Set Proactive Alerts

October 20, 2014 Brad Perry

header image of an old-fashioned, red telephone sitting on a table

Sometimes things break for [seemingly] no reason. Given the complexity of most site and application deployments in the age of mashups, social media, tag proliferation, big data, mobile, and quickly evolving standards, errors and breakage shouldn’t be so shocking.

Unfortunately, the breakage of web measurement code is harmful to business processes and early detection of these events is challenging. Unless bad data appears in your analytics platform at a high enough volume to be noticeable or it’s caught in a QA process, the proverbial smoke can go unseen until the forest fire is raging. That is, unless you’ve configured a proactive alerting system.

What is a proactive alerting system?

The mechanism for proactive alerting to web tracking is simple, although some forethought and design specific to each situation is appropriate. But the result is exactly what it sounds like – a continuous background testing process that verifies the deployment and function of tags. And when a test fails, you’re notified immediately.

An important detail to note here is that this process actually verifies the functionality of tags. This is important because tagging script can still exist on a page but either executive incorrectly (variables populated incorrectly) or fail to execute (technical error caused by javascript, network issue or something else). This test should be robust enough to detect changes in the technical environment as well as the marketing tool setup.

In their white paper, Web Analytics Demystified recommends that on certain pages, you’d be notified whenever a new tag is added, a tag is removed or absent, or when pre-determined variable is not populating correctly. We agree with this approach. Start by identifying high-value content. What content is most important to driving and measuring conversion? Where are conversion metrics most strongly influenced? Where is the most money being made and spent on the site? What is most visible to management stakeholders? Start here.

Proactive, Powerful and Practical

Most of our customers receive their proactive alerts through ObservePoint Simulations. Simulations test specific content for specific measurement tools at a specific interval. For example, set up an alert on:

  • A campaign landing page for search marketing tags with campaign parameters four times per day.
  • A video for the Adobe Analytics tags that fire at the start, 25%, 50%, 75%, and finish, once per day.
  • A three-step form, where depending on how a user completes the form, each step changes the experience. Variables are populated according to the experience chosen by the user in the previous steps. Confirm the functionality of each test in a pre-specified scenario every hour.

 

This type of proactive alerting is quite powerful and becomes a key part of measuring and improving data quality in the enterprise. Think you can use a data quality tune-up? Request a complementary audit.


This post is based on the white paper Data Quality and the Digital World by Eric T. Peterson, Principal consultant at Web Analytics Demystified.

 

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.

LinkedIn More Content by Brad Perry
Previous Article
Rule Your Tag Validation [Recap]
Rule Your Tag Validation [Recap]

This article discusses the importance of using rules to automate data governance, as explained at Validate ...

Next Item
ObservePoint Rules Engine - Build and Automate Rules Systematically
ObservePoint Rules Engine - Build and Automate Rules Systematically

Use ObservePoint's Rules to test your analytics and marketing implementation against the expected values of...