Monday, October 30, 2006

Web Analytics conversations

Last updated: March 11th, 2008
How do you follow the most interesting conversations about Web Analytics? Subscribe to the most complete aggregated feed of blogs discussing about web analytics!

Current conversations

Add to Google Reader or use your own reader RSS
All of those are also included in the Web Analytics Co-Op Search
Missing your blog? Post a comment and I will fix that!

Optimize for international visitors?

As quoted from "The Unofficial Google Analytics blog" in a recent post entitled "The promise of overseas visitors":
In general, people don't optimize their website for non-USA traffic. It never crosses many people's mind. Why would it? Who else would be looking for your site, anyway?
I find it a bit surprising there are still people who don't optimize with internationalization in mind. Unless you offer a very narrow and local content or service, I think international visitors should always be granted with respect. If you don't want or can't support international visitors (because of commercial or logistic constraints), that's fine, but let them know as soon as possible!

Here in Quebec we are very sensitive and accustomed to this reality: we speak french while being surrounded by english culture. How many times to we end up on sites offering a french version assuming we are in France? How many times do we end up on a site to be abruptly abandoned after many promising steps to finally reach the lethal questions: state and zip code. Sorry "canadian", your "persona non grata".

Just for the sake of discussion, while the blog where this post came from shows 70% traffic from the US, my blog, with similar interests, shows radically different geographic spread:

All things being equal, notions like "foreign" and "overseas" don't make much sense anymore... we're all a bit "foreign" and, well, we're certainly all "overseas" somehow!

Wednesday, October 25, 2006

Google growing larger than Microsoft?

Update, Novembre 6th: This post has been updated to reflect some comments following its post to Digg, shown as one of the top stories under Tech Industry News.

The news tip

The source: A small note on MarketingPilgrim refers to the Boston Globe own reference to Fred Wilson's summary of an HipMojo blog post entitled "Google To Boast Larger Market Capitalization than Microsoft in 2010" (wow! talk about blogs and cross-pollination!) stating in a very prophetic tone that at the actual rate, Google will have a larger market cap than Microsoft by 2010.
"Could be a peak into a crystal ball", Fred Wilson
The stock graph accompanying the article compares MSFT and GOO. Not too revealing for now...

My theory of evolution

Several months ago (Feb. 2005) I blogged about (in French) my own theory of evolution from IBM hardware supremacy up to Google information-age domination. The graph bellow emphasizes what I meant at that time:

  • 1960-80, The Hardware era: In the 60s, IBM ruled the computing world with their hardware. Naturally, they also provided the operating system and most applications.
  • 1980-94, The OS era: The 80s clearly brought Microsoft to the forefront. They didn't bother about the hardware (I remember running MS DOS and Microsoft Flight Simulator v1.0 to "certify" the machine was truly an IBM PC compatible clone). Quickly, Microsoft took the realm of the applications running on their OS, and in mid 90s the infamous "The Internet Tidal Wave" memo turned over Microsoft attention toward the Internet. See the two part article from BusinessWeek on the topic: "Inside Microsoft" part1 and part2. Microsoft was challenged by newcomers such as Netscape. Clearly, OS is not important anymore: applications are.
  • 1994-2001, The Big Extinction: Then eBay, Yahoo! and Amazon grew up to Internet mega-star levels. This period was also the era that ended abruptly in 2001: the Internet bubble. Extremes, chaos, broken dreams... but also lots of innovation. Many didn't survive, a some passed beyond the extinction point and fewer stand-up today. Now what? We care less about applications, more about connectivity (it's all about the hyperlink model) and access to information and services.
  • 1998- ,The Information Age: Google's mission "to organize the world's information and make it universally accessible and useful" is becoming a reality. First, a web search engine, than news, images, books, maps, database, payment, and recently, videos from the rising star YouTube and while being at it, why not bring some softwares to make it easier to control the information: Blogger, Picasa, Writely (now Google Docs) and Spreadsheet, and a lot more. Do we care about hardware, OS's and applications? Much less than before!
  • 2006- (still to be named): I can't go without saying something about Web 2.0, not as much about the technology, but about the philosophy of Web 2.0. So far there is no clear dominant player. Some would say it's starting to look like an echo wave of a tsunami. I would argue we are more mature, more serious about it and generally more cautious about the potential outcomes. There seems to be actually no player who could dislodge Google. Nevertheless, the concepts of Web 2.0 serves as a strong enabler of new innovations:
    • Network as a platform: allowing mashups
    • User empowerment: you own your data, anywhere, anytime, anyway
    • Architecture of participation and democracy: I'm a human being, not a cookie.
    • Richer user experience: if I can use it online without hassle, why install a software?
    • Social network aspects: "the power of many"
    Could the next wave be the ultimate democratization of information and software? We can think about Wikipedia and FLOSS. Those would be long dead based on past financial criteria, yet they are gaining enough momentum to exist as new economical model examples; they are largely based on the concept of "attention economy".

Can Google succeed as an innovative company?

When a company raises to the point of becoming a religious icon, there is always room for heated debates. Proponents of both all horizons refer to "stolen" innovations and "takeovers". Although they didn't invent the remedy to heal cancer, both companies often innovates in their own ways by repackaging existing technologies into new, easier and better products. In this respect, Google seems to have a cultural advantage over Microsoft.

Company maturity and volume

Microsoft is much more "mature" than Google. So are IBM, Oracle and Apple. Remember at one point IBM had cash flow, market cap, sales force, consulting services, everything Microsoft have today, yet, depending on the financial indicators, IBM has been shadowed by all the companies mentioned above.

The maths

If you speak fluent "finance", read carefully the original article from HipMojo and you will understand the mathematical aspect of revenues, profits and P/E. That's one interesting point of view and a very well articulated post!

My own conclusion

When I read MarketingPilgrim's note I immediately thought about my two top-of-the-list book: "The Tipping Point" from Malcolm Gladwell and "The Attention Economy" from Dr. Thom Davenport . In my opinion, the "tipping point" passed somewhere around the second quarter of 2005 when the sheer volume of "attention" shifted from Microsoft to Google, as shown in the Google Trends below:

The example above highlights Google Trends view, which is based on the search frequency of Microsoft and Google. Other tools can be used to get similar insights: BlogPulse by Nielsen looks at blog buzz for up to 6 months. Still, the trend is there: check out Microsoft, Google and Apple.

In my mind, there is no doubt Google will economically surpass Microsoft. And I'm not alone, as my running poll shows:

No doubt 43%285
Probably 25%165
Maybe 19%125
No way 14%91
666 votes total

Do you think Google will surpass Microsoft?
No doubt
No way

Free polls from

Tuesday, October 24, 2006

Google Co-op Custom Search Engine

Google innovates again with the possibility of creating your own specialized Google search, and this can be done cooperatively.

My first shot at it, of course, is to create a Web Analytics specialized search engine where all solution and service providers would be presented, along with specialists and practitioners in the field.

FeaturesGoogle Co-Op

Configuration is a snap and provides good parametrization of the various features:
  • definition and subject of your specialized search,
  • indexed sites which can be exclusive or given preference in the search results,
  • labeling the results, which ease refinement of search results,
  • tweaking the look & feel,
  • including a search box in your site,
  • managing the collaborative aspects (contributors & volunteers) and doing invitations,
  • setting up adSense
  • and uploading annotations easily built with Google Marker (trough OPML or XML format).


  • Current number of sites in the Web Analytics specialized search: October 24th, 42
    November 2nd, 80
  • Refinements: Blogs, Solution, Services, Reference
  • Collaborators: 1
    Important! If you want to collaborate to enhance and extend this specialized search, send me a quick email (shamel67(at) or create a profile (provide your URL) so I know who you are. Otherwise your collaboration request will be rejected!


Be warned that the Google Co-Op Search is still in beta... and it shows! After a few days of using it, here are some stuff to be careful with:
  • Collaboration is far from being complete. When someone asks to collaborate, you can't even see their email address. It is also possible to ask for contribution without a profile, which leads to some confusion about who is contributing...
  • When adding new sites, be careful to use only the most meaningful URL. That is, avoid putting "http" (useless) and specify "www" only if it makes a difference, such as "" vs ""
  • Surprisingly, there is actually very little stats you can get to know if your search engine is being used or not. I would expect some Google Analytics hooked up to it.
  • Of course, the concept only works if people actually use it and suggest sites to be indexed. I think the code to create a search box should be widely available to anyone who wants to use it on their site, not just the owner or contributors. Here's the code
<form id="searchbox_017598970026374442327:f6fozzzu_qk" action="">
<input type="hidden" name="cx" value="017598970026374442327:f6fozzzu_qk">
<input name="q" type="text" size="40">
<input type="submit" name="sa" value="Search">
<input type="hidden" name="cof" value="FORID:1">
<script type="text/javascript" src=""></script>
<!-- Google CSE Search Box Ends -->


Monday, October 23, 2006

Web 2.0: buzzword?

M. Mohamed Kahlain, un collègue de travail et stratège chez, participera bientôt à un déjeuner-causerie qui traitera du Web 2.0.


Certains experts ne jurent que par le Web 2.0 qui, par sa nature, influencerait les modèles d'affaires. Certains autres mettent en doute ce concept et attendent toujours de voir les retombées d'affaires de ce « buzzword ».
Deux stratégistes Web s'affronteront dans un débat afin de savoir si oui ou non le Web 2.0 est une avenue prometteuse pour les retombées commerciales de nos entreprises. Venez donc entendre Michel Leblanc, associé principal chez Analyweb, blogueur réputé sur les sujets du web 2.0 et du marketing 2.0 et Mohamed Kahlain, stratège internet de la Fédération des Caisses Desjardins et scéptique convaincu de l'inutilité du Web 2.0 dans un contexte d'affaires.
Qui a raison ? Qui a tort ? La réponse pourrait vous ouvrir des perspectives nouvelles!
Le débat sera animé par Gilles Lajoie, La Presse

Date : 01 Novembre 2006
Heure : 8h à 9h30
Type : Petit déjeuner
Conférencier : Michel Leblanc, associé principal, Analyweb et Mohamed Kahlain, conseiller vente et stratégies pour, Fédération des Caisses Desjardins
Lieu : Le Saint Sulpice Hôtel Montréal
Salle Le Moyne/le Ber
414, rue Saint-Sulpice, Montréal
Inscription :

Mon opinion:

J'aurais bien aimé participer à cette rencontre, mais mon horaire et la distance ne me le permettent pas. Comment pallier à cette contrainte de temps et d'espace? En parler sur mon blog, dans un premier temps, et qui sait, peut-être que quelqu'un pourra l'enregistrer et la publier sur YouTube? Peut-être même que l'expression de mon opinion résultera en une discussion et au plus grand partage de nos opinions respectives, et ce, pour le bénéfice de tous ceux qui s'y intéressent? Espace, temps, opinion, communication... voilà l'essence des principes du Web 2.0!
Prenons un premier exemple: YouTube est sans contredit un modèle d'entreprise du type Web2.0, voici d'ailleurs la définition qu'en donne Wikipedia (un autre modèle concret et utile de Web 2.0!):
YouTube is a popular free video sharing web site which lets users upload, view, and share video clips.

Je crois qu'il y a une distinction à faire entre certains concepts qui sont associés au Web 2.0 (ou Enterprise 2.0!?) et l'espèce de vague de folie qui rappelle l'ère de la bulle Internet.
Il y a des éléments indéniables qui ont une utilité dans nos entreprises et dans nos vies. Puisque la définition exacte de ce qu'est ou n'est pas le Web 2.0 est très suggestive, voici les caractéristiques du Web 2.0 retenu sur Wikipedia:

  • "Network as a platform": les concepts de "mashups" sont une concrétisation de l'impartition et de la spécialisation des services. Ceci permet en retour de se concentrer sur notre expertise d'entreprise ("core business"). Mon message précédent au sujet de MotionBased est une concrétisation d'une innovation qui n'aurait pas été possible sans le Web 2.0.
  • Contrôlé par les usagers: nous sommes en contrôle de ce que nous désirons "consommer" comme service et information: quand nous voulons, où nous le voulons, et sur le média que nous voulons, et plus encore, avec une diversité de sources et d'opinions à notre disposition. Comparez simplement la page personalisée de Google à ce qu'offre Canoë sur son portail: un offre la flexibilité du design et des sources d'information alors que l'autre...
  • Plateforme participative et démocratique: les technologies du Web 2.0 permettent de s'exprimer plus facilement, pour le meilleur et pour le pire, avec tout ce qu'il y a de positif et de plus noir chez l'Homme.
  • Interfaces riches et conviviales: il a été démontré depuis longtemps que la richesse et la qualité de l'expérience usager contribue à l'atteinte des objectifs d'affaires. Mais pour ma part, j'aime plutôt me placer dans la perspective "utilisateur": la qualité et la richesse de l'interface facilitent l'atteinte de mes objectifs.
  • Réseaux sociaux: le pouvoir marketing est entre les mains des consommateurs, qui peuvent influencer et exercer leur opinion librement et facilement. Bien utilisés, les concepts de Web 2.0 offrent des possibilités incroyables, comme le démontre la récente campagne de Dove, qui a fait rapidement le tour du monde, encouragé et alimenté par une communauté qui y a vu une valeur (le simple fait que je mentionne cette campagne sur ce blog est une concrétisation de ce concept de "réseautage social"!)


Je crois que tout comme les concepts de "ebusiness", qui ont d'abord été perçus comme des recettes miracles et une panacée, le Web 2.0 évoluera au niveau d'une intégration normale et toute simple au coeur de nos vies et de nos entreprises. Le Web 2.0 est une évolution logique (et non une révolution!) qui s'exprime actuellement à travers l'expérimentation, l'innovation, une part de chaos et d'improvisation, mais qui contribuera ultimement à l'évolution des modèles d'affaires et influencera nos vies. Je ne jure pas que par le Web 2.0, mais je ne peux mettre en doute que le Web 2.0 exercera, et exerce déjà, une grande influence sur nos vies.

Friday, October 20, 2006

Upcoming webinar on analytics by Prof. Davenport

The Attention EconomyI just got an email from Aquent on behalf of the American Marketing Association inviting me to register for an upcoming webinar talking about analytics. This webinar will be presented by Prof. Thomas Davenport, a renowned authority in the field and author of one of my favorite books: The Attention Economy.

Prof. Davenport wrote a paper entitled "Competing on analytics" and will soon publish a book on the same topic.

Here's the excerpt from the mail invitation:


  • Aquent and the American Marketing Association present a free webinar with analytics expert, Professor Tom Davenport: Competing on Analytics: Move Faster, Accomplish More, and Avoid Mistakes by Learning From The Best
  • Davenport, an international authority on business and marketing analytics, is the author of "Competing on Analytics", a critically-acclaimed Harvard Business Review article. While researching the article and a soon-to-be-published book of the same name, Davenport profiled over 25 early adopter organizations that now successfully compete on the basis of their analytic prowess. Beyond the broad strategies used by pioneers like Capital One, Procter & Gamble, Amazon, and the New England Patriots, he detailed the often painful tactical mistakes and missteps made along the way by these analytic leaders. In this webinar, Davenport will discuss what data-driven marketing is (and isn't), the broad strategies and specific tactics these early adopters believe were essential to their success (and what they'd do differently next time), and how marketers can be professionally successful in these tumultuous times. The webinar will be of interest to all business-to-consumer and business-to-business marketers, especially those challenged with making analytics work on a practical basis within their organizations.


  • October 31, 2006 @ 1:00 pm EST (45 minute presentation and 15 minute Q&A)


Professor Tom Davenport of Boston's Babson College
  • Author of critically-acclaimed Harvard Business Review article "Competing on Analytics" (being released as a book in Spring 2007)
  • International authority on analytics as a business and marketing strategy
  • Prominent consultant, speaker, and pundit

What participants will learn:

  • What data-driven marketing is (and isn't)
  • How marketing visionaries like Capital One, P&G, Amazon, and the New England Patriots are using analytics for competitive advantage
  • What specific tactics these early adopters believe are essential to their success (and what they'd do differently next time)
  • How you can personally succeed as a marketer during these tumultuous times

Who should attend:

  • Business-to-business and business-to-consumer marketers at all levels, up to and including CMO's, Vice Presidents of Marketing
  • Information technology professionals, particularly those in fields involving collaboration with marketing colleagues

How to register:

Thursday, October 19, 2006

Competitive analysis and monitoring

I believe there are strong links between web analytics and performance metrics. For example, if you experience a drop in conversion, how do you correlate that with the fact your infrastructure had some problems (performance, network, etc.)?

Another perspective is to be able to do competitive analysis: monitor competitors transactions and detect changes to a single page or a transction.

What I've been looking at recently are solutions such as Gomez, Keynote, Alertsite, Mercury End User Management and a bunch of other lower-end solutions (such as Site24x7 by Zoho).

However, it turns out that most of the tools are made for IT people who want to monitor and establish performance baselines (response time, SSL handshake, DNS resolution, etc.) and get alerted whenever something goes wrong. I have to admit the usability of those solutions is often awkward and not too intuitive.

I will continue to investigate this topic and post updates to my blog. Pending questions are:
  • what business tools (vs IT solutions) could be used for external end-to-end performance monitoring?
  • what tools could be used for competitive monitoring?
  • how to correlate infrastructure performance metrics with web analytics?

Tuesday, October 17, 2006

Web Analytics Wednesday (Montreal, Qc)

Suite au succès de la première rencontre de la communauté Analytique Web du Québec, nous répéterons l'expérience une seconde fois le 8 novembre 2006, de 18h à 21h au restaurant "Le Cartet",106 rue McGill, à Montréal (Google Map).

L'idée est d'instaurer la tradition de rencontres conviviales favorisant les discussions et les échanges au sujet de tout ce qui touche les statistiques Web, l'analyse du comportement des visiteurs, l'ergonomie, le design Web et l'analytique en général. Professionnels, utilisateurs, représentants de solutions ou de services se regroupent afin d'échanger et d'établir de nouveaux liens.

L'évènement est appuyé par la Web Analytics Association.

Si vous connaissez d'autres personnes susceptibles d'apprécier ces rencontres, vous êtes encouragés à leur en faire part!

P.S.: Nous cherchons un commanditaire pour prendre la première tournée.

CrazyEgg: heat maps for the mass

Following a post about CrazyEgg on LunaMetrics blog, I decided to subscribe and give it a try on my own blog. I'm already using Google Analytics, which was a snap to install and provides good insight. Although I find GA to be missing some important features and reporting capabilities when compared to the big 3 (Omniture, WebSideStory and Coremetrics), it's enough for a personal site and furthermore, it's free.

So back to CrazyEgg: subscription is free for small sites, then you define your goal (how long or how many visits?) and installation is a mater of copying & pasting a couple of lines of JavaScript in the right spot of your Blogger template. Lets look at each CrazyEgg reports after a couple of hours of tracking.


Provides a list of the clicked objects (text links or images) and number of clicks, the most basic information. While GA will tell you about the destination page name (the TITLE tag), CrazyEgg uses the link text itself, which can provide useful insight if your page have multiple links to the same destination, both uses different wordings.

Click Map

Neat color-coded overlay of link clicks. Already more revealing.

Heat Map

Probably the coolest feature: contrary to simple click map overlays, a heat map tells you exactly where people clicked. That's it, for my link "WaW in Montreal: 225% increase", we see the "heat" around the "225% increase". Another example is my post about "Web Analytics Wednesday - Québec", where it's clearly the word "Quebec" that was clicked. There was also some clear interest for my LinkedIn profile.

Simplified A/B testing

Just like CrazyEgg will tell you which links are performing better on a single page, it will also tell you which of two page design works better. This is really a simplified A/B testing technique. Again, the biggest merit of CrazyEgg is to bring this technique to the mass.


Blogs are often composed of other scriptlets or gadgets. In most cases, they will take the form of a SCRIPT reference or an IFRAME. IFRAME are really pages hosted osomeone else'sss web site and thus, what happen in there (clicks and any other interactions) is really out of your control and obviously, not measured by Google Analytics, CrazyEgg or any other tools. Script embedding techniques offers more control and will generally work with analytics solutions.


CrazyEgg is a simple and useful addition as a complement to any analytics solution, maybe especially bloggers using Google AnalyticsAltoughht other heat map solutions are typically expensive and a bit more complex to use, they are usually part of a planned and structured analysis process which includes analysis and skilled personnel to help you out. I would say that CrazyEgg, just like Google Analytics, increases awareness of the field of usability, design and analytics. Companies should still investigate to find the best fit for their needs.

Thursday, October 12, 2006

WaW Montreal: 225% increase!

Yesterday's WaW in Montreal was a success!

Attendance increased 225% from last month (from 4 to 9) with a decent
attendance conversion rate of 65% (9 out of 14). In terms of "sales",
we should expect two "converts" who are not already part of the WAA. (working with small numbers, it's easy to show impressive growth! that's why context is always so important!)

Attendees were from various horizons:
as well as users of WA from:
Topics discussed went from pure web analytics geek talks up to
performance analytics, usability, jobs and even departed at some point
to include quality of life in Montreal and the town's reputation to
have some of the best looking womens in the world.

To quote my friend Jacques Warren: "intelligent people, intelligent
conversations, no pitch, for once..."

We already look forward for the next get together! Some action items
for next month's:
  • reach out to more people
  • create a LinkedIn group for WA practionners in Quebec
  • publish a list of solution/service offering in our market
And who knows... maybe we could eventually host an E-Metrics summit in
the beautiful town of Montreal?!

Tuesday, October 10, 2006

GPS+Google Maps+Community=MotionBased

A colleague, fan of mountain biking, found a good way to interest me to his week-end expeditions:

Simply put: use a decent GPS, record your trip, upload your data and share it with the community. Bonus includes minute by minute tracking on Google Maps, along with elevation, speed, and even heart beat. Works from simple walks around the block to long lasting road biking, or, why not, even canoe, skiing, or a bunch of other stuff... someone could even think of street racing... Knowing you can take someone else's trip and try to beat it, this will certainly happen eventually!

While some argue the speed of innovation is slowing down, what we see in this Web 2.0 wave is often new and innovative ways of using existing technologies. I think MotionBased is a very good example of several Web 2.0 concepts: community focus, innovative use of technology, Web as a platform, innovative design, etc.

Check out their demo:

Monday, October 9, 2006

Mavericks, MBA and leadership

As an ebusiness MBA student, one of the course was "understanding the digital enterprise". An interesting topic that was raised is "what makes a good business leader"? One could think most answers would tend toward studying and applying the best business practices learned in management schools. However, what came out of this group discussion is very close to what Taylor and LaBarre explain in their new book "Mavericks at work". Guy Kawasaki posted an interview with the authors on his blog. Also, Taylor and LaBarre submitted a very nice manifesto on ChangeThis; a must read.

"Why the most original minds in business win?" Here's some interesting quotes from the manifesto:
  1. Being different makes all the difference: "Winning companies don't just sell competitive products. They stand for important ideas, ideas that shape the future of their industry, ideas that reshape the sense of what'’s possible among customers, employees and investors."
  2. Sharing your values beats selling value. "When just about everything keeps getting cheaper and better, offering customers something that's a little cheaper and better today won'’t win them over because something else is bound to be cheaper and better tomorrow." This point and the one above reminds me of Thomas Malone in "The Future of Work", when he says "Making choices that are wise, not just economically efficient, means making choices that are consistent with our deepest values."
  3. Nobody is as smart as everybody. Behind every maverick company is a set of distinctive and disruptive ideas... the new world of innovation is built on the principle that "“nobody is as smart as everybody."”
  4. The people are the company. Simply put "they are the building blocks of long-term prosperity."
In 20 years of career, I worked for numerous companies, small and big, some were acquired, some failed, some had good while others... had far less interesting managers. I haven't read "Mavericks at work" yet, but I'm sure I will be able to relate to some great people I have worked with. You know, those who have that kind of charisma, those who can rally people around them and motivate a team with a great vision. Those, when asked about the secret of their success, who will instantly answer "it's a team work". Keeping that in mind, and trying to apply those principles, I hope, helps me being a better employee and a better person.

Wednesday, October 4, 2006

Buyer intention and attention

A recent study by HFI talks about personalization troughout the purchase cycle. They go trough the typical purchasing process and sugest cautious use of personalization acording to each step:
  • Universal set: all possible product alternatives that could be purchased by the consumer
  • Awareness set: a subset of all possible alternatives, limited by consumer interest and cognitive factors (attention, comprehension, cognitive capacity)
  • Consideration set: small set of potential purchase alternatives, that the consumer carefully analyzes and compares
  • Choice outcome: the best alternative that the consumer arrives at, after eliminating the other alternatives from the Consideration set.
One incredible outcome of this study is the impact negative tone have on the purchasing behavior (for example "this item just for you" vs "this is the last time you will see this offer"). This came out as a strong positive factor. Could it be a culture difference specific to the Hong Kong market where the study was done?

Another aspect which I studied myself about two years ago is the impact of "attention" (see "The Attention Economy" by Thomas Davenport) on the purchasing cycle. One interesting thing to study is the type of attention awarded depending on steps of the purchasing cycle. What we see is an initial, shorter and voluntary attention span progressively going toward a very focused and narow attention, often revealed after a couple of visits, just before the final purchasing decision. Think about how you shoped for your last car and you will easily see an evolution in the way your "attention" was used. When combined to the classic RFM marketing metrics (Recency, Frequency, Monetary), Attention, what I would call the 4th dimension of RFM (now RFMA!) really becomes a new decision factor influencing personalization.

The challenge is now to be able to measure "attention" using web analytics techniques!

Monday, October 2, 2006

Google Reader and KPI

I went back to Google Reader today and noticed some changes. While parsing my new RSS aggregated feed about web analytics, I stumbled on an "Instant Cognition" post wondering how Google Reader decides if a post have been read or not.

One of the first thing I did in the new Google Reader incarnation is to go under Settings/Preferences and uncheck "Scroll tracking", which is said to behave like this: "In expanded view, mark items as read when you scroll past them." However, experimenting with this features tell me it might be a good idea after all.

How to determine if a post was read?

Even if a post is in the current focus, nothing tells us the person is actually reading it. However, in expanded view, we should notice the framed color around an individual post changes as we scroll. A nice touch from Google to remind us of the post currently being in focus. I think the idea of "onFocus" + time could have some meaning, but "time" should be ajusted based on the post length.

Which KPI to use in this case?

  • Ratio of focus time to post length (longer post/longer focus, shorter post/shorter focus, but ratio tells us how much attention was awarded to an indiviual post)
  • Feed depth (out of X new posts, how many were read?)
  • In this case, could the notion of "page view" be replaced by "post view"?
This always remind me of "The Attention Economy" by Davenport. We used to try to get attention to our sites, now we try to get attention to individual posts!

Online marketing effectiveness

A recent study from the CEFRIO sheds some light on our behaviour toward different types of online marketing activities. Some revealing points of this study are:
  • 93% of internet users ignore online banners and publicity in general;
  • 80% find it invasive, especially when used in a popup or overlay;
  • In terms of message type, 25% tolerates publicity in search engines, 21% find text links ok, 18% find banners to be fine, while only about 6% likes overlays.
  • However, traditional marketing can drive online trafic: 60% of respondants will sometime refer to links presented in magazines and 50% will sometime look at links presented on TV.
The study also reveals its own top 10 golden rules of online marketing:
  1. Do not take the user ostage - always offer a way to close or bypass it;
  2. Do not interfere - overlays or animations are to be used cautiously;
  3. Be respectfull - leave the choice up to the user;
  4. Be contextual - place your publicity on a site that have some contextual meaning to your offer;
  5. Be unintrusive - place publicity at the top or on the side;
  6. Publicity should be displayed transparently within the page flow;
  7. Use the medium - be different from other media channels;
  8. Be informative and subtle - never be coarse;
  9. Be funny or amusing - a pub that makes you smile can never be negative;
  10. Consent - for email marketing.