Analytics Horror Stories: Spine-tingling tales from Analytics & Reporting Experts


Tweet about this on TwitterShare on FacebookShare on LinkedIn


It’s that time again — time to get our spook on and recap some of the scariest big data and analytics nightmares we’ve experienced over the years. Join Sweetspot and our partners at MaassMedia in this thrilling and spine-tingling post, originally shared on MaassMedia’s Blog, as we reveal the stories that really keep us up at night:

Tagging Terrors

Improper planning around site production and inconsistency around the use of tag management can lead to some petrifying data analytics scares.

Let’s say you’ve meticulously planned a halloween costume for several months. You sewed the costume yourself, bought accessories, and tried out your makeup multiple times. You leave the house for the night of a Halloween party, thinking you’re prepared to win the best costume award. But when you take a quick look in the car mirror after arriving, you realize you’re still in work clothes, despite hours of planning.

We’ve experienced a similar scare when pushing analytics tags from staging sites to production.  Often, companies don’t have enough time and resources to make sure their staging sites match their production sites. Ultimately, what this means is that any tagging work done on the staging site may not function properly once it’s live. Ultimately, this will leave you collecting the wrong data, or no data at all. It’s frightening, panic-inducing, and downright scary.

Another halloween-alytics scenario to avoid is changing your implementation without thinking about how it will affect your reporting. This can result in “Frankenstein” reports.  For example,

if you change your page names, i.e. from “home” to “home page,” you’ll have to stitch together behavioral data manually from different locations within your analytics tool box in order to get cohesive reports. While at analysis level this task can turn into a monster, it’s often overcome at strategic level through the use of a reporting solution that lets you aggregate data from different sources and undertake cross-source deduplication at dimension level.

Finally, beware of phantom buttons. Phantom buttons are buttons installed on your site on the fly, without any IDs. These buttons can be hard to find, tack, and tag. Without an ID, a developer can spend hours looking through code to find the right button. If you’d like to avoid having to exorcise your buttons, always use tracking IDs.

Many of these scenarios can be avoided by creating a Solution Design Reference (SDR). An SDR can vanquish the analytics vampires from sucking the lifeblood from your reports. An SDR includes all technical requirements for the development team and any other important stakeholders, as well as information on what should be tracked and how — and will help you ensure consistency with your button IDs and naming conventions.

The Metric Graveyard

Or more bluntly put, where good metrics go to die. Many moons ago, in a less than cheerful meeting, we met a desperate and exhausted marketing manager. Her team had already given up on even the notion of reporting after receiving page upon page of disconnected marketing reports, spread throughout various Word and Excel files.

Disorganized? That’s an understatement. Useless? Even more so.

These reports were painstakingly built by hand with absurd amounts of disconnected data, none of which was directed towards any particular role, and without a single insight…

Scary? We’d certainly say so. But worse still, not a single report was ever taken seriously.

Marketers, don’t distress. There is light at the end of this dark and scary tunnel. In order to avoid sending your reports straight to the metrics graveyard, follow these simple rules:

  1. Report on relevant metrics by industry and role
  2. Customize your KPIs to the objectives of your organization
  3. Streamline data integration and delivery
  4. Develop consistent reports
  5. Always accompany data with insights

Nightmare on Analyst Street

Imagine living in a constant state of paranoia. As an analyst, you’ve probably heard about the Automated Reporting Monster. This monster has befriended your colleagues, whispering in their ears that it can do your job more efficiently than you. Some analysts find themselves within this horrific nightmare.

What does this mean for the analyst? Do their colleagues devalue their work? Could they even be replaced?

In an attempt to maintain control of their elusive expertise, analysts may guard ownership of the project and limit access to retain dependency. They may even opt for advanced analysis tools where their unique set of skills is required for operation and maintenance. Whatever path they choose, these worried analysts often trap themselves behind manual processes. Instead they could be using automated dashboard solutions to elevate their influence by delivering valuable insights and measurably impacting company outcomes.

Don’t despair analysts, this nightmare has no chance of becoming reality for one specific reason — you possess a skill set unlike any other: the ability to analyze. Your expertise in finding correlations and patterns and ingenious ability to dig deeper to provide data-backed conclusions are all invaluable. While automated reporting solutions deliver information around what has been happening to your stakeholders, they also provide you with the opportunity to show your expertise and exponentially increase the value of these reports. Adding conclusions from your analysis, thoughtful recommendations, and recording insights on how performance has been impacted by past decisions will propel you professionally.

Data-driven dread

As we’ve seen numerous times, fear is a paralyzing state of mind that can overwhelm even the bravest souls. Even in organizations that have seemingly got reporting down pat, that make data widely available, and customize delivery by role, executives are still weary…

What could be keeping these wise business professionals from using their data to really make a positive impact on their marketing performance? Decision paralysis is often the side effect of a larger root cause, be it strict company culture that does not encourage learning, or lack of clear direction.

If, for example, executives fear negative fallout from a sub-optimal decision, then they’ll obviously be reluctant to take any sort of action. Where culture, however, encourages testing and optimization to better performance, organizations are more likely to have executives willing to act on data-based insights.

Likewise, those organizations that have yet to define their true business objectives are less likely to foster risk-takers as they still lacking a clear picture of what they should be working towards. You cannot expect someone to act if they don’t have a real target.

So, in order to help stakeholders get over their fear of making data-backed decisions, it’s essential to empower them at an organizational level to feel comfortable taking risks and learning from their actions. Not only that but, they should be encouraged to align their metrics strategically with business objectives in order to allow them to visualize how their performance and decisions impact the company as a whole.


Would you like to hear more spooky stories from the analytics and reporting experts at Sweetspot & MaassMedia? Join us on Halloween, following the eMetrics NYC conference, for happy hour at the Houndstooth Pub in Hell’s Kitchen. Reserve your spot now!

Tweet about this on TwitterShare on FacebookShare on LinkedIn

Holly McKendry

Sweetspot Marketing Director. Wakeboarder & travel enthusiast. Communication Studies graduate of Texas State University, San Marcos.

Add a comment

Try Sweetspot today!

Not Another Dashboard.