Its not just customers who worry about privacy, but marketers too. How much should you know about your customers, and what is appropriate to discuss? If only your browser could talk… well, it actually can. And we all have more information on our customers than they tend to think we do. So how much is enough, and when is it appropriate to use that data?
How much customer data is enough?
This is one question that is often ignored by big data advocates looking to collect as much information as possible. While more information may help you make better decisions, there needs to be a balance between data, activity, and meaning. We advocate modesty and honesty in your approach to your data mining process. Just because you can find out information about your customers with tracking codes and data appends doesn’t always mean that you should. Just look at the current news and breach issues, and you see that tracking “because you can” is not a good position to defend. So the data mining process needs to advance your knowledge of the customer and your product and be accountable to that. If it does not help your understanding, you should remove it from your view. Not only will it help you understand your customer better, it makes the job easier and establishes trust with your customers.
The “creepy” data problem
So you know your customers’ favorite time of day to visit your site – is it OK to mention that in your email? No. The “creepy” factor of the massive data mining efforts gives us a distorted perspective on what we can use to communicate with our customers. Marketers need just enough data to provide value and keep pace with their needs. Predictive analytics is something we talk about with our customers, and it seems to be very attractive to many businesses. It’s the 21st century version of the crystal ball that can predict what is about to happen. While it’s slightly more reliable than a crystal ball, it’s not a real determinate of human nature. So make sure you use your data to help communicate some value to your customer when they need it. A good rule of thumb should be something like this: if they visited it, clicked it, looked at it, or gave it to you, you can use it. If you found it by your own efforts, it should be concealed. This includes insights that maybe came from the user themselves through your research. If they did not engage you with that data, don’t engage them. For example, you know someone regularly opens your email at 10am every week. But for two weeks, you see them opening it at 8pm. Sending them an email with the subject line “Working late?” would send the hairs on the back of their necks straight up. They did not give you that data, you took it and used it for your purposes. That’s creepy. And moreover, what value did that deliver to your customer?
Data should always be used to learn how to serve your customers better, not to control or persuade them. No matter how you approach data, knowing something about your customer that does not serve them will only hinder the relationship, not grow it.