Targeting Cart Abandonment by Email

A while ago I read an article called Four Ways to Improve Marketing ROI Through E-mail by John Rizzi, CEO of e-dialog. This is a good article for those who are trying to determine how to collect email, learn from email marking and email effectively. In his last point he says “Use Behavioral Targeting” to convert abandoned carts. He suggests using incentives to bring customers back to complete the cart they had abandoned. This is a great idea but I want you to be aware of following two issues before you jump into it.

  1. Lack of Email Address: If you don’t have an upfront email collection process it is very likely that visitors (customers) will leave even before they give you their email address. If that’s the case then you won’t have any email to target (You can still deploy anonymous on-site behavioral targeting. Check out my article on behavioral targeting).
    If you decide to put email collection up front it might cause cart abandonment rate to go up. You have to provide a very good reason to your customers on why they should provide you email even before they started buying anything or checking out. Like any other change on the site, I suggest conducting A/B testing before you start collecting email addresses for all your customers. If the tests do not show desired result you might be better off with on-site anonymous behavioral targeting.
  2. Backfiring of incentives: Let’s assume you have the email address and are ready to send an email incentive. As you already know the word spreads very fast these days. Most of your customers (visitors) will find out about your offers which could ultimately result in two outcomes:
    • If the incentive is not too enticing (such as free shipping) your customers (even regular customers) might find out about it and start abandoning the cart in anticipation of receiving that offer or they might just use the coupon or offer code given to them by somebody on the internet.
    • If the incentive is too good (such as $10 free for any purchase over $5.00, not sure why would you do that but I have seen companies giving free money just to get users to signup), the word will spread sending new customers to your site. So be prepared to handle the amount of traffic this viral marketing will generate and a possible bankruptcy.
      Appendix A shows what happened to Starbucks when they sent out an e-coupon to limited number of employees (or that’s what Starbucks thought).

So should you provide incentives to bring back customers who have abandoned carts? Yes I think so but think about all the pros and cons before you jump into it. Below are some of the steps that you should include into your process for using email incentives

  1. Select a sample (say 20%) of visitors, who abandoned the shopping cart, who will receive any offer (I am assuming you have already created and tested a process for upfront email collection).
  2. Test different offers within this selected group. Testing will show you which offer works and which ones don’t.
  3. You can use more behavioral data (and I encourage you to do so) to determine what offer will make sense to which visitor segments (create few manageable segments so that you can stay focused). E.g. A customer who abandoned at shipping step might be more interested in free shipping than a user who added products to the cart but then left without clicking on the final checkout button (provided the customer has given you the email address), a 10% off coupon might be a better offer for this customer.
  4. Unless you purposely want to engage in viral marketing, make sure coupons and codes can only be used by those for whom they were intended for and for specific period only. Also don’t forget to configure your web analytics tools properly so that you can measure effectiveness of these offers.

Note: If you provide users the same kind of incentives 2-3 times to a customer then he/she (most of them) expects it every time.

Appendix A: Starbucks Lawsuit
“Starbucks e-mailed the grande iced beverage freebie to a limited number of employees in the Southeast on Aug. 23, with instructions to pass it on to friends and family.
The forwarding turned into a frenzy as the coupon landed in thousands of inboxes and on Internet message boards – forcing the chain to reject scores of coupon-touting java lovers pouring into stores for the perk.” Source:

Originally Posted at : Targeting Cart Abandonment by Email – Web Analytics, Behavioral Targeting and Optimization by Anil Batra

What is Behavioral Targeting?

Behavioral Targeting (BT) is the ability to target users based on their behavior on the internet. Most commonly it used to target online ads but the technique can be very well used to target products and content.

Behavioral (Ad) Targeting promises to precisely target the audience that matter most. Hit the users with the right message, a message that they care about. It is all about audience.

How Behavioral Targeting work?

Users are segmented based on the content they view or actions they take on a site(s) and then are targeted with a message (ad) relevant to that segment.

There are two ways behavioral targeting has been deployed

  1. On-Site Targeting
  2. Network Targeting

On-Site Targeting:

The user are segmented based on content views or actions on one site and then are targeted on the site itself.
For example: Users who view 5 or more pages related to auto and view online interest rates page they can be classified as “In Market Auto Buyer”**. Once classified these users can be then targeted with messages from those advertisers who want to reach “In Market Auto Buyers”.

Network Targeting:

The user are segmented based on content views or actions on one site (usually advertisers) and then are targeted where ever they go on sites participating in the behavioral ad network.
For Example: A user views pages 5 or more related to Alaska tours on a travel site, this and then leaves the site without buying. This user is segmented as “Alaska Tour Buyer”**. A very valuable segment for the travel site. This travel site can then tap into the network and share their segment (based on cookie – more on this later) with others in the network and then target them with relevant message to bring them back to the site and convert the sale.

**(How users are segment and what they are called is totally dependent on how marketer wants to define their segment. There is no industry standard, however every now and then I have seen push for creating industry standard way of creating these segment. In my opinion it is too hard to create standards for creating segment as it totally depends on each individual business).

Originally posted at : Behavioral Targeting 101 – Web Analytics, Behavioral Targeting and Optimization by Anil Batra

A Primer on Multivariate Testing

In my last post on testing I received some great feedback from my readers. One such feedback was around “describing” the principles of MVT. The question was “It would be great if you would be able to get into further detail on the principles behind this (MVT) form of testing”. This question has become the motivation behind this persuasion. In this post  we will recap  the Why’s of testing and experimentation and then jump into the principles of Multivariate Testing and how it impacts business performance. We will also look at designing experiments with Multivariate Testing (Levels, Factors, and Orthogonal Arrays).

Sounds complicated? Don’t worry. The goal of this persuasion is to provide you with a straight forward description of Multivariate Testing so you can walk the walk and talk the talk in real world scenarios. If you stick with me,  I am sure  you will elevate your learning around MVT.

It’s worth pointing out that proactive testing and experimentation offers a sustainable competitive advantage. Majority of the digital properties evolve in response to their competition and that evolution is slow (and poor). Testing and Experimentation serves as an enabler to “bring the organization along” to:

Complement the “art” of web design (right brain) with “science” (left brain) by validating design decisions with statistically significant findings by identifying factors influencing your business objectives. More importantly controlled experiments can help you quantify the variables which have a significant effect on your desired response by screening out analysis of insignificant variables. (Correlation is not Causation)

Optimize site improvements at any stage  of the organizational evolution. If you are new to testing, you can hit the ground running by adopting simple frameworks; if you are already performing some testing you can turbo charge your improvement process by leveraging principles of successful experimental design techniques illustrated in this persuasion.


Less is More


Magnify the impact of Web Analytics. This is my personal favorite as it enables your group to be strategic (vs. ad hoc) and empowers you to accelerate the evolution of your digital presence through proactive discovery.



How do you define Multivariate Testing and What are it’s benefits?

This form of testing involves studying the effects of multiple factors simultaneously  in an efficient way so as to isolate the most important elements and combinations on a desired outcome. I discussed this at a high level in my prior post. This form of testing is ideally suited to Web Analytics (more than any other industry).

It also yields a significantly better set of results with fraction of the effort required by other methods.

Finally, the results for this form of testing are almost immediate with neglible costs when compared to other testing approaches.

…. and its advantages?

Efficiency: By significantly fewer testing iterations than A/B one can take less time and perform test and learn of key variables that drive conversion with minimal cost.

Optimizing Combination of Variables: Allowing for testing of possible interaction effects between variables.

As an example, here is how you can consider performing a Multivariate Test on Google Analytics homepage. In this case user can potentially test Position of Logo, Navigation Bar Type, Splash Page, and Selection of Clients to drive their desired macro (or micro) outcomes.

For example, Google may want to know which of the factors/levels (see description below) contributes to signing-up for their Analytics platform. (Is it a combination of client list, and splash page or none of them  because people sign-up irrespective of logo, nav bar, splash page, or client list).

(Annotations in image  example below)


Caution: Typically when a large number of factors are under study, inefficient researchers abandon the ideas of factorial designs and revert to piece-wise experimentation. Using this approach, one control experiment is performed then additional experiments are performed by varying each factor to a new setting while holding all other factors constant to the control value. This is a poor strategy since the effect of each factor is determined by the difference in response between the two runs. The experimental noise may overshadow the factor effects and there is no averaging of data to reduce the noise.

Another strategy often used when a large number of factors are under study is to arbitrarily, or by opinion, pick a subset of the factors and do a complete factorial design. This is even a worse strategy, since the whole purpose of working with a large number of factors is to determine which subset of the factors are most important. By guessing, it is very likely that one or all of the most important factors will be left out of the design.


What is required to conduct a successful Multivariate Test?

Real-time randomization of traffic to ensure statistical validity and prevention of non-randomization bias. For example, if a plant is switched from one supplier of raw-material to another and notices a deterioration in yield one explanation is that the new raw material is worse than the other but another plausible scenario is normal drifting of the process due to other things changing gave a worse yield after the switch so the raw material source had nothing to do with the drop in yield. (A valid test in this case would be a random switch back and forth several times between suppliers and then test the outcome).


Who and What is Taguchi?

Genichi Taguchi was the person most responsible for developing high quality Japanese cars. He developed a technique to find the sources of variance and the method (Taguchi)  is named after him.

Taguchi Method is the approach popularized to find sources of variance. It leverages Taguchi arrays which are orthogonal arrays that minimize the number of experiments needed to gather information for analysis.

Taguchi Method

What are Factors?

Factors are elements that you want to test.

In the case of Amazon Kindle you would want to test the Banner Color, Kindle Color, Price Text size, and Button.


What are Levels?

Levels are options for each factor.

In the case of Amazon Kindle you could have the Banner color in (Lime, Blue, or Crimson), Kindle color in (Black, White, and Grey), Price Text Size in (18 pt or 20 pt), and “Buy Now” button vs. “Order” button.


What are Orthogonal Arrays?

Orthogonal arrays are matrices that show which elements to include in an experimental run.

Run 16 => A2, B3, C1

Factor A Level 2, Factor B Level 3, Factor C Level 1


What is Full vs. Fractional Factorial?

In Full Factorial an experiment is run for every combination of levels. Example: 3 * 3 * 3 = 27 runs.

In Fractional Factorial an experiment is run for only a fraction of all possible combinations.


Comparison 9 runs vs. 27

Net, net, here are the 7 steps to complete your MVT journey

1. Determine measure you want to optimize. (Hypothesis generation)

2. Come up with possible changes.

3. Use number of changes (Factors) and options for each change (Levels) to detemine which runs/configurations are necessary.

4. Create a page for each experimental run.

5. Ensure that you minimize the non randomization bias by ensuring traffic is randomly sent to each of the test pages.

6. Run Tests.

7. Measure results in your favorite platform.

MVT Results Google Optimizer

(Source: Google Optimizer)


That’s it when it comes to nuts and bolts of MVT.  In a follow-up post I will share a live example from a site to conduct a potential Multivariate Test.

Contributed by Kanishka, originally posted at

Decreasing Page Load Time Can Increase Conversions

Do you know that load time of your site can have a big impact on your conversions? KissMetrics in it’s post “Speed Is A Killer – Why Decreasing Page Load Time Can Drastically Increase Conversions” writes:

According to surveys done by Akamai and, nearly half of web users expect a site to load in 2 seconds or less, and they tend to abandon a site that isn’t loaded within 3 seconds. 79% of web shoppers who have trouble with web site performance say they won’t return to the site to buy again and around 44% of them would tell a friend if they had a poor experience shopping online.


How to Test your Page Load Time

  1. Page Speed Site from Google
  2. Web Page Test:
  3. Page Speed Browser Plugin:
  4. Google Analytics Plugin:

How to Decrease Page Load Time

Here are some tips from Kissmetrics to reduce your page load time:

  1. Use GZIP compression – it can significantly speed up a site, reducing file size by as much as 70% without degrading the quality of the images, video or the site.
  2. Wrangle Your Javascript and Stylesheets – Have your scripts and CSS load in external files. This way, the browser only has to load the files one time, rather than every time someone visits each page of your site.
  3. Optimize Your Images – In Photoshop or Fireworks, you can use the “Save for Web” option to drastically reduce image size.
  4. Don’t Rely on HTML to Resize Images – HTML (and by extension, WordPress blogs), make it easy to resize the images on the fly before it is shown to the user.  But just because you load that smaller size, doesn’t mean it’s taking up any less room on the server.
  5. Web Cache Me If You Can – Content management systems like WordPress have plugins that will cache the latest version of your pages and display it to your users so that the browser isn’t forced to go dynamically generate that page every single time.
  6. Don’t Confuse the Browser with Redirects – A lot of 301 redirects piled together just confuse the browser and slow it down as it wades through the old destinations to get to the new one.
  7. Let the Network Carry the Load – Content Delivery Networks serve pages from the server located near uses geographical area, which means they get the site to load sooner.

For detailed post please visits,  “Speed Is A Killer – Why Decreasing Page Load Time Can Drastically Increase Conversions”

Ultimate A/B and Multivariate Testing Resources

Here are the resources to help you learn A/B Split Testing and Multivariate Testing (MVT).  We will continue to update this page so that you have the latest information. If you find something that we should include then please email us.


Tool A/B or MVT Price Site
Google Website Optimizer Both Free
Omniture Test & Target Both Varies
Visual Website Optimizer Both $49 and Up – Free Trial
Sitespect Both Varies
Maxymiser Both Varies
Ioninteractive Liveball Both $1295 and up
Optimost Both Varies
Vertster Both Varies
KaizenTrack Both $197 fixed price
SiteTuners Tuning Engine Both Varies
Webtrends Optimize Both Varies
Split Test Accelerator MVT $297 fixed price
Hiconversion MVT Varies
Conversion Multiplier MVT Varies
Unbounce A/B $25 and up


Blog Posts, Articles & Sites