Personalization: The thin line between being personable and being creepy

Erich Gerber, SVP EMEA & APJ, TIBCO Software, takes a closer look at the balance that organizations need to strike in using customers’ data to create better personalization.
Erich Gerber, SVP EMEA & APJ, TIBCO Software, takes a closer look at the balance that organizations need to strike in using customers’ data to create better personalization.

Remarkable advances in analytics, data accessibility, and new data sources are making personalization even more of a priority for marketers. Whether that’s an app, online or an in-person offer, companies are desperate to create human interactive experiences that delight buyers. Getting personal is a tricky matter of balance though. It’s nice if the restaurant manager remembers my choice of cocktail and serves it to me on the house. If he knows my birthday, that may be nice too, but he has crossed a line if he knows the year, birthplace and my parents’ names. This is the challenge we face in personalization and individualization: we need to get close to understanding the in-depth needs and desires of customers and prospects but not so much that our behaviour is deemed creepy.

Every smart company wants to strike that balance and a leading analyst firm suggests that nine in 10 companies already have personalization strategies. But nobody said it was easy and while overwhelmingly the data suggests that getting it right will lead to loyalty and the chance to upsell across channels, getting it wrong will send customers into the arms of rivals. Instapage found that 74% of customers feel frustrated when website content isn’t personalized. There may also be a demographic element in play: SmarterHQ found that 70% of millennials are frustrated by irrelevant emails they are sent.

Many of us talk a good game about personalization and it’s already a conference meme, often in company with talk about CX. The dirty truth though is that we’re really only taking baby steps: in 2019, McKinsey found that just 15% of CMOs believe their companies are making the right moves to building a successful personalization strategy. But they need to keep their eyes on the glittering prize because the wins when they come are big. McKinsey said personalization can halve customer acquisition costs, lift revenues up to 15% and increase marketing spend efficiency by up to 30%. We can confidently assume those numbers, based on 2016 research, will have since headed north.

The path to maintaining a happy balance: keep it relevant 

Personalization is a blink-and-you’ll-miss-it affair. People decide if they’re interested in a marketing offer in a matter of seconds… or even less. Get my name wrong or send me something I couldn’t care less about and you’re toast. 

Relevance is of the essence and we need more than tools to get us there. Marketing stacks abound and companies groan under the weight of shelfware but we need more than tech: processes need to be tuned too… and that means getting teams of smart people thinking. 

Perhaps they get too bound up on their missions to create the iconic 1:1 personalization at scale relationships, but they miss a lot if they don’t understand core demographics first. We’re all individuals but we also share a lot: age, sex, parenthood all shape us, whether we like it or not. Combine that with information we leave as breadcrumb trails behind us (streaming analytics covering website visits, purchases, service histories, etcetera). Companies can identify us and send us notifications, offers, stock updates, and delivery tracking. These are things we appreciate without feeling that the seller has crossed a line. 

If we serve offers that are in areas of known interest to an individual or a business we are helping them. The Spanish fashion retailer Desigual achieves this by recommending apparel items based on customer history and stated preferences, all stored in a central system, together with stock inventory and other relevant information. Similarly, If I am a loyal customer of a shirtmaker then an offer to buy three for the price of two may well appeal three years after my last purchase. Or, if I regularly purchase bulky items such as nappies or printer paper in-store then the offer of free delivery could be well received. The seller has persuaded me to cross channels, deepening my relationship with them while I have gained from newfound convenience. 

Next-product-to-buy triggers are also highly effective when they are truly relevant and refined over time for proven efficacy. Often though they require the human touch rather than a keyword finder in an algorithm: if I buy a shirt then I may well want to buy cufflinks or a necktie but a suit or some other item of clothing tagged under apparel? Less likely. Similarly, a well-written social media post that fits my reading history may grab my attention more than an email. Even in the 21st century, despite what Marc Andreessen has said, software has not eaten all of the world. 

The creep factor

If we instead use a scattergun approach and play ‘algorithmic hit and hope’ we will quickly and correctly be seen as tiresome spammers. Transparency is key. Only a minority of us understand the algorithmic and probabilistic minutiae of what is happening when we are the subject of sophisticated targeted marketing. We might not even understand if somebody tells us all the details but sharing high-level information about sources of knowledge would be a welcome start to building trust.

Companies need to focus on delighting their audiences. An augmented reality environment can help us try sporting equipment before we buy. A healthcare provider or health insurer might provide us with free wearables to monitor our pulse, heart or blood pressure. That may be an attractive offer, but first we need to know how that data is going to be used.

If scattergun marketing can leave a nasty taste in the mouth then overly targeted marketing can leave buyers feeling stalked and harassed. Being alerted to a deal when in store is fine but a funeral offer after a loved one has died is very much not. Sometimes it’s a matter of timing: a pop-up ad relating to a purchase I made a minute ago is too much too soon. Other known turn-offs include an inability to opt out, being targeted for too long and too often. False positives are also a live issue: if I buy perfume, it’s probably going to be a gift, not for me.

2019 survey by MIT Sloan Management Review found wide swings in sensitivity to personal data use with over-65s. The well-educated is much more likely to be concerned than under-24s and the least educated. The magazine urged companies to understand their specific audiences and to educate on the links between data collection and personalization offers.

READ MORE:

But actions can’t be one-off measures; they require a platform approach that lets them share data across lines of business and without silos. Data has to be accurate and relevant and available for combining data at rest and real-time data, depending on use case. Development capabilities must be fast and responsive. Companies today can’t weasel out of ethics by relying on small-font Ts and Cs and boilerplate text to excuse their behaviour. It may comfort those that abuse data to feel that public tolerance of creepy behaviour is perhaps greater than ever but that would be to court risk and downplay the intelligence and awareness of modern consumers. As a company, you need to apply the latest tools but also to behave systemically in a manner that is human, empathetic and tasteful or risk public exposure and poisoning your brand.

For more news from Top Business Tech, don’t forget to subscribe to our daily bulletin!

Follow us on LinkedIn and Twitter

Amber Donovan-Stevens

Amber is a Content Editor at Top Business Tech

Unlocking productivity and efficiency gains with data management

Russ Kennedy • 04th July 2023

Enterprise data has been closely linked with hardware for numerous years, but an exciting transformation is underway as the era of the hardware businesses is gone. With advanced data services available through the cloud, organisations can forego investing in hardware and abandon infrastructure management in favour of data management.