2014-01-September-Creepy-Bob.jpgThere's a fine line between effective, personalized marketing and an eerie feeling that “big brother” is watching. Unfortunately, being creepy is one of the most common perils of data-driven marketing. In an online environment, retailers can observe every single nuance of an individual’s shopping experience.

On the one hand, there are obvious perks for doing so, including the fact that carefully tailored marketing experience can lead to cross-sell, upsell and retention opportunities. Personalization simply works -- and in study after study, consumers say that they like it.

But marketers need to be careful. Because when personalization goes too far, it stops being helpful and just gets, well ... creepy.

A sample scenario: you’re at a grocery store. A sales associate notices you spending a lot of time in the specialty food section, looking at gluten-free options. You go to check out and put flour-based cookies on the counter. The associate says, “I noticed you looking at our gluten-free display earlier. I just wanted to let you know that this brand has gluten.”

Helpful, right? Definitely -- especially in a more extreme case like this one where some personalization could help you avoid an allergic reaction.

But what if the associate says something very different … just as observant, but not nearly as appropriate? For example, what if she happens to overhear you talking to your wife on your mobile phone, then suggests that you bring home flowers for your date night too?

Not nearly as helpful. And a good deal more intrusive.

Of course, while it is certainly creepy when these types of situations happen in a brick-and-mortar setting, it’s far more common in the world of digital marketing. Companies store mounds of data that track customers’ personal preferences and have access to much more intimate details.

The good news for digital marketers? There are simple steps to help organizations achieve the right level of personalization, without stalking their customer like a predator. Take a look:

1. Put data architecture in place that supports customer insight

Do you have a history of every interaction you’ve had with every customer? If you don’t, you are missing out on incredibly useful information. If you don't keep data history, you won't know what’s been said to that customer before, or how they responded. This leads to repetitive marketing, annoying ads and your brand eventually becoming a nuisance instead of a desirable vendor.

The solution is simple -- store the data! And analyze your customer’s preferences so that you can target them properly and in a way that will make them respond positively. 

2. Develop a governance infrastructure predicated upon awareness and restraint

Do you have a way to manage communication frequency across all customer interaction channels? It is important to take note of not only which channel you are using to target an individual, but also how often. If not, your messaging may become a burden.

Therefore, set up a system of checks and balances. Put in place an intuitive governance infrastructure that tracks this frequency, and depending upon the awareness and interest level of the individual, the infrastructure is capable of filtering out unwarranted communications.

3. Observe how your customers respond 

If you’re really listening, your customers will tell you exactly how your organization’s marketing is performing. The trick is having a tool to help consolidate online insight with offline interactions. Without this functionality, you run the risk of seeing only half of the picture.

In general, customers are aware that companies are tracking their behaviors -- and are okay with that reality when it improves their shopping experience. Still, they don’t want to be reminded of it every time they open their browser.

But with a few simple steps, you can provide a customized consumer experience for every individual --staying between the lines of helpful and intuitive -- without over stepping into the creep zone.

Title image by fineartlady (Flickr) via a CC BY-NC-SA 2.0 license