Wednesday, September 26, 2007

Do we care if it's Web Analytics 2.0 or 3.0?

Heck! We might live on different (virtual) planet, but I'm fascinated by our top web analytics minds gently arguing about which version of web analytics is most promising. I have a lot of respect for Avinash Kaushik and Peterson and reading both of their recent posts had me stop, think for a few minutes, and go back to my web analytics 1.0.

Web Analytics 1.0

From the Web Analytics Association web site:
Web Analytics is the measurement, collection, analysis and reporting of Internet data for the purposes of understanding and optimizing Web usage.
It's a good start! The web is the center of the universe, everything revolves around it. But it's amazing how many web sites are so badly constructed that we don't even need web analytics to optimize them. And then, the relatively small percentage of companies who addresses web analytics are barely using what's at their disposal...

Web Analytics 2.0

According to Avinash Kaushik:
"1) The analysis of qualitative and quantitative data from your website and the competition 2) to drive a continual improvement of the online experience that your customers, and potential customers have, 3) which translates into your desired outcomes."
Now we look at what surrounds us to understand our own role in the universe. Already better!

Web Analytics 3.0

While Peterson doesn't give a clear definition of what would be Web Analytics 3.0, his examples stresses "location" is key to Web Analytics 3.0:
"the next technology era will be characterized by our collective ability to access the Internet anyplace, anytime"
Ok! Now we will be able to know where we are, and where everyone else is...

Sorry, I digress!

It's always interesting to see where things could be heading, what they should be in a not so distant future. But as soon as we realize what we call the "Web" and the "Internet" exists only as part of a bigger ecosystem, where energy and communication flows from the real to the virtual, and vice-versa, we can only come to the conclusion that merely measuring the medium and it's immediate surrounding is not enough. Avinash focus on the online experience and Peterson finds the holly grail in the ability to precisely locate the source of the interaction. In my humble opinion, they are both right and wrong.

The Web and Internet ecosystem encompass quantitative and qualitative elements, physical and virtual organisms, online and offline interactions that are functioning together within legal, ethical and technological constraints. From that angle, things like a website, competition or location can't, by themselves, explain the complexity of what's going on. They can merely improve the science of analysis that will eventually lead to better insight.

It's too easy to fall in the "e-pit" where anything and everything becomes a reason to stick an "e" in front of a word. We shouldn't do the same in the web analytics world.

In the end, "web analytics" is to "analytics" what "ebusiness" is to "business". Doing one without the other is a waste of time and energy.

Shocked? You think I'm crazy? Did I miss something? Talk to me!

Monday, September 24, 2007

Instances of Referrers Report in Omniture

As a follow up to an inquiry on the Yahoo! Group about getting Visits by Referrers and an earlier post I did about Instances vs. Visits in Omniture, here's what you can do to get Visits by Referring Domain:
  1. Inference: Is the absolute number of Visits "that" important? Can you infer that the trend of Instances would be similar to the trend of Visits? In a quick test I did using Discover, I found the trends to be very similar but the scale was a couple of times bigger for Instances. In the end, your Top 5, or Top 10 (or the sites that brings 80% of your traffic) will remain the same regardless if you look at them by Instances or by Visits.
  2. Configuration change: Ask Omniture to do a configuration change for your account in order to track visits instead of instances. However, this might have some drawbacks in other reports as well. Here's why Omniture chose to show Instances instead of Visits: "First, multiple referrers can occur on a single visit. Second, potentially no referrers can occur on a single visit." (See KB #1046 and #711) (The second point is a bit weak in my point of view... if there is no referrer, there is no instance... same difference)
  3. Discover: As noted in a previous post about Discover, it goes way beyond what SiteCatalyst offer, and will allow you to get Visits by Referring Domain.

Wednesday, September 19, 2007

Register! Omniture training in Montreal

Here's a unique opportunity to get trained on Omniture with fellow web analytics practitioners from Quebec! Following the success of the first Omniture Café in Montréal, I've been discussing with Omniture to have training in Montréal. So here it is!

What: SiteCatalyst User training
When: October 15-16th

What: SiteCatalyst Advanced User training
When: October 17th

Registration: Omniture University Regional Training
Location: Complexe Desjardins in Montréal
Hotel: check out the Hyatt Regency

Do it quick, last time I checked there was only 8 seats available!

Why you should attend

  • If you are an Omniture user and never had this training: it's a must!
  • If you already had training, you should attend anyway:
    1. the course content will be tailored to our needs (training attendees will be able to pick the most relevant topics to them beforehand),
    2. and you will be able to share and learn from the local network of web analytics practitioners.
If you have questions, feel free to post a comment or email me and I'll try to help.

Anil Batra interview's me

In previous posts I referred to the sense of community and mutual collaboration that (still) characterize the web analytics industry. I'm thinkful to Anil Batra for doing his series of interviews with web analysts.

Anil interviewed me about my current position, how I ended up here, my background and my take on education and how to get in the field.

Hope you'll find it interesting, and of course, your comments are always appreciated!

Monday, September 17, 2007

Caravanserai of modern times

Last night my wife and I traveled to a different world, that of Loreena McKennitt unique musical environment. It was literally an immersive experience that psychologists call "a state of flow". All elements were carefully orchestrated to provide the highest degree of emotional impact: a pleasing and rewarding experience that made us travel to a different time, a welcomed escape from our fast paced and crazy life.


One of the song from the Ancient Muse album is entitled "Caravanserai". A caravanserai was a roadside inn where travelers could rest and recover from the day's journey. Caravanserais supported the flow of commerce, information, and people across the network of trade routes. I didn't think about it last night, but this morning I woke up still surrounded by the sound and the mood and it got me thinking about the contrasts of those times and today's age of virtual companies, remote communications, avatars and virtual worlds.
By supporting the flow of commerce, information, and people across the network of trade routes, the Web as become the Caravanserais of modern times.

Friday, September 14, 2007

Omniture is at it again! Now it's about Disney

It seems nobody else took note of this announcement: the Walt Disney Internet Group abandons WebSideStory (VisualSciences) and goes with the full range of Omniture products, as highlighted earlier in this short press release.

From my knowledge of the web analytics market, it's a huge blow at VisualScience as Disney was probably (one of) their biggest and oldest client. As you can imagine, Disney is not just your next schmuck web site, it's a conglomerate encompassing Disney Online, ABC, ESPN, and a bunch of others and one of the world's most renowned brand. In August, Nielsen/Netratings ranked Disney Internet Group at the 9th position of the most visited sites, with over 32M monthly unique visitors.

Blog advertising? Act locally!

I was contacted by renowned Quebec blogger Michel Leblanc (in French) who suggested to put advertising on my blog. His idea led to the creation of a new ad network specialized for the local market and fostering links between Quebec's leading blogs. There are a couple of unique aspects to it:
  • It's really targeted to local blogs and local readers, with ads about local events and local companies, what François Laroche (in French) calls "social proximity"
  • It will creates links between leading blogs and globally increase traffic to them
  • The ad format is unique, a 145x145 Flash pub that will be well positioned in the page layout
  • The new ad network is managed by, a leading business magazine in Quebec, owned by Médias Transcontinental, largest printer in Canada and leading consumer magazine publisher
  • The sponsoring of participating blogs and the revenue model is also quite unique
  • Each blogger keeps his own autonomy while being part of a larger "elite" group
For the reasons above, it's much better for me than Ad Sense, which hasn't proved to be very effective. A great initiative that gives a hand to local bloggers. So we'll see how the pilot run goes starting September 20th.

Thursday, September 13, 2007

Montreal's September WaW wrapup

I'll do just a quick recap as it's getting late and I've been mostly "unplugged" for the last 36 hours. If you were there, feel free to add your grain of salt!

Should have been there

Twenty people from Montreal's Web Analytics community met last night, some new faces, some old timers, and always a couple of people who misses THE most important get together of the web analytics community in Quebec! And the way it's going, the discussion is much broader than web analytics: web strategies, emarketing, design, etc. I now have 128 people on the invitation list, so 20 out of 128 isn't so bad, but I'm sure this will expand radically in the coming months.

About half of the group was composed of web analytics practitioners while the other half were consultants in the analytics, SEO/SEM and design field. The mix of people was also interesting, ranging from web analytics practitioners to SEO/PPC specialists to managers. At some point, I glimpsed two conversations at once, corporate managers discussing fishing gears and scheduling a trip on the left side, and other people challenging themselves to see who would be able to increase their ranking the most on Google on the right side! Hmmm...

Who said web analytics was hard?

We started with a presentation of a couple of slides from Peterson's RAMP presentation (we lacked time to present everything, but the idea was also to sparkle discussion, which it did!). RAMP stands for Resources, Analysis, Multivariate Testing and Process. While all agreed that "web analytics is hard". Without denying the great contributions and the wisdom of M.Peterson, the discussion took an interesting turn, with some points of view that challenged the de facto "guruness" level of what is put forth in Peterson presentation ("guruness", for lack of a better term, qualifies what comes from a Guru's mouth).

Is it worth it?

Another interesting discussion turned around the web analytics maturity level of companies. While "internal" web analytics practitioners strive to convince their boss of the critical value of web analytics, consultants basically take the stance that if a client doesn't engage in web analytics, it's probably not a client you want anyway. With the current lack of resources, the strategy might make sense! Those who don't recognize the need to "compete on analytics" are eventually condemned to die.


I took a few minutes to talk about what's going on in the our web analytics community:
  • WaW: Mark your calendars for the next Web Analytics Wednesday: October 17th, in Montreal. Send in your suggestions for a good restaurant!
  • Omniture Training in Montréal! Following the success of the first Omniture Café held in Montréal a couple of months ago, I was able to work out a very interesting proposal with Omniture. I'll post more about this very soon, but if you are an Omniture client or offering consulting, book your agenda for October 15,16,17!
  • Jobs! There's some great openings in Montreal: Coradiant is looking for a product manager, Canoe is looking for a data analyst and Acquisio is looking to extend it's team.
  • WASP: (I had to do a quick plug for my Web Analytics Solution Profiler!) I explained what the tool does and talked about the upcoming release, which is actually ready... but I need to find some time to package it and update the web site. So look for it soon!
  • eMetrics Marketing Optimization Summit, Washington D.C.: If you are in the field, you MUST attend eMetrics at least once.
  • WebCom Montreal, November 14th: Simon Rivard from Canoe, Jacques Warren and myself will be holding a panel on web analytics.
I might have missed something! Let me know if I did!

Tuesday, September 11, 2007

WaW, September 12th, Montreal is tomorrow!

Don't forget Web Analytics Wednesday is tomorrow! Wednesday, September 12th in Montreal! More info can be found here.

So far I have 15 confirmations from a wide array of companies: Hydro-Quebec, Bell Web Solutions, Bell Canada, Archambault, Vidéotron, Acquisio, Desjardins, Networld Media. Ranging from decision makers to web analysts, SEO specialists and consultants.

There is no formal agenda, but here's some topics that will be discussed:
  • Sébastien Brodeur from will present "why is web analytics so hard?"
  • I will present a quick overview of the Persuasion Architecture concept
  • And I have some exclusive scoops to share tomorrow!
It's not too late to register! Just send me a quick email, see here for location details.

Conversion doesn't start where you think

We might have implicitly known about this fact, but this time it is quantified in a newly released study by ComScore/Forrester: 50% of those who start a transactional process actually have no intention of completing it.

No wonder we often see 2% or 4% conversion rates when using a magnifying glass to look at our checkout process or lead generation forms.

That's why Persuasion Architecture stresses that conversion actually begin much earlier, and also why it's important to measure more than the Conversion Funnel. Conversion Beacons, Points of Resolution, Waypoints and Driving Points needs to be understood and measured. This doesn't mean you need a dashboard will a gazillion of KPI's! At the executive level, you still want to watch Purchase and Buyer conversions (see E.Peteron's "Big Book of Key Performance Indicators"), but at the tactical level, you need to dig much more to find the root cause of abandonment.

That's where a systematic improvement process kicks in. I was first introduced to SixSigma when working for a manufacturing company where it was first applied for production of recreational vehicles, and than extended in all spheres of the business. SixSigma was the rallying cry for the whole company (maybe sometimes too much!). But I can appreciate and apply this methodology to web analytics. In SixSigma terminology, I'm some kind of a Green Belt :)

When faced with a financial transaction, the 50% or so who didn't expect to complete the process gave the following reasons:
  • 23% wanted more product information
  • 19% where not ready to apply
  • 14% wanted to see if they qualified for the product
Even when they had the intention of actually completing the transaction online:
  • 12% changed their mind (what's missing is "why?")
  • 11% had privacy/security concerns
  • 11% wanted to speak to a human regarding the product
To sum it up, "As more consumers research and purchase financial products on the Web, the importance of understanding application abandonment will increase", said Forrester. You bet!

Friday, September 7, 2007

Market consolidation: Omniture + Offermatica

Omniture is certainly getting some higher ground in the web analytics sphere with the announcement of the acquisition of Offermatica.

With TouchClarity, Genesis, SearchCenter, Discover 2.1, enhancements to SiteCatalyst, Omniture is aggressively expanding. The challenge will be to continue to maintain and enhance flagship products while integrating all the recent acquisitions. But this is certainly a good move in a crowded market burgeoning with new innovations.

And between all those moves, this should be a signal to web analysts that their job will continue to expand to new horizons; this is great and exciting! Reporting wasn't enough, analysis is the value. Now, A/B and multivariate testing, SEO, PPC and emarketing campaign management, behavioral targeting, surveys and other types of analysis, without forgetting core system data integration are becoming even more important. To employers, it's also a trend to watch as web analysts demand increase and expertise remain scarce in the market.

Still a lot of exciting time ahead!

Tuesday, September 4, 2007

Tutoring for the UBC Award of Achievement in Web Analytics

Just got confirmation that I will be tutoring the course "Introduction to Web Analytics" from the acclaimed Award of Achievement in Web Analytics. This activity, complementary to my job of ebusiness architect and web analytics practitioner, is a great opportunity to share & learn.

Last May I had the chance to present the "Web Analytics for Site Optimization" workshop at eMetrics, a really condensed version of the course that was just a glimpse of the excellent curriculum put forth by the UBC. Tutoring will allow for a much deeper dive into web analytics.

The online model is especially well suited for professionals and a great way to increase one's value in the market. I'm sure my experience with Laval University's online eBusiness MBA has given me a solid ground about the e-learning advantages and challenges. I look forward to start the next session with a first group of about 20 students.

Since good news never comes alone, while browsing my ULaval emails I noticed I will be listed on the honor roll for a second year in a row for achieving higher grades in my ongoing eBusiness MBA (averaging over 88% after at least 50% of the program).

Web 2.0 in the enterprise: a no and a yes

In a recent article, Gartner warned us about the security related to Web 2.0 in the enterprise. Whenever Gartner says something it's like a bird song to the ears of management, especially IT management. Although a good article, it is sure to raise a red flag to managers who don't fully understand what the heck is Web 2.0 and more importantly, the philosophy that goes with it. Frankly, I found that article to be alarmist and I expect it to be used as a "proof" that Web 2.0 is not so great for enterprises, especially from an IT perspective.

Since I wouldn't like to get caught naked by my boss (well... in both senses of the term!), I recommend the fine article from ZDNet: A checkpoint on Web 2.0 in the enterprise by Dion Hinchcliffe.

For sake of discussion, I'm including two charts from the full article. Although long, I strongly recommend it (I already said that, did I?).

"Web 2.0 is an underlying set of principles... that have reached a tipping point that’s enabling brand new business models, unleashing a wave of innovative products, influencing public behavior on a large scale, and in particular, resulting in entirely new types of online businesses."

"... like SOA or SaaS, the ideas represented by Web 2.0 will take years for the majority of the world to embrace and make effective use of. ... Many of these will require serious soul-searching in business, not to mention overcoming the Innovator’s Dilemma, the latter which raises the interesting question of how do you adopt a new way of business that is at odds with your current way of doing business?"
In short, Web 2.0, just like the Web about 15 years ago, is challenging several aspects of the business functions, including IT. It shouldn't be dismissed by fear of the unknown or hiding behind a comfortable status quo that is bound to haunt us for the coming years. The best approach might be to pick a few quick wins, explore and learn even if its only for sake of initiating discussion and sparkling a cultural change... Which would already be something interesting for companies that have a conservative culture.

Job in Montreal: Research Data Analyst for

Canadians, and especially Quebecers know about Canoe, one of the heavy weight of the media industry in the Canadian market. I've heard they are looking for a Research Data Analyst to work in the Marketing/Research team.

The rest of this post is a transcript from the job posting (in French). If you are interested in this job, send me a quick email and I'll give you the contact information.

Description de rôle

L’analyste en recherche Internet, de niveau intermédiaire, est relativement indépendant, mais fait affaire avec de nombreux clients à l’interne. Il doit avoir d’excellentes aptitudes de communication verbales et écrites et doit travailler pour le client, d’abord et avant tout. L’analyse de données fait partie intégrante de ses fonctions. Le candidat choisi doit être capable de répondre à plusieurs demandes de clients à la fois et de respecter de strictes échéances.


  • Fournir des analyses de données et seconder l’équipe des ventes dans la préparation de présentations dans un délai raisonnable.
  • Créer des présentations exhaustives sur l’analyse et les catégories.
  • Fournir des mises à jour à l’équipe des ventes pour souligner les éléments importants de Canoë.
  • Créer des rapports réguliers/mensuels. Fournir des statistiques pour les bulletins de ventes mensuels.
  • Fournir des rapports de statistiques pour le service de la comptabilité ou autres.
  • Fournir les données nécessaires aux dirigeants (président, v.-p. exécutif, directeur financier, etc.) et aux autres services.
  • Gérer les ressources de recherche pour les statistiques (où les présentations et les documents de ventes peuvent être accessibles facilement).
  • Les sources de données comprennent : comScore MMX, PMB et Omniture.
  • Optimiser les statistiques et le classement dans comScore Media Metrix. Envoyer les changements d’URL (ajouts de sites, changements de catégories) à comScore.
  • Renouveler les noms de domaines de Canoë et de Quebecor.
  • Personne clé pour le PMB pour le retour à l’échantillon.

Objectifs personnels

  • Travailler passionnément pour le bien de nos clients et de nos employés
  • Remettre en question le statu quo
  • Démontrer de la confiance en soi et envers les autres
  • Accepter ses responsabilités ouvertement
  • Écouter de façon active et s’exprimer de façon claire et directe

Critères de rendement

  • Production de rapports réguliers
  • Niveau de satisfaction des clients à l’interne (représentants, gestionnaires)
  • Gestion des URL de comScore Media Metrix (tous les sites)
  • Production de rapports de catégories
  • Remise des projets selon les échéanciers [> 85 %]

Formation, habiletés et expérience

  • Diplôme collégial ou universitaire
  • 3 ans d’expérience en analyse de données
  • Expérience en analyse de résultats de recherche
  • Connaissance du français très importante
  • Aptitudes de communication orale et écrite éprouvées
  • Expérience dans une entreprise du domaine des médias un atout