Updates from Peder RSS

Peder 2:06 pm on Jun 20, 2011 Permalink | Reply
Tags: hosted service, privacy,   

On May 26, 2011, minutes before a midnight deadline, President Barack Obama extended the USA PATRIOT Act (Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism Act) of 2001 which gives authorities of post 9/11 powers to search records and conduct roving wiretaps in pursuit of terrorists.

Should we care?

In an effort to protect the American people, the Act allows for unprecedented access to personal information stored on computer systems and in other formats. Perhaps the most controversial section of the Act is section 215 which is commonly referred to as the “library records” provision because of the wide range of personal material that can be investigated. It allows the US government to secretly request and obtain records for large numbers of individuals without any reason to believe they are involved in illegal activity. Items that can be searched include “books, records, papers, documents, and other items,” which includes dumps from private-sector computer databases.

This is likely a concern to many organizations wanting to protect the privacy of its customers and users. And for companies using web analytics tools, the collection and storage of that data in the US provides government authorities unfettered access to that information. Furthermore, the US government can access considerably more information than any web analytics tool would be capable of. For example, the US government can cross-reference the IP address assignment records for any given IP address for any given time. This allows them to access name, address, phone number, account numbers, etc. Not even the best web analytics software has access to this depth of information. There may not be much US based organizations can do to prevent such access, but those located in Canada and abroad can choose not to use services which store collected information in the US.

Web analytics tools such as Adobe SiteCatalyst, Google Analytics and Webtrends On Demand all store visitor information on US servers. As an alternative, Webtrends offers software on premise to ensure collection occurs within the organization and within its own country borders. Since hosted services free staff from implementing hardware solutions to support installed software, they continue to be popular options for many. For clients who don’t want their data stored in the US but also don’t want to deal with software installations, Unilytics offers managed services, whereby we install and manage the Webtrends implementation. This is a very attractive solution for organizations concerned with the location of collected data, but who are also attracted to a hosted solution.

The number of court approvals for business record access jumped from 21 in 2009 to 96 in 2010. While that remains a relatively low number, it poses an uncertainty that many organizations are unwilling to accept. It’s that uncertainty that causes the issue.

It may be that the greatest risk to organizations allowing data to be stored in the US is to their own reputation. The threat to privacy may be more perceived than actual, but that may be reason enough for organizations to not want their data stored in the US.


Peder 7:05 pm on May 25, 2010 Permalink | Reply
Tags:   

I am regularly frustrated with the desire by many clients to track statistics such as the number of visitors to a site or the volume of page views consumed. It’s time to break free from all that nonsense (and noise)! Why would anyone care about the top 10 pages report? Or for that matter, the top 100 or top 1000! They deliver no value to the question: Is you web site working and doing its job?

So, why do we have a web site? As I have stated earlier, it is for one of two purposes; to either make money or save money. You are either in the business of promoting something or encouraging visitors to find what they are looking for instead of more costly options such as calling your business for support.

If we can agree that visitors are brought to your site for a specific purpose then learning if they are accomplishing those specific tasks are indications of how the site is performing.

Measures, which illustrate visitor behavior and traffic volumes, are not indicators of web site success. Increasing hits, page views, and visitors may show that marketing campaigns are driving more visitors to your site, but not whether it’s performing well or not.

Instead, focusing on conversions is the key to success. Conversions are much more than tracking successful shopping cart check-outs. Conversions extend to non-commerce activities such as:

  • Filling in a form
  • Downloading a document
  • Signing up for an event
  • Registering for upcoming seminars or conferences
  • Renewing membership or paying membership dues
  • Signing up for continuing education classes
  • Consuming key content (measured by how long visitors spend on a specific page)

Once we begin tracking the activity visitors are doing relative to what are objectives are for the site, we can learn how our site is performing and what changes need to be made.


Web analytics should begin with a plan. Unfortunately, most don’t. And the plan should start by answering the most basic of questions; why do we have a web site?

All web sites can be categorized as having one of two objectives; either to make money or to save money. Sites which make money are not only those in which shopping carts are presented. Rather, any site involved in the promotion of products or services have as their underlying purpose to increase brand awareness and encourage purchases on or offline.

Sites which attempt to make money include:

  • lead generation
  • brand awareness
  • eCommerce
  • Social networking
  • Entertainment

Conversely, sites which attempt to save money are those which encourage self-service. It decreases corporate costs if web visitors can successfully download a document, register for assistance or fill in an online form rather than contacting a toll-free number and asking for the same to be done or mailed out.

Sites which attempt to save money include:

  • >Customer education
  • Self-service
  • Customer service
  • Informational
  • Intranet

Understanding the objective of your site allows you to construct and establish key performance indicators (KPI) you need to track. Studying raw web analytics reports is meaningless.

Measures, which illustrate visitor behaviour and traffic volumes, are not indicators of web site success. Reports showing page views, most popular pages and visitor traffic do not indicate whether a site is performing well or not. No organization, with the possible exception of media firms, should fundamentally care if page views and visitor numbers increase. To extract real meaning from increases in raw numbers like page views or visitors it must be put into some context. That’s where KPIs come in.


Peder 10:12 pm on Dec 1, 2009 Permalink | Reply
Tags: SDC,   

More and more organizations are implementing an alternative technique to collect web site traffic information rather than relying on web server log files. This technique is called client-side data collection, or data tagging for short. Webtrends’ implementation is referred to the SDC (SmartSource Data Collector). Data tagging solves many problems associated with web server log file analysis.

Implementing data tagging requires some development work to ensure that data tags are inserted and maintained on web pages.

With data tagging, web traffic data is more accurate because traffic normally hidden by cache or proxy servers is tracked. IT administration is eased because data collection is centralized in one location versus site data being dispersed among several log files from multiple web servers that may also be geographically dispersed. And web data can be collected from specialized applications, such as application servers and browser applications (e.g. Macromedia Flash).

In short the SDC has the following advantages:

  • Generates logs that are optimized for Webtrends that are as much as 90% smaller than traditional access logs.
  • Produces a single centralized SmartSource file rather than separate log files for each web server. This essentially eliminates the administrative headaches associated with gathering logs from multiple, geographically dispersed web servers. The SmartSource file can even contain hits from multiple domains (the domain name can also be passed as a query parameter), allowing visitor behavior to be analyzed across an organization’s sites or even partner sites provided they permit your tags to be included on their pages.
  • Provides information that is difficult or impossible to obtain with log files. For example, data tags linked to your SDC can be included in your banner ads placed on other sites. SmartSource tags can also be inserted into Flash applications, permitting a hit to be entered into the SmartSource file for each event fired in the program. This means visitor activity within Flash applications can be analyzed just like visitor interactions with HTML-based pages.
  • Web traffic data is more accurate because traffic normally hidden by cache or proxy servers is tracked. In many cases, web server log files do not accurately represent the actual interactions visitors have with a web site. Proxy servers are one of several examples of how analysis results can be distorted by web server log file data collection. Proxy servers deflect page views from web servers by caching the most frequently requested pages. Local caches have a similar effect, handling browser requests through locally cached pages rather than making repeated requests to the web server. In doing so, these page views are not recorded in the web server log files.
  • Creates a cookie for more accurate reporting. Cookies ensure visitors are tracked as they navigate and return (if using a persistent cookie) to your site. This enables the most sophisticated features of Webtrends such as Scenario Analysis, SmartView, Path Analysis, and Custom Reporting.
  • Acts as a filter: you only tag the pages you need reporting on.
  • Bots and Spiders don’t need to be filtered out and / or scrubbed from the logs resulting in more accurate reporting and requiring less CPU processing power.
  • Enhancing reporting capabilities through META-tagging (i.e. tracking revenue, correlating Paid Searched Terms with conversions, …)
  • Reporting on specific events within a page through the use of the dcsMultiTrack JavaScript function:
    • Tracking PDF downloads
    • Tracking dynamic / Web 2.0 events on pages (i.e. DHTML or any browser-supported event)
    • Tracking events within Flash movies

Client-side data collection is quickly growing in popularity as the superior approach to collecting web visitor behavior information. It provides greater reporting accuracy and lower administrative overhead. Organizations should carefully analyze the costs and benefits of data tagging versus web server log file analysis, and determine which method will best meet your insight needs.


Peder 1:10 pm on Oct 9, 2009 Permalink | Reply
Tags: customer, engagement, model   

Customer Engagement Model

There are many factors and activities which determine the success of online properties. Understanding how these actions affect the overall message and how they interrelate is of significant importance.

Mass media marketing such as television or radio spots may reach a wider audience but may have little or no interest to many. How many of you fast-forward through commercials whenever possible? Conversely, other activities such as social media may reach fewer people but the direct dialog with customers can mean a stronger bond and impactful message. Cross-channel analytics allows us to investigate what the interactions between these different activities and the influence one activity has on another.

To better understand the effect marketing and online activities have on the success of an organization and their web site, Unilytics has created the CEM (Customer Engagement Model). It illustrates the relative worth of different marketing activities with respect to Customer Reach and Customer Value. “Customer Reach” refers to the relative number of people our message will get through to. “Customer Value” assigns a rough measure of how personal the message is to each individual recipient when it reaches them.

Customer Engagement Model

CEM is divided into three activity sectors. In the first sector we unilaterally ACT with no real feedback from our customers. This one sided communication includes mass marketing activities such as television, newspaper advertizing and radio spots. These may reach a very large audience, but will likely not resonate with the majority of those who hear or see the message. Email marketing can also reach many and we see it as being more impactful to individuals as we can tailor the message based on previous results. That is also why it overlaps with the next sector.

In the second sector, we have the ability to REACT to our customers and make changes to our communication and marketing message. We’re beginning to have a two-way conversation, but we can only ‘hear’ from large groups of people. There are several marketing initiatives which fall into this sector including web analytics, 1:1 marketing, A/B testing, voice of customer and search marketing including SEO and display advertising. They all allow us to react to customer feedback and change our messaging to better serve customers. Because we are able to gather feedback and improve our offer, these activities tend to be more impactful and result in improved customer value, while these activities tend to reach fewer people. Again, some overlap into the third sector.

In the third sector, we INTERACT with our customers and we market products and services in direct response to customers concerns and needs. We now have a one-to-one conversation with our customers and their involvement is greatest in this sector as is the value of the communication. We fully interact with customers by using such social media tools as Twitter, forums, and various others; we react to their concerns and correspond directly with them. However, unless the dialog becomes viral, the number of people we interact with tend to be very few. By using CRM solutions, we can establish an ongoing dialog with customers and create an ongoing relationship, placing this activity also in this sector.

Some may challenge the placement of the various activities. Furthermore, this is not meant to be a comprehensive list of activities. The point is that each activity has its value and they affect company-wide marketing success differently. Each contributes to the overall success of the message. No single event or activity is positioned in the top right corner of the model where we ultimately strive to be. Our goal is to reach everyone with a comprehensive, interactive dialog so we can serve them fully to their specific needs and desires. Such a tool does not currently exist. But with a combination of marketing activities we can get there as a whole. Reaching that goal requires an integrated effort so we understand how one activity affects another. Cross-channel analytics is crucial in this endeavor.

Page 1 of 212