Fact: Data-driven communications break. Whether you’re a veteran data engineer, mid-level developer, or new to email marketing, we all have those WTF (What the Facts?!) moments. Problems, conundrums, and troubles are inevitable in data-driven marketing. Data stops syncing. Emails display wonky …ahem, Outlook. Dynamic components break. Personalization loads incorrectly, or not at all. Audience processes crash. Those are just a few things email marketing pros are faced with investigating and troubleshooting. While these issues are applicable to a variety of tools and systems, this blog series is primarily written for Salesforce Marketing Cloud users.
Before We Get to the Mad Sciencing, Meet Dr. Frankendata
Greetings and grand salutations!
I am the enigmatic Dr. Frankendata. Welcome to my laboratory, a blog series that will delve into the recesses of my knowledge and experience in data-driven CRM (customer relationship management) and digital marketing. In this series, I’ll share tips and tricks that have proven successful in my own data investigations. I hope the advice saves you time and, although failure can be a good teacher, I want to help you avoid learning the hard way.
Please forgive my excess personality. I’m unashamed to say that it’s a combination of IQ, ADHD and a penchant for theatrics. Keep things interesting. 🧐
Most people know me as Christopher Kayser, Sr. Solutions Architect at Response Labs. My coworkers bestowed the moniker of Dr. Frankendata on me in reference to the “mad sciencing” I do with data and SQL. I’ve been in the email marketing industry for nearly a decade, rising from entry-level email marketer to my current role. Although I’ve dabbled in dozens of other systems, my focus has been on Salesforce’s CRM, Marketing Cloud and suite of tools, and Oracle Responsys Campaign Management platform and Interact suite of tools.
3 Tips for Starting Your Data Investigations
Tip 1: Take initiative but communicate.
At the beginning of any data investigation, alert others and determine a timeline before asking for help. Whether it’s a co-worker, team member, your manager, or other relevant parties; if someone is impacted by the data failure, make sure you alert them.
Tip 2: Watch your time.
The amount of time you spend investigating should be determined by the criticality, urgency, and impact of the issue. Data problems can be time-sucking black holes. Starting a timer. My rule of thumb is not to spend more than two hours without checking in with colleagues. Taking initiative to solve problems yourself is a good thing (probably part of your job), but it’d be crazy to sink a ton of hours into the investigation if others might be able to help or lend insight.
Tip 3: Don’t panic.
Data investigations can be scary, frustrating, and complex. Like printers, email marketing platforms can smell fear. When a process or dynamic element breaks for no apparent reason, anger is a natural reaction. It’s okay to scream out, “this stupid thing should be working!” But don’t give in to the Dark Side. There’s no shame in taking a few minutes to simmer down and breathe. Then grab your energy-inducing beverage of choice and dial up the music. I recommend Rush’s Tom Sawyer, Glory Hammer’s Hootsforce, or Sabaton’s Winged Hussars. Also, find yourself a towel—to dab away the blood, sweat, and/or tears that are sure to flow. Then put your head down and start investigating!
I’ve just skimmed the surface of how to deal with data issues and how to investigate them. In Part 2 of this series I, that is Dr. Frankendata, will go deeper into some best practices. If you’re in the middle of a particularly vexing problem, here’s a teaser list to follow in the meantime:
- Follow the data backward, and then forward.
- Find the break. Where does the data stop behaving as it should?
- Try turning it off and back on.
- Consider breaking it into multiple steps.
- Break it yourself and rebuild it.
And that, folks, wraps of the first part of Dr. Frankendata’s mad science knowledge share for data investigations. Until next time, thank you for your time and for indulging me to the end of this post.