Putting a Face on Retail Analytics

Big-data analytic solutions can provide brick-and-mortar stores with the same kinds of insight online tools provide to e-tailers. Physical stores gather data from loyalty cards, in-store cameras, smartphones, and even smart carts. Now, Synqera, a retail technology startup, has added facial recognition technology to the arsenal.

Saint Petersburg, Russia-based Synqera delivers customized marketing content for customers at the moment of decision. In July, according to a press release, it began piloting ďintelligent retail technology" that leverages big-data and real-time customer response in partnership with Ulybka Radugi, one of Russia's largest cosmetics chains.

Synqera sells software for checkout devices or computers that can personalize marketing messages based on a customerís gender, age, and mood, measured by facial recognition. How does it work? Synqera gathers data about customers from sources like loyalty cards and combines it with real-time information -- like the items in a shopper's cart and the shopper's facial expressions.

It then applies analytics to create a personalized promotion on the spot, which can be delivered in the store on video displays, or outside the store on the customerís smartphone or mobile device.

One promotion does not fit all, and the company maintains that knowing a customer's history and buying preferences can help retailers tailor promotions and increase the odds of making a sale.

The company considers facial recognition an important part of the process. Say a customer isnít smiling. The system can access her purchase history and suggest an appropriate purchase as a pick-me-up, possibly with a special deal on the item. Another possible tailored response is "a funny picture on the checkout screen."

A recent Guardian article stated, ďToday's shoppers crave personalization -- they want to feel as though the shopping experience has been created with them in mind. For that reason, video should feature a call to action that prompts shoppers to give their personal details.Ē

But is a blanket statement like that really accurate? The Guardian article cites a 2012 Accenture study, claiming the study "found that 49% of consumers in the U.S. and U.K. are happy to offer data in return for a relevant shopping experience, so stores should use video to prompt signing up for membership, a loyalty card or a mailing list."

What the study actually states is that 49 percent of consumers in the US and UK "are receptive to their favorite stores or brands using their tracking data to inform their future purchases and make them aware of product availability." That's not quite the same thing, is it? And one way or another, the reality is that stores still need to be sensitive to the privacy concerns of the other 51 percent.

Ekaterina Savchenko, Head of Marketing of Synqera, tried to address those concerns in an email interview. She told me that Synqera doesnít handle customer data, which remains in the hands of the retailer, and that customers have to opt in to allow the retailer to access the personal data from their loyalty cards. She explained:

No card -- no personal data. We may still see the facial expression and predict gender and approx(imate) age of the customer through the device's cameras and use it like triggers for content targeting -- equal to the data on the weather forecast in the moment of purchase -- but we never match this data with customer personal details. Please note that online stores do the same for years: (they) target ads based on the demographics and shopping history of the customer.
Savchenko confirmed that customers have to opt-in to participate in the solution's email, text, and mobile app functions.

While in-store facial recognition may make some people feel creepy -- as if they are being stalked -- itís not all that different from training a salesperson to pay attention to a customerís taste and the physical cues of her behavior for a personalized sales pitch. If youíre the kind of person who likes to be greeted by name and offered something that your history indicates you would like, then you may not mind personalized communication on a computer screen. In time, the practice may become as common as those ubiquitous purchase suggestions from Amazon.

Do personalized offers make you smile? Or do you value privacy over customized shopping experiences?

Ariella Brown,

Ariella Brown is a social media consultant, editor, and freelance writer who frequently writes about the application of technology to business. She holds a PhD in English from the City University of New York. Her Twitter handle is @AriellaBrown.

Hiring Trend: Machine Learning Reduces Bias, Increases Applicant Pool

Unilever is using machine learning tools to increase the pool of entry level applicants and then test and evaluates candidates in the pool to cull those who are the best fits. Adding machine learning tools reduces bias, increases diversity, and makes the whole process more efficient, according to Unilever.

Analytics and School Attendance: A Laundry Story

Appliance maker Whirlpool suspected there was a correlation between student attendance and access to clean clothes. Here's the story of how the company placed washers and dryers in schools and tracked the difference it made in student performance.

Re: Not sold
  • 9/27/2013 1:29:26 PM

I agree. Let them track someone who doesn't care.

Re: Not sold
  • 9/27/2013 12:51:11 PM

This sounds terrible! I would not like to be tracked in any manor nor would I like my store habits monitored.

Facial aging
  • 9/18/2013 10:08:37 PM

I do llike personalized offers it shows the retailer is thinking about me! However, I have heard that facial recognition can loose its accuracy over time based on aging etc.How would they address this issue?

Re: Keeping offers personal
  • 9/17/2013 9:12:49 AM

@Norren Though I admit to being absent-minded myself, I would recognize what I usually buy and what I don't. So something unusual would make me think, "Was  I planning to buy this for a particular reason?" But I think such an app would be most effective with husbands whose wives usually do the shopping. They're the type of men you are apt to see on weekends in supermarkets with their phones either glued to their ears or held in front of their eyes as they check on what they're supposed to buy because they don't know what the usual thing is. 

Re: Marketing Tools and Analytical Value
  • 9/17/2013 9:08:07 AM

@Louis, definintely, they have to not just think of ways that may increase sales but of what the customers' reactions may be. The assumption at work here seems to be that if they don't mind the tradeoff of discounts for data collection in loyalty cards, they won't mind this either, particularly as it is supposed to offer them what they would be interested in buying.

Re: Keeping offers personal
  • 9/17/2013 9:04:33 AM

@SaneIT that would be devious. It would also have rather a gaslighting effect on some of us, as we would think, "I don't remember writing that down, but it's here, so I must have. I must be losing my mind." People may just go back to scribbling a list of items on the back of an envelope the way I do then to avoid such mind games. 

Re: Keeping offers personal
  • 9/17/2013 1:52:35 AM

Great idea Beth !  Sorta like "Airplane Mode".   : )

Marketing Tools and Analytical Value
  • 9/17/2013 1:50:53 AM

Do personalized offers make you smile?

It depends on the product, if it is something I am interested in then yes otherwise it is spam.  But I think this could be a very interesting slant to analytics, Ariella.   How do marketing tools ( which are ultimately used for their Analytical value) affect the customer ?

I don't think many companies take the time to consider this.

Re: Keeping offers personal
  • 9/16/2013 7:36:36 AM

That's true.  There was an ad a couple years ago that featured a guy at the supermarket looking at his phone for a grocery list and things kept popping into his list.  At first it looked like maybe a spouse was adding to the list then it started getting junk food item after junk food item and the commercial cuts to two kids laughing as they edit the shopping list from a laptop.  I could see an app doing something like this or adding brand names for sponsored products.

Re: Keeping offers personal
  • 9/13/2013 10:02:27 AM

I'm afraid so many of us are absent minded that such an app could deceive shoppers of all ages.

Page 1 / 3   >   >>