Webanalytics general

Web 2.0 Expo San Francisco

April 22 - 25 are the official dates for Web 2.0 Expo in Francisco. I'll be attending and will be around all week. Besides  meeting with one business partner and a quick Google office visit, not much is planned out, even that there are a ton of evening events. So if anybody is around, shoot me an email and we'll probably can meet up.  Also if you don't have a ticket, and just want to visit the Expo, use this coupon websf08geh and get $100 off.

59% Of Online Users Are Not Comfortable With Web Analytics?

According to a Harris Interactive study... "a majority of U.S. adults are skeptical about the practice of websites using information about a person’s online activity to customize website content. However, after being introduced to four potential recommendations for improving websites privacy and security polices, U.S. adults become somewhat more comfortable with the websites use of personal information.

  • A six in ten majority (59%) are not comfortable when websites like Google, Yahoo! and Microsoft (MSN) use information about a person’s online activity to tailor advertisements or content based on a person’s hobbies or interests. A quarter (25%) is not at all comfortable and 34 percent are not very comfortable;
  • The remaining 41 percent who say that are comfortable with websites tailoring content is split between 7 percent who are very comfortable and 34 percent who are somewhat comfortable."

So should website's be open about their privacy settings?


After four privacy/security policies were introduced, U.S. adults did change their opinions:

  • By 55 to 45 percent, a majority of U.S. adults indicates that they would be more comfortable with companies using information about a person’s online activities to provide customized advertising or content;
  • Interestingly, once the privacy/security policies were presented the percentages of those who are very comfortable increases only very slightly to 9 percent from 7 percent. The percentage who are somewhat comfortable given the privacy/security policies increases more significantly to 46 percent from 34 percent;
  • Similarly, those who are not at all comfortable decline to 19 percent from 25 percent, and those who are not very comfortable decline to 26 percent from 34 percent."

Extremely interesting study.

Proactive Chat. Good Or bad?

Since Woopra's Chat system made a lot of noise a few days ago, today news from Omniture. InstantService and Omniture joined forces and enable a Proactive Chat to Omniture and InstantService clients. This is how it work: "InstantService InstantInvite employs a rules-based engine (instead of human monitoring) to identify shoppers with greatest potential to buy. This unique and innovative approach presents important value at the front and back-end, ensuring privacy for shoppers and increasing agent productivity. The solution also delivers informative data related to chat transactions which enables companies to better assess online customer behaviors, and make adjustments in real time without changes to site code.

Rules for triggering a proactive chat session are based on visitor site interactions and behaviors and commonly include: which pages they have visited, how long they remained on a page, where they came from, which keyword they used to find the website, which items they have in their shopping cart, and whether they are repeat buyers.

Based on these user-defined rules, InstantInvite automatically presents shoppers with a customized invitation to chat live with a customer service representative. With this automatic connection, agents can instantly engage the customer with offers for assistance and incentives in addition to up-sell and cross-sell opportunities."

Also according to InstantService a study showed that...

"-- 20% of web chats result in a completed purchase -- Web chatters spend approximately 35% more per order -- 30%-40% of Web site traffic includes self-service shoppers (consumers who make purchase decisions without vendor assistance) "

I was not really convinced by  Woopra's  chat system, b/c it might be one of these things that can go too far and turn users rather off than driving them into a sale.

Omniture and InstantService however might be a solution that is more based on metrics and I believe it could work for certain kind of sites (e.g. sites that sell customized products that need some consultation...).

Anyways...any online marketer should test a proactive chat and see if it works for them. Reliable studies will probably show up not within the next few months, b/c every website is just too different.

The shift to paid search

A few days ago classified website Kijiji, a subsidiary of Ebay, announced that they received 2.3 million U.S. visitors in January and became the 6th largest classified website in the US. Within 6 months Kijiji became about 10% of the size of Craigslist ( 26.7 million unique visitors in the U.S). For me these numbers are not very surprising. Kijiji did an incredible job outside the US and their paid search team is one if not the best in the world.

On top of that Google and Co. try to squeeze more and more money out of search and treat their (large) paid advertisers better than the rest of us as good as possible.

Especially when the advertisers sites include contextual advertising like Adsense, it's known that the advertiser  can receive larger payouts than the general Adsense partner (I lost the source, but Ebay's Gumtree got a special deal with Adsense, so I assume that Kijiji does as well).

Overall the results are not really a surprise, but what's way more interesting is that most of Kijiji's US traffic is not organic (at this point). It's a matter of time (and money) until Kijiji beats competitor Craigslist. Even if most people sitting at the peanut gallery predict that it won't happen.

Nowadays long-tail bidding for large advertisers can be done in such an effective way, that organic traffic becomes obsolete. Especially for sites, that live of niche long-tail keywords like Craigslist.

I would not call Kijijis' PPC bidding arbitrage, but the shift to paid search in combination with contextual advertising is a sign that the industry "matured". The time when a lot of small sites made it big through organic traffic is certainly over.

Microsoft doesn't like the way we measure Clicks

So not only a bunch of startups, but also Microsoft criticizes the way we measure advertising: "Microsoft Corp. today announced Engagement Mapping, a new approach to managing and measuring the effectiveness of online campaigns that goes beyond the current “last ad clicked” standard. For the last decade, virtually all ad campaign reporting methodologies associate sales, leads and Web traffic simply to the last click or ad exposure. Engagement Mapping takes into account for the first time all the various online touchpoints and interactions a consumer experiences before an eventual sale."

Ohhhh goodness. I guess the chaos is almost perfect and I am not astonished that all these companies (incl. Microsoft) question the way we measure advertising. There is just too much money in the business and everybody wants his share of Google's the cake.

But don't we remember the CTR issues for years? Coremetrics for example was is able to measure not only the last action that triggers a lead/subscription/sale, but also can go further back to previous actions.

Anyways, Microsofts statements certainly support new web analytics companies like NuConomy, who now might have a good shot at gaining some traction.

How do you qualify for a web analyst position?

Anil Batra asked a few web analysts about skills, that they think are required for a Web Analyst. Here my 2 Cents: A web analyst needs to have advanced statistics knowledge. Period.

A web analyst needs to have a background in SEM, SEO, Usability and overall Web development(programming languages, wire frames, Web 2.0, data mining, databases...).

Additionally advanced Excel knowledge and presentation skills are key to succeed in this profession.

To summarize this: A Web Analyst should be a trained Analyst, Search Engine Marketer ,Search Engine Optimizer, Social Media Optimizer, Online Marketer, Usability Expert, Digital Artist, Web Developer, Communication Expert, Excel Pro in one person.

Click-Through Metrics Is Dead As Well...

I really start loving the web analytics industry more and more. Every other week a new well-funded analytics startup, every month the death of another metrics. Today Comscore announced Heavy Clickers Distort Reality of Diplay(!) Advertising Click-Through Metrics.

"The study illustrates that heavy clickers represent just 6% of the online population yet account for 50% of all display ad clicks. While many online media companies use click-through rate as an ad negotiation currency, the study shows that heavy clickers are not representative of the general public. In fact, heavy clickers skew towards Internet users between the ages of 25-44 and households with an income under $40,000. Heavy clickers behave very differently online than the typical Internet user, and while they spend four times more time online than non-clickers, their spending does not proportionately reflect this very heavy Internet usage".

I guess I should have added CTR for Display Advertising to my last dead metrics post Revolutinizing Web Analytics post.

KPI's From Iceland

Iceland (=Lýðveldið Ísland= Island in German) still exists and is even the place of a popular Internet Marketing Expo called RIMC. I went to Iceland a while ago, but didn't participate at the RIMC. However Dennis Mortensen from Indextools gave a presentation called "The difference between a KPI and a metric". Enjoy: [slideshare id=253633&doc=the-difference-between-a-kpi-and-a-metric-1202240162965806-3&w=425]

Web Analytics Wednesday

Quick note for everybody that works in the Web Analytics field: Visit the Web Analytics Wednesday meetups. On Wednesday Thursday evening it was my 2nd WAW meetup in NYC and Eric Peterson, aka House author of Webanalyticsdemystified , attended and even took the time to talk to everybody.

Having never met him before, I was surprised how much fun he is and I only can imagine how valuable it would be to get him as a consult in your company (Yes, you can hire him here).

But not only Eric, also pretty much all the other web analysts are really laid back and the usual SEO BS talk, that you often hear at conferences, was completely absent. Probably metrics driven people are just a little more mature.

Overall it was a really great event and besides having free drinks food, I really enjoyed networking with the brightest minds in the industry. Small meetups like this are so much better than attending big conferences. Especially when you take insights home, that you don't get anywhere else.

Thanks again to everybody who came up and chatted with me (Mark, Joel, Marshall....). Also thanks to the sponsors Optimost Opinionlab and Webanalytics demystified Inc..

It Just Doesn't Add Up

Just found a great example in Richard Rothstein's book "The way we were" about how we a lot of people misinterpret data: " [A] Siemens manager noted that American youths (at plants in Florida, North Carolina, and Kentucky) scored higher on his company's apprenticeship exams than comparable German young people who followed the same curriculum and took the same test...."

"These employment-test anecdotes are suspect, among other reasons, because they are not representative or properly controlled samples. It is not known, for example, from the Siemens manager's speech whether Siemens's pay scales in the United States are relatively high while in Germany they are relatively low: if this were the case, it might be that Siemens's apprenticeship programs in America enticed the cream of our non-college-bound high-school graduates, while similar programs in Germany attracted the less able. "

Lessons learned:

1. Get your hands with the data dirty before you come up with assumptions that don't make sense.

2. Always use a sample size calculator

[3. American schools are not as bad as newspapers want to tell us]

Eric Peterson in NYC

Just for everybody who wants to meet web analytics god  Eric Peterson and grab free drinks  should read this: Eric Peterson, Web Analytics Guru, author of Web Analytics Demystified, The Big Book of KPIs, Web Site Measurement Hacks, moderator of the yahoo web analytics user forum and CEO of Web Analytics Demystified Inc.,will be at the Heartland Brewery (6:00 to 8:30)1285 6th Ave at 51st Street tom..

Impressive collection of titles...wished I would be that smart :-)

Revolutionizing web analytics

When page loads became irrelevant last year (due to Ajax and Co.), it was time to come up with something new. The Audience 2.0 system, that the guy(s) from Alenty.com developed is pretty cool and could revolutinize the way we currently measure things. Alenty's audience measurement tool allows to measure time, content visibility and even better "banner visibility". Real-time!

So what's so cool about that?

At this point most advertisers sell banners on a CPM or PPC basis. While PPC gives some great indication about ROI, a CPM deal is often extremely difficult to measure and often not very interesting to small advertisers.

Alenty's new measurement tool (they should come up with a cool name) can change the way that banners are sold and measured:

Banners could be measured by Visibility (visible on computer screen), Time (only when mouse or keyword activity is recorded) and Surface (what's the percentage of banner surface, that is visible on user's screen).

Banner 468x60:


Banner Skyscraper:


Audience measurement:


If this new technology becomes adapted by one or more big vendors, we might see a change in the online advertising world.

The Facebook web

Usually I don't post Facebook or Myspace news unless they are related to Analytics. But what Facebook recently announced is just once more pretty smart: Facebook released a Javascript Client library that enables developers to extend their application to their own website.

"Rather than building your applications strictly within Facebook you can now extend the full functionality of the platform to your own website and leverage Facebook as the tool for managing members and their relationships." Source

This means that a lot of companies now will build a Facebook app and let it run on a separate domain. The companies can then drive traffic from their "real website" to their Facebook app and vice versa.

After this new release, I believe that Facebook's exposure in the web will increase drastically. Marketers will probably also love this. They now have the ability to crossmarket their Facebook widgets via search/print/radio... and via the Facebook platform.

I am not sure how far it will go (I never developed a Facebook app), but if the applications get better and better over the years, we won't be far away from a "Facebook web".

Kind of reminds me on the good old phpbb days and their open source code approach. First developers through the world developed modules/hacks for the CMS until it turned into a high-end solution, that anybody with a little PHP knowledge could customize.

Then the next wave of modules helped webmasters to offer  modules /hacks, that increased the exposure in the web.

Facebook did all that, but even opened up a little more. Great move!

Are IP addresses personal?

Germany's data protection commissioner, Peter Scharr, leads the EU group preparing a report on how well the privacy policies of Internet search engines operated by Google Inc., Yahoo Inc., Microsoft Corp. and others comply with EU privacy law. He told a European Parliament hearing on online data protection that when someone is identified by an IP, or Internet protocol, address "then it has to be regarded as personal data."

The big search engines see this probably different, but I understand that privacy is still a big issue. Probably the future will show how this stuff will effect the web analytics industry.

Google Checkout Trends

Just saw that Google has a new (?) checkout trends tool. It allows you to check what people are buying online. I am not sure if this data gives much info (not many E-commerce sites use Google checkout yet), but ....hey..it's free, so you might want to use it to get an idea of what's going to sell in e-commerce: google_checkout_trends.jpg

Multi-Channel Analytics - predictions for 2008

Joseph Carrabis  over at Allbusiness.com predicts Multi-Channel Analytics as the trend for 2008: "One thing that's come out clear and simple is that for many clients web analytics is...umm...dead. If not dead at least not the be-all and end-all that it once was.

I realized that fewer and fewer (and in the past year, exactly two) clients are coming to NextStage and asking us to help them improve their web performance....

I'm not the first one to write multi-channel is important. I think I'm the first one to write that web analytics is a subset of multi-channel analytics and multi-channel will dominate. The allure of web analytics was that it was directly measurable...kind of. I'm still not convinced that what people are measuring directly equates to what people will do (predictability is what it's all about as far as NextStage is concerned)" Entire article here.

I agree that more and more large companies are interested in "multi-channel" analytics. Still I believe that the majority of companies haven't even touched advanced web analytics and the whole industry is far from being dead. I think it is kind of the opposite: The web analytics industry hasn't reached it's peak.

Web Analytics predictions for 2008

Ok...we are already in the year 2008, but Rich has some great predictions, that I had to comment on: 1.  Jan 2008 - Core Metrics acquire Web Trends.

***** possible, but I am not convinced.

2.  Mar 2008 - Amazon.com gets into web analytics!

************ I definitely would love to see that. Amazon's Alexa is a pretty powerful tool, but pretty much undervalued. They are in the business for a while and I would love to see them coming up with some advanced features (like Compete's search analytics).

3. May 2008Disaster for Microsoft Site Analytics.

************ Let's see how that goes. I've tested Gatineau a while ago and I like the tool, b/c it is not the usual 0815 interface. Gatineau is still in private beta, but it's time to come up with something for everybody.

Overall I am not so sure how the demographics feature will really change the game. I am not sure how accurate the demographic data from hotmail is. So...I don't think it will be a disaster....I'd rather think it won't have as much of an impact as Google Analytics had, when it launched.

4. June 2008 - WebTrends continues its spending spree!

***ok...don't know too much about WebTrends.

5. August 2008 - Omniture offers free analytics solution

*********Definitely would love to see that. Omniture vs. Google would be nice matchup.

6. December 2008 - Big acquisition shock.

********** Google buying Omniture! I can imagine it, but I doubt that it will be necessary. The majority of websites will be happy to keep on using Google Analytics / Sitemeter / Gatineau....