ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

Who Moved My Cookie?


Let’s make a deal.  I promise not to bore you any more with information that has no relevance to you.  I promise to stick to topics that interest you.  I promise not to repeat myself.  And I’ll throw in another promise: I won’t talk about you to anyone else.

All we need to do to get started is one little thing: You have to tell me about yourself.

Have we got ourselves a deal?

Online marketers are amassing more and more tools to speak meaningfully to carefully targeted audiences, and survey data suggests that consumers are at least reluctantly willing to agree that they would prefer targeted, relevant messages.  But that doesn’t mean consumers are really going to play along.

The latest data about cookie deletion and rejection has set off a round of debate and hand wringing.  Are consumers deleting cookies every month?  Are they blocking cookies even before they are installed?  How are marketers to deal with these recalcitrant consumers?

JupiterResearch kicked things off with its report that 39 percent of internet users were deleting their cookies at least once per month. The implications for anyone involved in online advertising, website analytics or online marketing are obvious. If users are regularly deleting their cookies, that’s going to distort the picture of their online activities and reduce marketers’ ability to serve up targeted ads or marketing programs.

Then, data from Atlas DMT suggested consumers actually deleted cookies far less frequently than they (the surveyed consumers) said. But the web analytics and measurement firm later came out with a retraction, essentially affirming the original Jupiter findings.

Next, research survey company InsightExpress weighed in with new survey results that supported the original Atlas DMT theory, i.e., that consumers didn’t know what they were talking about. For example, while 77 percent of consumers in a survey said they know what a cookie is, only 30 percent could provide an accurate description. Further, while a large percentage said they deleted their cookies on a regular basis, when asked to prove it by physically deleting their cookies on their computers, only a small percentage could do so.

WebTrends capped the ongoing debate this week with a report on rejection rates for so-called third-party cookies, which track a site's unique visitors and their responses to marketing campaigns and website promotions.  Rejection rates ran at 12.4 percent in April 2005, WebTrends said, up from less than 3 percent in January 2004.  Outright cookie rejection is far more serious than occasional cookie deletion, WebTrends argues, because it has an immediate affect on the accuracy of measurement.

All in all, it’s clear that a significant chunk of consumers are blocking or removing cookies, either actively or through the use of installed software.

This shouldn’t surprise marketers.  However much consumers might buy into the notion of targeted ads, and recognize that there is a value to them, privacy continues to be of the utmost importance to most people, and an instinctive distrust kicks in whenever they are asked for personal information.  Those ingrained attitudes aren’t going to melt just because marketers have neat new tools to work with.

Consider the results of a 2004 survey by the Ponemon Institute that probed consumer attitudes about divulging personal information.  Only a handful of information categories were acceptable to more than half of the respondents -- home location and telephone, gender, email and IP address, and education.  This kind of bare-bones information isn’t nearly enough to unlock the door to highly targeted ads.

It seems almost counterintuitive that consumers would be more protective of ephemeral information like reading preferences than of real world data such as phone numbers and home addresses.  But that gets at the heart of the privacy issue: consumers recognize that they live in a highly connected world, and realize that much of the fundamental data of their lives is probably only a mouse click away if anyone wants it.  To preserve their sense of privacy, then, they may feel even more, not less, protective of largely symbolic information like religious affiliation or taste in literature.

Privacy is a hot button issue.  A November 2004 survey of US Internet users by ReleMail, an email privacy certification service, found that 96 percent said that email privacy was important to them.

Many internet users are doubly on the alert because their online behavior is tracked at work. 

With the boss looking over one shoulder and a marketer looking over the other, it should come as a surprise to no one that a fair number of Internet users are going to look for ways to lock out prying eyes.

Even so, there are still plenty of users who seem to understand the quid pro quo – trading personal information for a better online experience (i.e. free content supported by more relevant advertising information).  A TRUSTe survey found a general understanding of why companies collect information -- and nearly half of the respondents said they trusted companies to safeguard it.

The latest data on cookies is not the end of the debate, or the death knell for targeted marketing, but a reminder that new technology and tools will only go so far when they come up against human nature.

Ezra Palmer is the editorial director of eMarketer. eMarketer publishes data, analysis and market projections focusing on e-business, online marketing and emerging technology. Founded in 1996, eMarketer aggregates, filters, organizes and analyzes data from more than 1700 research firms, consultancies and government agencies around the globe.

Return to Hella lame or hella tight email marketing

Beyond email: widgets, sponsorships and content; oh my!
Of course not every brand can break through the email clutter, so it's important to stay up on new marketing opportunities to reach teens. The buzzword of 2007 has got to be "widgets." Our clients want to know more about them and whether or not they should utilize them now or in the future. The problem is most clients don't have the library of content necessary to push new stuff out to a widget every day or week. So the best option for them would be to sponsor or skin an existing widget. 

Wait a minute... back up... just what is a widget, anyway?
Wikipedia defines a web widget as "a portable chunk of code that can be installed and executed within any separate HTML-based web page by an end user without requiring additional compilation. They are akin to plug-ins or extensions in desktop applications. Other terms used to describe a web widget include gadget, badge, module, capsule, snippet, mini and flake. Web widgets often but not always use Adobe Flash or JavaScript programming languages." Blah blah blah. Let's look at a picture instead.

(click to view)
Because content is king in this digital age, social networking sites like MySpace and Facebook allow embedded widgets that bring relevant content to teens' profiles. On Facebook, a widget/application called "Food Fight" enables teens to buy virtual food and throw it at their friends' profiles. To earn money to buy food, they answer marketing research questions like "What kind of computer do you use as your primary computer?" and "Which of these new FOX TV shows are you going to watch this fall?"

Interesting, right? They answer questions to earn fake lunch money to buy virtual food to "throw" at other teens' profiles. Why? For fun. (By the way, in the Food Fight widget/application, I found out you can fling poo. Take that, Jane Goodall.)

Put on your thinking cap
Unfortunately, marketing in the post-email digital age isn't easy. There's no set media plan you can plug into (yet) and results are still difficult to quantify. But there's a bright side for marketers willing to put in the extra effort and try something new: You face less competition in the crowded digital media landscape and you might just be the first in your category to reach this fickle audience where they live.

That's great in theory, but what would be a specific marketer's opportunity in this new world of widgets? Let's brainstorm just a few quick, random solutions. Is your thinking cap on?

If your product is a food, why not talk to the creators of Food Fight and negotiate to have your food item added to that application? What if you market a cleaning product? Again, the positive brand interactions possible by associating Bounty paper towels or Mr. Clean with Food Fight are endless.

Or maybe you're selling kitty litter. BunnyHero Labs creates virtual pets that teens add to their MySpace profiles and send to friends. Sponsor the adoptable cat and feature your kitty litter in the virtual pet widget. Or create an original widget that's a kitty litter Zen rock garden.

At the very least, voluntarily add the logo of your product or service to the HotList application on Facebook. It's free and more than 1,120,194 users have already added it to their profiles, to list the products and services they like.

Thinking outside the inbox in a post-email world doesn't have to be scary. Just watch out for the flying food.

Tiffany Young is executive creative director at Smashing Ideas Inc. Read full bio.

Step 2: Asking the right questions
Once you've determined the goals of your project, you'll want to develop targeted research questions to identify the most effective ways to move toward those goals and to assess how well you're doing in achieving them. One important thing to consider is how you'll use the answer to your questions. If the answer is not helpful in meeting your business objectives, then there's not much point in asking the question.

Here are some examples of the types of the questions you may want to ask, and how they relate to different goals.

Goal: Increase sales
Research questions: What types of customers do I have? Which customers are the most profitable? Who are the new customers to target?
What we hope to accomplish:

  • Identify different types of current and future customers and the best ways to reach them.

  • Learn about relative spending patterns to help you develop a targeted website that addresses the needs of your most valuable customers or prospects.

  • Explore the demographics of your audience: young or old, social networkers or web introverted, web-savvy or non-technical, loyal to your brand or cruise-on-by looking for the best deal.

Goal: Increase value of the brand
Research questions: How is my brand perceived?
What we hope to accomplish:

  • Assess baseline awareness and perception to identify problem areas and allow you to measure changes as you increase loyalty, raise awareness, or drive people to your site.

  • Map the competitive landscape to identify the features that are associated with your brand and its competitors, and the emotional reactions to your brand and others.

Goal: Grow customer base
Research questions: What will attract people to my site? What will keep visitors engaged? What features provide the greatest opportunity for growth?
What we hope to accomplish:

  • Identify the features most likely to lead to conversion so you can focus your resources on the areas that will be most effective.

  • Assess the relative importance of various features and how satisfied customers and prospects are with the offerings available to them to identify areas of maximum potential.

Goal: Increase repeat customers
Research questions: For frequent tasks, what's the most efficient experience for my repeat customers? For new tasks, what's the easiest path to learning?
What we hope to accomplish:

  • Identify the optimal experiences for both first-time and expert users to make it more likely that they will return.

  • Compare pages across your site to see which are most effective and which lead people to leave your site, so you can create a consistently positive site experience.

  • Learn how visitors are using the features available to them so you can optimize user experience and encourage the actions that are most valuable to you (such as conversion events).

Step 3: Using the right methodology
Mapping the best research tool or tools to the question you want answered involves understanding the strengths, weaknesses, and costs of each technique. There are several ways of distinguishing between the types of data you can collect. All of these should factor into your choice:

What data sources can I access?
There is a variety of different data that you can collect from your website (using web analytics), from your customers and prospects (using interviews and surveys), and from the internet at large (using click stream data or assessing social media discussions). The type of data that will be most appropriate depends on your research questions, your budget, and your timeline.

Do I want emotional, cognitive, or behavioral data?
Emotional data reflect individuals' feelings and emotional beliefs -- data that are well suited to brand investigations. Behavioral data reflect individuals' actions and are the usual focus for usability testing. Cognitive data reflect individuals' understandings, intellectual beliefs and attitudes, and are needed for understanding the person's mental model of your site.

Is my question best answered by observation or self-reporting?
data are based on observation, either directly watching what people do, or examining the impact of their behavior on such things as market inventory or click stream data. Self-reported data are goals, attitudes, intentions, etc., mainly extracted from what people say.

In general, self-reported data are more cost effective than observed data (the exception is data from your own website), but are subject to a variety of biases. Respondents may deliberately lie, can answer randomly, or may simply have poor insight or memory about their own behavior. 

Do I need quantitative or qualitative data?
Qualitative data, characterized by open-ended queries and small sample sizes, allow for deep exploration and insight in a variety of areas, including many that are uncovered during the data collection process. Quantitative data, characterized by close-ended questions and large (often statistically significant) sample sizes, are useful for estimating the prevalence of specific variables or quantifying differences across groups. Qualitative is good for brainstorming and idea generation, while quantitative is best for proving something.

There are inexpensive quantitative data tools available, and team members may be able to design and conduct interviews. Thus, it is possible to gather and analyze both quantitative and qualitative data quickly and inexpensively. However, collecting and interpreting data effectively require trained professionals. You're more likely to glean useful information if you invest in good software and experienced personnel to design the research questions, develop the appropriate instruments, and analyze the data.

Is my question best answered using a combination of data?
Different types of data can be combined in a number of ways to best achieve your goals, for example: 

  • Quantitative, self-reported emotional data, such as a survey in which you ask your customers how they feel about your product. This combination can be a strong strategy for branding goals.

  • Qualitative, observed behavioral data, such as a user testing session in which you present your new social media platform to a small group of people and note how each of them interact with it. This combination can be useful if your goals include engaging customers and potentials using Web 2.0 applications.

  • Quantitative, observed cognitive data, such as a small pop-up quiz on your site where you ask a large number of visitors questions about your product and see how much they know. This combination can be a great way to assess baseline and benchmark data as you move toward a goal of raising awareness about your product.

What's my budget and timeline?
Last but not least, different research techniques come with different price tags and time constraints, and have differing returns on investment. A research project may be best served by one more-expensive technique that is comprehensive and thorough, or several less-expensive "quick hit" techniques that can be compared against each other.

Putting it all together
Now that we've identified the steps involved in determining which research technique will best allow you to reach your business objectives, let's use a hypothetical case study to illustrate the approach.

A major airline decides that it wants to reduce overhead costs by improving online ticket sales. The business goals are to increase ticket purchases and reduce overhead by increasing successful completions (so there are fewer calls to the help center and abandoned purchases). Movement toward goals is easier to assess when the goals are quantified. Let's assume that the hypothetical airline wants to increase ticket sales by 2 percent, and reduce call center frequency by 2 percent.

Now that the business goals have been well-defined, we can start formulating research questions:

1. What kinds of customers currently buy flight tickets online? What types of new customers can be convinced to start doing so?
Increasing sales requires retaining current customers and adding new ones to the mix, so we need to know about both market segments before we make design changes. If the research identifies those customers who can be newly attracted to online tickets sales, a site can be designed to appeal to them.

2. What type of purchasing process will maximize conversions and successful ticket sales?
It's important to attract new people into the purchase funnel, but it's meaningless if they abandon the process part-way through. If roadblocks are found in the purchase funnel, then those blocks need to be fixed.

After defining the questions, it's time to develop a research methodology. A variety of techniques can be used to address the questions articulated above:

  • For the first question, a survey is conducted and the data gathered are used to create quantitative personas. The personas identify what's important to current customers, and to prospects that may be ready to start buying tickets online.

  • For the second question, a usability study of the current purchasing process is conducted to find out where people are dropping off and why. Web analytics can also be used to learn which parts of the existing site lead to user abandonment. All the information gathered helps to identify the most important features and guides the redesign of the purchase process. The data can also inform marketing techniques to drive people to the first purchase page.

  • Finally, the effectiveness of our changes needs to be assessed. To address this, multivariable optimization is applied to the new design after launch. This technique allows the airline to continue to tweak small changes in things like buttons or text choice, and continue to learn about what features are most likely to lead to sales (and which features users aren't interested in).

When starting a new design project, take the following approach:

  • Collect the business goals and objectives: What are you trying to achieve with a new project or initiative?

  • Identify what you don't know and what's most important to research.

  • Select a research tool that will tell you what you need to know, and will give you actionable results and intelligence. For a full list of tools, consult this appendix (PDF).

Targeted and well-designed research is just as critical as a targeted and well-designed website. You are the best judge of what information will provide you with meaningful, actionable information, but you may want to work with someone who has expertise in selecting appropriate data collection techniques to ensure that you get that information and analyze it in a way that meets your goals. Given the array of traditional and digital tools available, it is worthwhile to take the time to consider the method or methods that will give you the greatest returns.

Joanne McLernon is experience design consultant and Jenny Gutbezahl is senior strategy analyst for Molecular.

On Twitter? Follow iMedia at @iMediaTweet.

Social expansion

Let's say you've got a solid content creation team and content marketing plan in place.  You've got your email newsletters running, your blog established, plus Facebook, Twitter, and other social media outlets. How do you expand your following? How can you significantly increase your traffic, clicks, and ultimately your business? It's not as easy as we'd all like it to be. Of course, creative copy, viral videos, and intelligent content certainly increase share-ability and visibility, but there's more.

There is another valuable asset that many companies have not truly tapped into: employees. More and more employees want to participate, but the current company environment may not be encouraging or providing incentives for them to share, and in fact, may even be dissuading them. Most employees like talking about what they do, and this authentic person-to-person communication creates more trust than buying impressions through media purchases. In order for employee engagement in social media to be successful, companies need to have a content marketing plan in place.  

Companies need to have employee social evangelism policies in a marketing plan to provide some rules of engagement. And this is not as daunting as it may seem.

Success in action

For example, Citrix is one of the companies trailblazing the way, engaging its workforce in employee social evangelism efforts, with phenomenal early results. It launched a program bringing its employees on board and making it easy to share interesting, informational, or entertaining content via its existing social media outlets. Here's the traction Citrix has experienced so far:

In the first 30 days, starting with 43 employees, it saw:

  • 200 social content shares

  • 615 clicks on shared content

  • 419 site visits

  • Four leads collected for webinars

Now, how can an individual or small social media team compete with that? Being able to measure these results provides marketers with a solid indication of what's working.

So in summary, start with a solid content marketing plan, build quality content creation teams, and leverage channels including email and social media. Then expand your voice beyond the social media team, making it easy for your employees to socialize all the great content you're producing.

People are naturally social, and your employees are already active across multiple social networks. Make sharing content you care about easy for them, and you'll see a dramatic increase in your brand awareness, while providing value to customers and prospects. 

Eric Roach is co-founder and CEO at EveryoneSocial.

On Twitter? Follow iMedia Connection at @iMediaTweet.

Ezra Palmer heads eMarketer’s editorial group, managing a team of researchers, interviewers, writers, analysts, forecasters and editors as they collect, analyze and contextualize data from thousands of research sources worldwide for the...

View full biography


to leave comments.