Four Basics of Tag Deployment

September 29, 2014 Brad Perry

We recently conducted a series of audits on Internet Retailer Top 500 and Second 500 web sites. This research uncovered a number of interesting findings that we have decided to discuss in several blog posts. Although we found obvious differentiators of top sites, there were also issues common to all sites.

There are some basic elements of data quality management, and these continue to be a challenge for every company. Management of these fundamentals builds a solid foundation of data quality in digital marketing and improves the ROI of the entire digital marketing technology stack.

In order for Javascript tag-dependent tools, such as web analytics, to collect and report the most accurate data on each page, four basic conditions must be met. If these four conditions are not met, data becomes corrupted, inflated, inconsistent, and overall unreliable.

  1. Each page on the web site must render one (and only one) instance of a given web analytics tag. When tags are missing, data is missing. Likewise, multiple renderings will inflate data. On average, 2.8 % of pages are missing a web analytics tag. 95% of the time, these tags are one click from the home page.
  2. Each tag must be set up to send data to the correct reporting account, sometimes referred to in vendor-specific language such as UA or Report Suite ID. If the data isn’t sent to the right place you won’t get the information that you need. Also, a related privacy risk is introduced when deprecated account IDs are not removed from the deployment. On average, there are 2.5 reporting accounts set up for each site, but 8 sites had more than 10 accounts.
  3. Variable parameters and values must be configured correctly. Particulars of variable configuration vary by vendor, but generally the page name, namespace, and custom variable data must be configured consistently across the site. This includes naming conventions and is detail driven. Capitalization, spaces, and punctuation matters.68 sites are using more than 100 unique variable paramaters across their deployment – how is this being managed?
  4. Tags must execute without error and with consistent timing across all browsers, network connection types, devices, and geographic locations. This means that tags must not be affected by javascript errors or network problems; that they must have similar placement across the site and relative to other tags; that they are programmed correctly in the TMS; that they are using synchronous / asynchronous appropriately, etc. This is easier said than done and may prove to be a challenge for your organization when it comes to the time and resources you have to allocate. Over 1/3 of top retail web sites have javascript errors on greater than 50% of pages.

Data quality management is a complex topic, and tags can be deployed many ways. But if these four bases are covered, you’re building a solid data quality foundation that will help your website provide accurate information your company needs.

ObservePoint audit helps companies measure and improve digital marketing data quality. Take a free test-drive with a mini-audit.

About the Author

Brad Perry

Brad Perry has been Director of Demand Generation at ObservePoint since June 2015. He is, in his own words, “unhealthily addicted to driving marketing success” and has demonstrated his unrelenting passion for marketing in various verticals. His areas of expertise include demand generation, marketing operations & system design, marketing automation, email campaign management, content strategy, multi-stage lead nurturing and website optimization.

Previous Article
Defining Governance for Data Quality
Defining Governance for Data Quality

This article discusses the importance of governance for data quality, and the need for tag management. Also...

Next Article
10 Steps to Digital Data Quality Nirvana
10 Steps to Digital Data Quality Nirvana

This article outlines how to manage data quality, and the ultimate cost of poor data. See the ten perscribe...

Get a free 14-day trial with ObservePoint

Start Your Trial