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

Making Web Analytic Standards


There are very few standards in the field of web analytics. The field is relatively new; it was only a few years ago we were trying to decide what to call it ("e-metrics" was the favorite until people realized it was too close to "emetics"). This is why most online activity is still not measured at all, and why there are few standards. At this time there is little agreement on what should be measured. Where there is agreement, there are often differences in how that metric should be calculated. In extreme cases, the same system may vary in different places, according to the preferences of the individual programmer who wrote that section of code. This means some systems can generate different numbers for the same thing, depending on how that number is accessed by the user. In other cases, the same metric is given different names by different systems, leading to confusion over terminology.

In practice, the degree of confusion is bearable because there are not many software vendors in the marketplace. However, if you find yourself in a dispute over numbers, you'll often find this is because they were calculated in different ways. It's always advisable to begin any such dispute by comparing methods of data gathering and calculation. If you do have such a conversation, you need to anticipate that some issues may occur for which we do not have answers yet. One which keeps coming up for me at the moment is "how much does it matter where the tracking tag is placed?" I know placing JavaScript tracking tags in the is less accurate than in the . I'm pretty sure the degree of variation depends on browser and operating system, but I couldn't put numbers to it. There is even less certainty about the amount of difference placing the tag at the top or bottom of the makes, but many people suspect it does matter. Needless to say, there is no standard about how this should be handled, and vendors are unlikely to provide even product-specific guidance-- most of them aren't aware this is an issue.

Agreed international standards are essential to ensure we all measure the same things, and do so in the same way. At the same time, it's important to ensure vendors support these standards, and implement them in their software.

There are a number of bodies involved with the development of standards for web metrics; the Audit Bureau of Circulation Electronic or ABCE (www.abce.org.uk), Joint Industry Committee for Web Standards or JICWEBS (www.jicwebs.org), and the International Web Analytics Association or WAA (www.webanalyticsassociation.org). 

Initial development of web metrics came out of online advertising in the 1990s. Since online advertising needed to prove its numbers in order to get paid, it was the first industry to see a need for metrics. Hence the first organization to form and develop standards was ABCE.

ABCE is based in London. It is formed from a coalition of large websites, agencies and a few of the larger web metrics vendors. In the late 1990s it developed a set of standards focused around the measurement of ad delivery. For the last few years ABCE's primary role has been using those metrics for auditing. ABCE will also certify metrics systems as meeting their standards. This means they will trust the figures the software provides when they audit. This doesn't mean they will refuse to audit your site if you use a different product; it just means you'll get a discount on the audit fees if you use a product they have certified.

It is important to understand that ABCE is not interested in providing a comprehensive set of metrics for measurement of all online activity. They are "owned" by their members who determine their agenda. Their members are in the business of online advertising, so the only metrics they need are those related to ad delivery. The majority of the definitions created by ABCE are of little use for most site analysis and include email distribution, interactive TV, PDA activity and so forth. However, some of the fundamental web metrics definitions, such as visit, page impression and unique visitor come from ABCE.

JICWEBS is the real heavy hitter in this field. ABCE provides the definitions for JICWEBS.  JICWEBS standards are fed into the International Federation of Audit Bureau of Circulations, or IFABC. IFABC standards are supported in 34 countries, including the United States, Canada and most of Europe. 

Not all ABCE standards are supported by JICWEBS at present. The web metrics defined at JICWEBS are Page Impression, Unique User, Visit, Visit Duration, Search, Stream, Stream Duration, Ad Impression and Ad Click. Notice there is no definition here for behavior over time (such as Repeat Visitors), anything related to browsers (such as User Agent) or anything related to links (such as Referrer). If you want to calculate frequency capping over longer time periods, or traffic between sites, you're not going to be dealing with anything which has an agreed standard.

ABCE and JICWEBS had effectively ceased development of standards by 2002. They had enough for their purposes, and extending them any further would have placed pressure on their members. Sources inside ABCE told me at the time members had no desire to create metrics that could compromise the income of advertising outlets or force analytics vendors to create new features in their products. Remember, the aim of ABCE and JICWEBS is to support auditing-- to enable you to compare two potential outlets on a common basis; not to create a detailed, or even accurate, picture of online activity. It doesn't matter if the numbers are a little inaccurate, so long as the inaccuracy is the same for everyone.

The International Web Analytics Association was formed in 2005. Part of the reason for formation was to provide standards for web metrics. Most web analytics professionals recognized the limitations in ABCE standards, but unless you lived in London, were prepared to pay over $5,000 a year to join ABCE, and were ready to fight the established status quo, there wasn't much you could do about it.

The WAA Standards Committee contains a variety of members representing the complete spectrum of those who work with web metrics, ranging from senior analysts at Yahoo!, to private consultants and ad people. It started by adopting most of the relevant JICWEBS standards. Since then, the committee has been quietly developing new standards, meeting once or twice a month in an online conference to argue and debate, then finally agree. It could be said they have been a little too quiet, in that they have neglected to publish the standards they have agreed upon. In late 2005 two members, Jason Burby and Guy Creese, published an excellent discussion paper containing a proposed list of KPI's on the WAA site. The standards committee has since been working on ratifying or tuning some, but not all, of those metrics. The initial set of standards they aimed to ratify has been completed, but as yet they have not been published.

The standards committee is now working on the next set of terms. These are Page Views per Visit, Page Exit Ratio, Call to Action Conversion, Average Order Value, Average Items per Order, Visits per Visitor, Frequency, Recency, % of Content and Cost (per campaign, visit, transaction, et cetera).  It is to be hoped they will soon publish the list of standards they have already ratified in a definitive document.

The future
The creation of WAA prompted ABCE to start developing standards again, and for a while it looked like ABCE and WAA would start creating conflicting standards. However, both organizations have recently agreed to work together to prevent this happening. At this stage the agreement is nothing more than a statement of principle. It remains for these organizations to develop formal methods for regular liaison, and to define mechanisms for joint ratification.

Bearing in mind that there are very few published standards, it is not surprising that vendors of analytics software have each been doing things their own way. WAA is just now starting to think about creating methods of certification for analysts, and I would expect this to extend to software at some stage in the next few years.

In the longer term we can expect a full and comprehensive set of standards to emerge, though this will take several years. If you are working with web analytics at this time, bear in mind you are working in a field where most people can legitimately "do their own thing." If you find someone disagrees with you about key numbers, start by establishing a common language. If JICWEBS and WAA can't give you the metrics you need-- agree on some standards of your own.

Brandt Dainow is CEO of Think Metrics, creator of the InSite Web reporting system. Read full bio.

It's easy to think of an album split into tracks and sold individually, but what about higher-priced items? Thankfully, all these users have accounts -- accounts that are unique. Because of this uniqueness, you can stack micropayments together, like coupons, collecting them until you're able to make a purchase.

A teenager may not have the hundred dollars he needs for the hot new sneaker, but he does have, on average, 115 friends on social networks that wouldn't think twice before sending a buck in a click to help their friend get to the sneaker promised land.

Virtual gifts

This could be an entire series of articles unto itself, but virtual gifts are today's "Too Good to be True" story for interactive marketers.

The premise is utterly simple: Create a small icon that people can save somewhere on their social networking profiles.

As simple as the concept sounds, sending people small icons is generally untracked as an overall trend, but it's $15 million a year if you're Facebook (report by Lightspeed ventures). Balloons, care bears, images of drinks… if it sheds a sliver of light on someone's personality or relationship, it's often worth a click and a buck.

The Lightspeed report shows a tremendous disparity between free virtual gifts and their dollar-costing compatriots. Their bit of tracking shows free gifts given in the hundreds of thousands, as virtual gifts costing a dollar going at the snail's pace of a few hundred a day, with demand usually driven seasonally or around holidays.

From a brand perspective, both scenarios are a win. Give a free gift, as Axe has done with the "do not disturb" door-hanger-looking virtual gift, and a few hundred thousand folks will show them off to their entire network of friends. Charge a dollar and give audience members the feeling of exclusivity within your brand, and get hard return on investment for your development costs.

Put in the context of a widget, your brand can become a virtual gift hub. Every gift (if given "publicly;" gifts can also be given so that only the sender and the recipient can see them) not only posts to the recipient's profile, but to the pages of all the people in their network as an update. Best of all, it uses resources you already have in place.

Your strategists have already outlined all the psychographics that tell you what perceptions your audience members hold to help define their outward-facing personas. Your designers are constantly coming up with ways to explore those perceptions visually. Make a widget with a sharable selection of icons that can be transferred to any number of social platforms, and watch the engagement take off.

The fact is, people love collecting and sharing these trinkets, these artifacts of their personality and friendships. And your brand becomes the conduit for this emotional connection. Naturally fun and viral, when paired with micropayments, there's no end to the revenue and brand-building possibilities.


Don't groan. Yes, engagement is the cliché du jour to describe all kinds of good audience interaction. However, clichés don't become clichés unless everyone has accepted a certain universal truth about them.

One shining example of widget engagement comes from an unlikely spot: the University of Illinois, Chicago.

Initially looking for an efficient digital communications channel, as any locally based campus would, the athletic department pushed its chips onto a widget strategy that has paid off exponentially.

Again, the concept is simple: a slideshow of sports action, sports news tickers, a button to play the fight song on demand, and links to buy gear or tickets on the UIC website.

Creating a Flash widget that can be posted to just about every social media platform as well as desktop platforms like Apple's Dashboard, UIC received more than 100 percent ROI even before launching. And just showing the comps to sponsors garnered deals with AT&T and Adidas.

Since launch, the response has been astounding. Thousands of people from all over the social media map have installed the widget. In the first month alone, it garnered more than 10,000 interactions.

Even more impressive is the engagement. Month to month, the average engagement time per user is 22 minutes. You did read that correctly. Every month the audience for this widget spends as much time with it as they do a network television show. What's the production budget for 22 minutes of televised programming?


Over at Gogo, we've got a brand that's launching and positioning itself as the nation's provider of Wi-Fi internet to commercial airlines. As a new brand, it's in need of engagement and exposure.

Working with partner agency Upshot, Gogoinflight.com features a game in which you're a cell tower beaming beautiful Wi-Fi signals to the planes overhead.

This widget also continues to pull new downloads every single day, creating an advertising network of advocates money simply can't buy as effectively. To the delight of everyone involved, Gogo has given early adopters and evangelists a convenient tool to spread the word, with Gogo maintaining complete control over the messaging -- and fun -- as they quickly build equity with the people who mean the most to these early months of the brand.

But as any brand-launching marketer knows, frequency is crucial to establishing brand recognition in these media-saturated times we live in. On average since launching the widget, on every game-play, the Gogo gamer is exposed to 80 branded messages.

An added benefit of this widget strategy is understanding where the audience "lives" online. Initially, the thought was that the most play would come from the major players like Apple and Facebook. While these two monsters certainly earn a good share of the interactions, the largest groups of downloaders and active users are actually coming from hi5 and friendster. This is incredibly valuable insight to strategists looking to plot the future of the brand's interactive channels.

There are great things happening all over the world of widgets these days, clouded by the fog of terminology and, well, basic awareness. But there's gold in those hills. All we need to do as brand strategists is to look for simple ways to engage users on their terms.

The return, at the very least, is valuable insight on user habits, using the very same talent you're currently employing to make banner ads and microsites. At the very most, you'll be creating a network of advocates pushing engagement, awareness and significant revenue with every click of the mouse.

If there's one takeaway I hope you act on, it's just to get involved. Sign up on these networks. It's free to check them out and fun to experience. Just being polite and interested will help you to discover new insights and tactics to activate your audience that you won't find any other way.

Michael Leis is VP of Emerge Digital (formerly Publishing Dynamics), and writes What’s The New Media Buzz?

Embracing the role of marketing technologists

Many in the industry still staff their teams in traditional agency structures. This follows some mixture of account management, planning, and creative. But what is missing and absolutely necessary in today's world is the role of the marketing technologist.

This is not just a technical resource nor is this a traditional digital strategist by agency standards. These individuals have evolved to bridge technology and marketing, combining the best of both worlds. They understand the nuances of how to integrate complex elements such as SDKs or APIs while also understanding the nuances of mapping feasibility in support of the creative big idea.

This is the individual who understands how to "hack" native platforms and offer new perspectives on integrating contextual programs to the delivery model that will create the most impact. This role is critical moving forward as we head into a world that will be more system-centric than in years past.

Exclusive access and partner ecosystems

A number of my associates from various agencies acknowledge they want to own "the big idea" and "innovation" and are reluctant to understand that truly groundbreaking programs and first-of-its-kind campaigns can come through co-creation with publisher and technology partners.

This is achieved through investing in and cultivating key relationships that lead to alpha and beta access and advanced insight into product roadmaps. This is essential to stay ahead of the competition. Partners also bring additional strategic and creative resources that are deeply immersed in their platforms and can enable agency teams to seamlessly align creative ideas and themes and apply them in new and unique ways.

Although many agencies view their shop as the center of the creative universe, the fact is everyone is a publisher and a creator. Those that adjust and understand how to leverage branded content, co-created content, the integration of content into existing programs, and displaying curated content for impact have a distinct advantage when it comes to creating compelling experiences that consumers want to engage with.

Innovation as a service

I have seen a number of agencies try to implement tactics like opening an "innovation lab," doing a quick beacon test or hacking a platform to drive good PR and out-of-the box client presentations, but in order to truly transform a business, it is critical to understand how to integrate innovation into an existing, sometimes very linear organization.

This starts with having a focus on mapping to the startup ecosystem. Aligning creative teams with access to new and emerging products and organizations can lead to a breakthrough program or partnership. It's also important for agencies to have a role in the startup community. I have participated as a formal advisor for many startups over the years and have found this to be a good method to stay in tune with the latest developments in categories of interest.

In the early 2000s, many agencies carried large tech staffs to execute programs. Now the modern agency is normally very light when it comes to development and user experience resources. A large number of agencies do not have a chief technology officer or a senior tech leader who can serve as a bridge between an organization and tech heavy partners.

Also, many agencies are not set up to invest in the creation or co-creation of new products. At times, it can be advantageous to actually create a product, but most agencies are not structured to support product creation.

The final and most important point about agency innovation is having the expertise to guide clients to understand when innovation has hit a threshold that ultimately converts from emerging to fully integrated.

Proprietary data assets

There are a lot of agencies that talk about data, but the reality is only a few have proprietary data assets that cover attitudinal, behavioral, demographic, and most importantly, transactional data that can lead to personalized connections with consumers at scale.

Even fewer agencies are structured to capitalize and convert data into "smart data" that can create impact for a client's business. This comes back to how agencies are currently structured. The biggest weakness I have seen in many shops is a gap in analytics and attribution. This is critical to understand impact and, ultimately, success.

Also, there is a misconception that leveraging data is only a single piece of the planning process. With so much information being created on a daily basis and consumers sending up need-state signals at every turn, it is critical to be set up to capitalize on connections across the non-linear path to purchase.

This can be in the form of data narratives that inform creative and behavioral elements aligning channel strategy and message architecture as well as optimization of media and attribution of program effectiveness.

Having access to proprietary data as well as the structure to support the analysis, both from a human capital and machine-enabled analysis, is critical to uncover true insights that can fuel a deep and more relevant connection with a consumer.


Agencies of the future are going to be more consultative and highly dependent on technology and data. The staffing structures will need to shift from relational to knowledge-based, and those that can find the right balance while maintaining the consumer in the middle will have a distinct advantage.

Tom Edwards is chief digital officer, agency at Epsilon.

On Twitter? Follow Edwards at @BlackFin360. Follow iMedia at @iMediaTweet.

"Retro television, old microphone from 50s and headphones" image via Shutterstock.

Brandt is an independent web analyst, researcher and academic.  As a web analyst, he specialises in building bespoke (or customised) web analytic reporting systems.  This can range from building a customised report format to creating an...

View full biography


to leave comments.