Friday, May 30, 2008

Shopping Cart Abandonment and Conversions

Shopping cart abandonment is a huge challenge for online merchants. eTailers are constantly struggling to understand how they compare to other eTailers and the reasons of cart abandonment so that they can improve their conversions.
Yesterday I came across some interesting survey data that will help shed some light on the average abandonment rates, conversion rate and the main reasons for cart abandonment.

Industry averages

The following charts are from MarketLive Performance Index Volume 4

Q1 2008

Note: “1-and-out” is commonly referred as “Bounce Rate”. I also conducted a survey on Bounce Rates, you can find the results at

Reasons for Cart Abandonment

A study conducted by Paypal and comscore showed the following reasons for cart abandonment

Chart Source:
  • 43 percent of consumers didn’t pay for items in their shopping carts because shipping charges were too high

  • 36 percent of purchasers didn’t pay for items because they felt the total cost of the purchase was more expensive than anticipated

  • 27 percent of shoppers didn’t pay for items because they wanted to comparison shop at other Web sites before making a purchase

  • 16 percent of consumers didn’t pay for items because they could not contact customer support to answer questions

  • 14 percent of shoppers didn’t pay for items because they forgot their usernames and passwords for their store accounts created with the merchants

  • More 20% shoppers didn't complete purchases because their preferred payment option was not offered on the merchant's Web site

  • 21 percent did not complete online purchases because their wallets were not easily accessible

You might also be interested in Targeting Cart Abandonment by Email

Looking to fill your Web Analytics or Online Marketing position?
Try WebAnalytics360 Job Board

Thursday, May 29, 2008

5 Step Process to Ease Privacy Concerns Regarding Behavioral Targeting

According to an article in Reuters, Stanford Group Co., a financial services research group, concluded that Federal, state and consumer discomfort with Google Inc, Yahoo Inc and other companies tracking consumers' online behavior could slow the growth of Internet advertising.

"We think the growing government scrutiny is likely to make it easier for consumers to opt out of behavioral tracking, which in turn will reduce the number of web surfers that can be reached through behavioral advertising," the group said in a statement.
"The momentum toward disclosure/opt-out has negative implications for the rapid growth of online advertising," the group said.

I don’t completely agree with this the conclusion and statements made by Stanford Group Co. As I have written in past, any company that wants to engage in Behavioral Targeting needs to first build trust with the consumers before they start targeting them.
Perfect example is Most of the Customers do not mind being tracked by Amazon because they trust Amazon, they know that Amazon is collecting the data and that data will be used to provide better experience on the site and also to provide customers with timely and relevant offers from Amazon.

Yes, I agree that in short run providing easy opt-out options will result in lower reach but in long run publishers, advertiser and networks will build a trust that will provide more value than the value lost by sacrificing reach.
In my opinion, the momentum towards privacy will increase over time. So rather than complaining it is time that networks and publishers (and even advertisers) start taking action to build trust with their customers (visitors).

I provided a 5 step process to ease privacy concerns related to Behavioral Targeting, online advertising or On-Site in my post titled “ Consumer Awareness and Attitudes about Behavioral Targeting, here are those 5 steps.

  1. Build trust with your consumers

  2. Educate them what Behavioral Targeting is and how you collect the data and use it

  3. Provide them a compelling reason to allow you to collect their data

  4. Build an opt-in model allowing users to control what data they want you to use

  5. Give users a way to easily opt-out of Behavioral Targeting

Do you agree? Disagree? Question? Comments? Chime in.

Looking to fill your Web Analytics or Online Marketing position?
Try WebAnalytics360 Job Board

Tuesday, May 27, 2008

Pay-Per-Click Behavioral Targeting

Last year I wrote extensively about major online players Google, Yahoo, Microsoft and AOL using Behavioral Targeting. Microsoft and Yahoo have been very open about it but Google has not admitted openly about engaging in Behavioral Targeting.

I speculated about Google entering Behavioral Targeting, you can read my past blog posts at Recently, Google Analytics Data Sharing option reconfirmed my belief about Google entering Behavioral Targeting and using all the data it collects via Google Analytics (and other applications) to better target the ads.

A recent survey by (Search Engine Marketing Professional Organization) found that four out of five advertisers are willing to increase their online advertising budget in order to add behavioral targeting to their pay-per-click campaigns.
The survey also found that
  • 57% of online advertisers were willing to spend more on demographic targeting, such as age and gender

  • Advertisers, on average, would pay 11% more for both behavioral and demographic targeting

  • Some 40% of the respondents said they are not currently targeting or retargeting searchers but they plan to in the next 12 months

Considering these stats and all the things that I have observed (see my past blog posts), I won’t be surprised if Google, Yahoo, Microsoft, AOL and ASK soon start offering behavioral targeting to it’s Paid Search (PPC) advertisers. To get the most of out of behavioral targeting, they won’t limit the behavioral data to search only, they will most likely use every data point they collect about visitors in various places including browser toolbars.


Tuesday, May 20, 2008

Charter Communications and Behavioral Targeting

Charter communications, an internet service provider (ISP), announced this week that it will share it's customers web browsing data with NebuAd, to show ads based on customer’s web browsing behavior.

(Note: I wrote about the ISP based behavioral targeting, NebuAd and privacy issues related to ISP based behavioral targeting in my previous blog posts).

Charter communication sent a letter to its subscribers in Fort Worth, Texas, San Luis Obispo, California, Oxford, Massachusetts and Newtown, Connecticut telling them it will collect their web surfing data and use to show ads related to their interests. Charter said it will start testing the system within 30 days and will make a decision whether to introduce it to its 2.8 million Internet customers a few months after that.

One big issue I see with Charter’s Behavioral Targeting is that they are automatically opting in the users. Even though they have sent the letters to customers how many their customers will actually read those letters?
As I have written in past, companies need to build trust with consumers, show them the value in behavioral targeting and then use an opt-in system. Charter is clearly taking the route that other companies have taken which, in my opion, will raise huge privacy concerns.

If anybody who doesn’t want to be tracked will need to explicitly opt-out of the system by providing their name, address etc. The opt-out uses a cookie, so if a consumer deletes a cookie or buys a new computer or uses a computer other than the one which has cookie then he/she will be automatically tracked. What I did not understand was, why do consumers need to provide their name, address etc to opt-out if it is just relying on a cookie. Can anybody from Charter please clarify this?

In march, I wrote about British Telco, coming under fire for engaging in Behavioral Targeting; let’s see what kind of repercussion we see with Charter communication.

Wednesday, May 14, 2008

5 Steps to Successful Behavioral Targeting

If you are a reader of this blog then you know what Behavioral Targeting is and how it is the latest (not really) idea that everyone is embracing. Ad network, web analytics tool vendors, agencies etc. all now claim to have behavioral targeting capabilities. Weather they really have the behavioral targeting capabilities or not is a topic of another blog topic. Some other like Google do behavioral targeting but do not talk about it. (You can read my past blog posts on behavioral targeting

For all those who want to get into Behavioral Targeting or have been involved with behavioral targeting but not seeing the results or just want to make sure they get the most out of their behavioral targeting efforts, I have put together my 5 step process for successful Behavioral Targeting. (I talked about these steps in detail in my session on Behavioral Targeting at eMetrics). These steps apply to both Behavioral Targeting via ad-networks and on-site behavioral targeting.

5 Steps to Successful Behavioral Targeting

  • Define your goals for Behavioral Targeting

  • Define your customer segments

  • Define Success metrics

  • Find the right tools/networks

  • Create a culture of continuous Improvement

Define your goals for Behavioral Targeting

This is the first and very important part of your Behavioral Targeting strategy. It is no different than any other campaign that you run, however, it is often not paid due attention to. Unless you know what you want from the Behavioral Targeting you will not be successful or in other words you will always be successful but not the same way that you should be.

While defining your goals for Behavioral Targeting, think about things like why are you even thinking about Behavioral Targeting (BT), are you looking for short term short term ROI or do you have a long term plan, are you trying to engage BT for Branding or direct response, are you just looking to drive traffic, acquire new customer, retarget the customers who have already been to your site? Having clear goals in mind will help you define your customer segments, success measures and set the right expectations among all the stakeholders. It will also keep you vendors in check. Document your goals and make sure all the stakeholders agree on those.

Define Customer Segments

Once you have the understanding of the goal of you Behavioral targeting efforts, think about what customer segment you want reach. How will you define these customer segments? What characteristics or behavior make them the right target. Some of the segments might be very obvious if you have thoroughly thought about your goals and some might need more thinking. Reaching the right segment at the right time and at the right place is the promise of Behavioral Targeting but if your segments are not properly defined then you will not reach the right customers. Notice that I have not talked about any tool or network yet. You should define your segment before you talk about tools because you want the tool/network to work based on your goals and segments and not the other way around.

Define success metrics

Unless you know how to measure success you won’t know if Behavioral Targeting is really working for you. Think clearly about what success means to you? Is it number of visits or is it conversions or something in the middle. Unless you have a predefined action defined you will not know if you are reaching the right customer or not. Also think about ROI, why do you want to pay extra for BT campaigns or engage in on-site BT activities. The ROI model will help you predict the return you will can expect from Behavioral Targeting and if it is worth your time, effort and money. During this process you might find that Behavioral Targeting (based on your goals, segment and success criteria) is not right for you. If that’s the case then it is time to go back to step 1 and rethink why BT.

Find the right tools/networks

Once you are equipped with the goals, segments and the success measures you can make an intelligent decision about a choosing the right Behavioral Targeting vendor (tool or network). As I said above, make sure the tool/network is capable of delivering based on your goals and segments and not the other way around.

Create a culture of continuous Improvement

As you drive users via Behavioral Targeting, you need to make sure that the creative, landing pages and conversion paths are optimal. Even if you reach the right customer, if you ads, landing pages and conversion paths are not persuasive you will not uncover the full potential of behavioral targeting. Conduct A/B or Multivariate testing on ads, landing pages and conversion paths to create the optimal conversion path. Remember that optimization is an ongoing process, it gets better with time. You have to create a culture of continuous improvement. You will never get to 100% conversion but every little improvement can have a huge impact on the bottom line.

Also check out my 5 steps for easing the privacy concerns with Behavioral targeting.

Questions? Comments?

Friday, May 09, 2008

Web Analyst Interview: Jacques Warren

Continuing my series of interviews with Web Analysts, here is an interview with Jacques Warren.

What is your current position and the name of the company you work for.

I am an independent consultant in Web Analytics. My company’s name is WAO Marketing, which stands for Web Analytics & Optimization (well, I guess I’m not so good at branding!), but it is in fact a one-man show. I intend it to stay that way.

How long have you been working in this fields.

I have been in Interactive Marketing for 12 years now, and I decided to focus all my time to Web Analytics 6 years ago. I knew then that it was impossible to be a good Web marketer without making use of all that data. I still strongly believe that analytics, not just Web Analytics, will deeply impact marketing, and how we do business in general.

Tell me about you Interactive Marketing experience, prior to web analytics.

I was in interactive marketing, mainly in agencies. I did have a small company in the 90’s specialized in localizing Web sites in Japanese, and promote them to that market. For many years, Japanese was the second most used language on the Web after English. I believe it lost this title to Chinese, but Japan is still by far the second largest economy. It made sense then to use the Web in the context of international business, although what I was doing was too niched to make a living. My academic background is Sociology. I did some graduate studies, but the desire to travel got into the way of my thesis, and I never finished it. However, I believe that background helps me a lot today, having trained my mind on abstract thought, analysis, and synthesis.

Why focus on Web Analytics?

How can you manage it if you don’t measure it? Back in 2002, after getting fired for not selling enough Web development projects (ah! Those darn 2000 – 2002 years!), I knew that companies could not keep indefinitely throwing money at the Web without starting the hard business questions about the value of all that. It was the beginning of the web normalization; it was time to treat it as another way of doing business. And if it was business, well, then it had to be accountable.

Tell me more about when and why you created your own company?

I actually created my company in April 2007 when I decided to leave Bell Business Solutions, a Bell Canada subsidiary, where I had started the Web Analytics service offer in 2002. We were working with external clients. I had a really good time then, because we were discovering this new field, trying to make it a revenue stream through consulting and reselling WebTrends, and just plainly educating Web managers about the benefits of Web Analytics. So you could say that I have always been on the consulting side of WA. I’m a consultant at heart. I believe I have the required qualities: expertise, communication skills, empathy, and a deep understanding of organizational contexts. If you can’t understand how well (or badly) a company can integrate what you bring it, chances are good that the project will fail. I will confess something to you: getting independent was the best decision I have made professionally in a very long time.

What are you responsibilities? Describe your typical work day.

Well, of course, finding work is, in theory, a good part of what preoccupies me. I say in theory, because my only merit since I left Bell has been to answer the phone. I am amazingly lucky that work comes to me.
So, my main responsibility is to bring maximum value in usually very short time to organizations who need to evolve in Web Analytics. I do a lot of consulting in KPI, dashboard, analysis work, and WebTrends implementation and training. I work with other applications (Google Analytics, Omniture), but only do sophisticated implementations with WebTrends. I was a Premier Reseller at Bell for several years, and although I am not a technical guy, I had to deeply learn
the product. Without my planning it when I started, I turned out to be one of the top WebTrends people here in Canada.

Besides doing the work I am paid for, I spend a good part of my time in meetings, and interacting with managers. Web Analytics is now understood as being crucial, and people want to learn as much as they can, and understand how they can make it a part of their daily practices.

What, in your opinion, are the skills that you think are important for a web

Having an analytical mind? Yes, sure, of course. But a “synthetical” one too, although that word doesn’t exist. Interpretation resides at that level; the big picture that is constructed through all that slice and dice. You must master your framework, here Web marketing. I believe I am an excellent web analyst because I am first and foremost a great Web marketer. Well, this is not the most humble thing I have muttered in years… But it’s true!

Do I need to stress the importance of communications skills? Analytics is done in such context, politics, ambitious people, P&L obsession, etc., that what you communicate is never purely objective, even though “I’m not saying it; the numbers are!”. You will expose people, plans, projects. Sure, you will as well demonstrate success, but companies are run by humans, who have a strong tendency to like status quo, and status.

What, if any, education or work experience helped you in your job.

I am of course an avid reader in everything Web Analytics, but also in business, marketing, etc. I also attend training, such as Stephen Few’s workshop on data visualization. That was Stephen first public class last June, and it had a big impact on my work. As for work experience, well, I had dozens of projects done when I decided to fly solo; that of course made the decision way easier, since I was offering deep expertise and experience to the market from day one.

What education is lacking, education or experience that would have helped?

More statistics.

What web analytics/online-marketing books have you read and/or own?

All of them, and most white papers out there. Of course, one needs to read
outside one’s field to find stuff that could be applied. Recent readings are:

The Power to Predict

Fooled by Randomness

The Black Swan

Super Crunchers
, and

Competing on Analytics

What were the major challenges you faced or are facing in this industry?

Adoption, adoption, adoption. I have been at it for six years now, and doing a lot of education here in my market (Province of Quebec in particular, but everywhere I go). Although I am very happy about how much awareness our field is getting these days, I am still amazed at how little still many companies do with the information they get. Acting upon the findings remains what it is all about.

How do you make sure you are learning and growing in this field.

I read a lot of books as I mentioned before. I also read all the blogs (but not always all the posts) in the field. I also watch for conferences and seminar that will help me grow. Paying for all that with my own money, I am very discriminating, and choosey; I wish I could go to more. If you happen to work for a company that gives you a good budget for that, don’t hesitate to use it

Do you have blog? If yes, what kind of article do you write.

I actually have several, Analytics Notes ( in English where I am more interested in the relationships between analytics and marketing culture.
Web Analytique & Optimisation ( in French, where I discuss the more basic stuff. The Big Integration ( in English about the data integration stuff. It’s also got a Forum ( WebKaiseki no Kiso ( in Japanese about basic stuff. But I have neglected that one for a while, since writing in that language demands me a lot of time.

What is your advice to aspiring web analysts?

Understand the web and Interactive marketing first and foremost.

Thank you Warren.

Tuesday, May 06, 2008

Site Optimization, Behavioral Targeting and eMetrics

I attended eMetrics on May 4th and 5th.
On May 4th, I taught a full day class on Web Analytics for Site Optimization. This class was a part of the WAA base camp series.
This class showed the tricks, tools and techniques needed to improve the websites, increase engagement and improve conversion rates. Some of the topics that were discussed in this class were
  • Building a measurable site

  • Designing for navigation vs. conversions

  • Improving online persuasion

  • KPIs for ecommerce, content sites, and intranets

  • Understanding visitor behavior

  • A/B & multivariate testing

  • Using advanced metrics such as frequency, recency and latency to increase customer engagement and conversions.

Next day I took the topic of optimization to next step and talked about Behavioral Targeting and how you can use it to increase your online marketing ROI. The topics I covered in my presentations
  • What is Behavioral Targeting?

  • How does it work?

  • What Determines Behavior?

  • What can you target?

  • Why should we use it?

  • Description and examples of Network and Site-Side Targeting

  • 5 Step Process for you to successful Behavioral Targeting

  • ROI Model – Is BT for you?

  • Privacy Issue surrounding BT, results from a survey on privacy and a 5 step process to ensure you don’t run into privacy issues

It is clear that the interest in this topic is increasing. It was evident by the use of this term in several other presentations at the conference and the attendance in my session.

If you missed this presentation and are interested in learning about Behavioral Targeting email me. If you attended the session, I would love to hear your comments and questions.

I wish I could have stayed for the all 4 days of the conference but I had to come back to Seattle to take my US citizenship exam. I am happy to announce that I have I successfully completed the citizenship exam.