­suade or con­vert a vis­i­tor to a customer.

3. Mea­sur­ing vis­its is based on fairly estab­lished indus­try standards.

I explained each of those rea­sons in detail. I also con­tin­ued to explore the rea­son­ing behind my the­ory in other posts, includ­ing one where I laid out 15 rea­sons why all unique vis­i­tors are not cre­ated equal.

As I pointed out then, users access the Inter­net via a vari­ety of browsers and a vari­ety of com­put­ers.  Also, mul­ti­ple users can access the inter­net via a vari­ety of browsers on a sin­gle com­puter.  Users delete or accept cook­ies on var­i­ous browsers and var­i­ous com­put­ers. At any given point in time, then, these sce­nar­ios are being played out by your vis­i­tors to your web­site.  They are inher­ent in unique vis­i­tor counts and, by their very def­i­n­i­tion, make unique vis­i­tor counts com­pletely unreliable.

Today, the issue gets even more com­pli­cated.  Users access the Inter­net not only by a vari­ety of dif­fer­ent com­put­ers (their home com­puter, their com­puter at work, their husband’s or wife’s com­puter) but by a vari­ety of dif­fer­ent devices includ­ing Black­ber­ries and iPhones.

Times are chang­ing, and the world of web ana­lyt­ics must change, too.  If you didn’t take my word for it, back in 2006, that vis­its was a more accu­rate mea­sure than unique vis­i­tors, think about what Susan Decker of Yahoo said.  And Susan wasn’t alone either – sev­eral other large media con­cerns have explic­itly or implic­itly gone this same direc­tion.  Then, go back and read or re-read my 15 rea­sons unique vis­i­tors are not cre­ated equal.

Con­sider that this might be just the first indi­ca­tion of a sea-change that is tak­ing place in the world of web ana­lyt­ics. New devices, new tech­nolo­gies and new ways of using the web con­tinue to cre­ate the need for con­stant vig­i­lance on the part of web ana­lyt­ics com­pa­nies.  We can’t afford to sit back and rely on what has worked in the past.

That’s part of the rea­son I was so pumped when I read about Yahoo’s shift.  It means we at Omni­ture are mov­ing in the right direc­tion.  We’re suc­cess­fully stay­ing on top of the changes the web world is encountering.

I’ll con­tinue to explore these changes, and to share my thoughts with you.  Hope­fully, they’ll help you to more eas­ily nav­i­gate the choppy but excit­ing waters of a Web 2.0 world.  In the mean­time, let me know your thoughts, even if you dis­agree.  As always, I look for­ward to read­ing your comments.

Tagged with →  
  • http://tuckerc.blogspot.com Tucker Chris­tiansen

    Con­grat­u­la­tions Matt!

    Vis­its are defi­antly the bet­ter met­ric. Ive noticed at BYU that quite a few of the busi­ness stu­dents have their web browsers set to auto­mat­i­cally delete cook­ies every time they close down the browser.

  • Pingback: links for 2008-05-21 « David Black

  • http://FutureNowInc.com Howard Kaplan

    Hey, at least we never called you wacky ;) I apol­o­gized pub­licly on our blog ear­lier today. You’re right, we went beyond scathing, and unnec­es­sar­ily so.

    Next time you’re in NYC, let us know and we can con­tinue this con­ver­sa­tion in per­son. Who knows, we may even buy dinner…

  • http://www.google.com jimmy

    nice post

  • http://www.bttradespace.com Alex L

    Fan­tas­tic!

    On behalf of all the down-trodden web ana­lysts out there, can I request that you please keep bang­ing on about this. The sooner that the business/marketing users real­ize that UV does not equal ‘peo­ple’, the sooner we can get on with the job of pro­duc­ing MEANINGFUL data.

  • http://www.arthurfreydin.com Arthur Frey­din

    Hi Matt,

    I’ll begin by say­ing that I agree using vis­its as a more trust­wor­thy met­ric than unique visitors.

    What I don’t under­stand is how your real­iza­tion relates to Omni­ture “mov­ing in the right direc­tion”. How does Omni­ture evan­ge­lize the vis­its met­ric exactly? I’ve been on a num­ber of conference/demo calls with Omni­ture and have a great friend that is a sales exec @ Omni­ture and have never once heard them ped­dling the vis­its met­ric — Omni­ture was always a open-box tool; they pro­vide us with the data and we do what we want with it.

    So, with that said, I’m not sure that Omni­ture is in a posi­tion to be evan­ge­liz­ing a cer­tain KPI — that’s bet­ter left to its clients.

  • http://blogs.omniture.com/author/mbelkin Matt Belkin

    Thx for the feed­back Arthur. And you’re right, Omni­ture pro­vides an opti­miza­tion plat­form that is essen­tially “met­ric agnostic”.

    So it’s not sur­pris­ing that you wouldn’t hear many of my coun­ter­parts evan­ge­liz­ing vis­its in their daily inter­ac­tions. That said, given our vast indus­try expe­ri­ence, our clients often look to us for con­sult­ing on which met­rics to use, and how to inter­pret their data. The Vis­its vs. Unique Vis­i­tors ques­tion has arisen for years, and many cus­tomers ben­e­fit­ted from our 1:1 expla­na­tion of why Vis­its tends to be the bet­ter of the met­rics. I elected to post this to the Omni­ture blog because I felt all our cus­tomers could ben­e­fit from this perspective.

    Thx again!

  • http://michelinguide.com Steve Hunt

    Matt– if one con­verts from a met­ric of “uniques” to a met­ric of “vis­its,” what sort of ratio would you expect to see. For exam­ple, if one orig­i­nally wanted to reach 100k uniques, what would be a rea­son­able tar­get for “vis­its,” all other things being equal?

  • Kim Davis

    The fact that some­thing may be eas­ier does not mean it is auto­mat­i­cally *bet­ter*. If so, we would not try to progress…
    We need to give con­text, i.e. includ­ing lim­i­ta­tions, as part of our infor­ma­tion. Oth­er­wise it remains sim­ple data.

    Two rea­sons why vis­its give an incom­plete pic­ture:
    — they don’t account for a cru­cial piece of infor­ma­tion to retail­ers: loy­alty.
    Con­sider try­ing to get the ROI of a web tool sup­posed to aug­ment rev­enue by mak­ing users come back. Cal­cu­lat­ing the Rev­enue per Visit, you get a smaller num­ber since more users come back, but obvi­ously don’t buy every time, even if total sales go up. Only the Rev­enue per Vis­i­tor would prop­erly account for the added loy­alty effect.
    — they’re not counted when cook­ies aren’t enabled (mean­ing reports may have more vis­i­tors than vis­its. Which do you trust? Neither…)

    So we have to con­tinue per­fect­ing that information…

  • Pingback: Website Optimizer and Event Tracking for e-commerce | Web Analytics Blog | Web analytics

  • http://www.danielhollerung.com/ Daniel Hollerung

    From what I can tell, vis­its as a met­ric is bet­ter than unique vis­i­tors in 2011 too. With more mobile browsers, web browsers and share IP addresses it is per­fectly clear and still rel­e­vant today.

  • http://www.connetu.com/ Paul Web

    We can­not always be sure what a sup­posed top dog at a com­pany like Google or Yahoo says. Now that Susan Decker has been fired from Yahoo, we know now that she was unable to help Yahoo get to Google’s stage. For web ana­lyt­ics, there can never be a ded­i­cated tech­nique to boost traf­fic. Times always change, so do the ways to improve traffic.