Should You Personalize Your Paid Media Content?

By Kirill Gil, Persado Global Solutions Consultant for Paid Media


Personalization is all the rage these days – you hear about it at every marketing conference and read about it in every marketing trade publication and blog, including this one. But as someone who has been working on direct response campaigns for almost a decade, I have come to think of personalization a little differently. Will a countdown timer on a display ad increase conversions? Does using a local landmark as the image on a Facebook ad increase the number of calls a company gets on an offer?
 
The fact is, I have not seen many companies question whether or not a personalized element will improve a campaign’s performance. They just know it’s best practice and, in today’s competitive landscape, try to hyper-personalize down to the individual in an effort to stand out. But this can lead to hyper segments that have such low response rates that it becomes impossible to test and implement new, more relevant messages. So, how personal is too personal in the social media landscape? These four gut checks will help you decide.   


Test the Effect of Personalization on a Macro Level Before Implementing It


Some brands, such as chain restaurants and stores, may have locations all over the country — and, in many cases, the globe. But that doesn’t mean these brands have to create ads with location-specific creative for each spot. When brands do this, it’s hard to tell if all the resources needed to generate hundreds of ads are worth it because it becomes so granular. This, in turn, kills the whole purpose of personalization since it renders a brand unable to tell whether inclusion of the individual’s elite status, qualifying bonus offer and call out of the limited time deal are actually better than a generic message in driving conversions. As a result, everyone is busy updating the hundreds of personalized messages, but the company has no quantifiable data to understand the associated ROI.

Instead, start small by testing 10 cities with specific creative to see if it moves the needle before creating an ad for each city. If it works, go for it, but if it doesn’t, don’t dry up creative resources by designing hundreds of ads. 

Look for Micro-Level Outliers to See If Groups Should Be Separated 


Sometimes, brands may need to segment further. Say a store is offering a $100 discount on purchases of $200 or more. Audiences in Boston, New York and Washington D.C. are converting like crazy, but people in Seattle aren’t biting. Perhaps they didn’t want the discount. Try running additional tests, such as adding other west coast cities to see how they perform against east coast cities. Maybe it’s a regional thing, or maybe it’s just Seattle. If Seattle is truly an outlier, it doesn’t mean the store shouldn’t offer discounts to people in the area ever again, but it’s certainly something to keep an eye on. 

Keep Testing. Even When Hyper-Segmentation Is Necessary.


There are certain businesses that need to create multiple segments. What do you think a baseball fan in New York will respond better to: An ad with New York Yankees star Aaron Judge or one with David Price of the Red Sox? Probably the former, but I still encourage testing. Perhaps a city with a team that isn’t doing too hot would prefer to see the Yankees’ breakout star than anyone on their team in an ad. Or, switching to football, maybe Philadelphia Eagles fans respond better to football-centric ads reminding them they’re Super Bowl champions, while Browns fans respond better to deals on free hot dogs (because they likely aren’t going to the games for the football).

Do a Sanity Check

Finally, take a step back and ask yourself: Is this going to scare people or inspire them to act? A realtor client I worked with at a previous job wanted people to see images of their actual houses in a display ad with a value attached to it. Maybe that would make them work with him to sell! Or, maybe the targets would be incredibly creeped out. It’s never a good idea to get too personal with someone on the Internet — especially when it comes to randomly calling out their addresses or family members. 

Bottom Line: Personalization is an incredibly useful tool, but there’s a such thing as going too far. Start small — it will help your sanity and that of your customers.