In today’s episode, we’re tackling a truth that can be unsettling for many marketers: sometimes, your data is just plain wrong. You’ll learn why discrepancies in your analytics are more common than you might think and why that’s not necessarily a reason to panic. I’ll explain the difference between “reliably wrong” and “unreliably wrong” data and how understanding this concept is crucial for making informed marketing decisions. Tune in to discover how to identify the quality of your data and gain valuable insights, even when it’s not perfectly accurate.
Can’t see anything? Watch it on YouTube here.
Listen to the audio here:
- Take my new Generative AI course!
- Got a question for You Ask, I’ll Answer? Submit it here!
- Subscribe to my weekly newsletter for more useful marketing tips.
- Subscribe to Inbox Insights, the Trust Insights newsletter for weekly fresh takes and data.
- Find older episodes of You Ask, I Answer on my YouTube channel.
- Need help with your company’s data and analytics? Let me know!
- Join my free Slack group for marketers interested in analytics!
Machine-Generated Transcript
What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for watching the video.
In today’s episode, let’s talk about wrong data—data that is wrong, incorrect. I was having a discussion last week with some colleagues about web analytics data being wrong. Did you know what was showing up in the, in the web analytics did not at all match what was in the CRM or the marketing automation system or this or that?
And all these systems have different data, there’s always a variance. It’s very rare to see a company that does not have variances in the data. Why is that? Well, there’s a bunch of different reasons. For example, ad blockers that can, that can prohibit a system from recording data. There’s VPNs, there’s also just secure browsing environments—all the different ways that people can obscure the data.
When I worked at the old PR agency I used to work at, we had one client that was in the cybersecurity industry. They had almost no web analytics data. Why? Because their entire audience was cybersecurity professionals that basically had fortresses around their computers. So, you wouldn’t, you could collect nothing, you couldn’t collect absolutely nothing. Their big, their web analytics was just one big question mark.
But that got to the discussion of, I think, a critical point about data: It’s okay for data to be wrong, particularly numeric data. It’s okay for data to be wrong, if it’s reliably wrong. What I mean by that: if what’s in your web analytics system, say, is 20% off from what’s in your CRM, but it’s always 20% off in the same direction, then directionally you can still use web analytics to forecast. You just have to adjust it for 20%.
Because, whatever reason, what’s in your CRM will always be 20% higher than what’s new web analytics, that is what we call reliably wrong data. And reliably wrong data is a good thing. It’s kind of like a car where, it just pulls to one side. And, the moment you get behind the wheel, you know just to put a little extra pressure on the wheel and drive the car that way and we’ll go straight. It is reliably wrong. That’s fine.
Where we run into trouble is when things are unreliably wrong, where the variance is different day to day. So, maybe your web analytics is off 20% today, but tomorrow is off 2%, and the next day is all 5%. The next day, it’s exactly on target. When that happens, when it’s unreliably wrong, you can’t forecast, you can’t map trends, you can’t make decisions from it because you don’t actually know what the value is.
So, when you’re diagnosing your data—either by yourself with the help of AI—you’ve got to first answer the question: Is the data wrong? By how much? And is it reliably wrong? If you can identify that your data is reliably wrong, then you’re, and you can, you can determine what the exact variance is, you’re off to the races, you’re good to go.
If your data is unreliably wrong, no amount of modeling or forecasting or anything is going to fix that problem, because it’s fundamentally broken. So, keep that in mind as you’re starting to work with your data. Reliably wrong data is okay. Unreliably wrong data is useless.
That’s going to do it for today’s episode. Thanks for tuning in. We’ll talk to you next time. If you enjoyed this video, please hit the like button. Subscribe to my channel if you haven’t already, and if you want to know when new videos are available, hit the bell button to be notified as soon as new content is live.
♪ ♪
You might also enjoy:
- Mind Readings: Most Analytics Data is Wasted
- Fireside Chat: Geraldine Deruiter on Food, Feminism, and Fury
- Almost Timely News: Principles-Based Prompt Engineering (2024-02-25)
- Almost Timely News, February 11, 2024: How To Evaluate a Generative AI System
- Mind Readings: Hacking Social Media Algorithms
Want to read more like this from Christopher Penn? Get updates here:
 Take my Generative AI for Marketers course! |
Leave a Reply