How Predictive Analytics helps get the ‘Bikini Body’

Julia: “Looking forward to our much awaited trip to Ibiza @Kate @Jen. 3 weeks to go! Sun, here we come!!”

Julia just updated her Facebook status. She sounds thrilled on social media but is in fact petrified. She knows that they will all be wearing bikinis at the beach and she does not think she has a ‘Bikini Body‘.

She picks up her smartphone and frantically searches the web for ‘How to get a Bikini Body in 3 weeks?’ She is overwhelmed by images of airbrushed models and diet regulations recommended by fitness experts. She feels low after learning this new found information on the effort required for her desired transformation. In this state of mind, her ‘go to’ is either comfort food or shopping. The latter seems like the wiser option to her.

She logs on to her favourite online shopping store, Fareys[1], and starts browsing items she would require for her upcoming trip. She looks at the range of bikinis, saves her shopping basket online and decides to pay later that week after discussing with her girlfriends.

Later that day she meets a friend for a cup of coffee in her local town and a notification pops up on her smartphone. It’s from Fareys. She clicks on it and finds that the store has listed bikini options, accessories to accentuate the bikini, SPF quantities etc. in an e-mail entitled ‘For your beautiful Bikini Body’. It has also made recommendations of the closest stores with availability of in stock bikini colours.

Is someone reading her mind? Fortunately, no. The world as we know it has become smaller as a result of the convenient connectedness of the internet. The immediacy and accuracy of real-time data transfer has become the norm. Fareys uses Predictive Analytics (PA) to estimate what its customers want and how they will behave. Their Analytics search engine combines location data with calendar entries, emails, web search history and more to predict what a user will do next and offers relevant help and information at each step without ever being asked. Retail industries have used PA for several years to work out a person’s customer score which would determine, for example, how likely they are to buy a particular product. This is based on algorithms that combine various snippets of information to form a prediction.

The smartphone scans calendar entries to work out where a user is, or should be. It combines this with location data, taken from the phone or tablet’s GPS unit, as well as posts on social networks, email information and more. It will then present the user with recommendations designed to offer support or help it thinks they need. A recommendation could also be created to show what the weather is like where they are going to and even advise taking an umbrella. Back in 2012, researchers from University of Birmingham devised an algorithm that could successfully predict a person’s future locations down to 20 metres (source: University of Birmingham).

In Julia’s case, Fareys has used real-time analytics to know that she has a holiday planned and where she is.  Fareys assigns every customer a Guest ID number, tied to a unique identifier that becomes the key that stores a history of everything they’ve bought and any demographic information Fareys has collected from them or information obtained from other sources, for example, articles browsed on social media. Using that, our fictitious retail giant looked at historical buying data for all the ladies who had signed up for Fareys’ bikini sales in the past.

They found that lots of people buy SPF lotion, but one of Fareys’ analysts noticed that proportionally more women on the bikini registry were buying SPF lotion, sarong wraps and hats around their holiday period.

They were able to identify several products that, when analysed together, allowed their analyst to assign each shopper a ‘bikini prediction’ score. More importantly, they could also estimate Julia’s holiday to within a small window, so Fareys could send coupons timed to very specific times of the year.

Do you like your Facebook feed? What about those recommendations on Amazon, Spotify and Netflix?

These companies use associative learning algorithms which might reveal, for instance, that customers who bought a cocktail shaker and a cocktail recipe book also often buy martini glasses. This information can be used for predicting the likelihood that a particular user will make a particular purchase. These types of findings are often used for targeting coupons/deals or advertising. Facebook uses the number of photos tagged, friend requests initiated or accepted, comments, likes etc., to predict future engagement for a user.

While the retail  industry has proved to have a greater appetite for PA, other sectors including Non-Profits, Agriculture and Real Estate still have a long way to go. PA can be used in any industry which has a supply chain value e.g. in agriculture, and to reduce building operating costs or improve energy efficiency by adopting smart-building technology e.g. in the real estate sector.

A group of researchers from Goldsmiths, University of London, published a study whose results show that nudity can have a positive impact on our mental health. ‘Body image dissatisfaction is a serious, global problem that negatively affects life satisfaction‘ (source: Goldsmiths, University of London). Our body image is a learned behaviour, shaped by our families, peers, culture, and most definitely by the media. The way Julia feels after looking at those air-brushed models is a result of her perception of her body image. Interestingly, body image is affected more by your self-esteem than your actual physical attributes. Even those who look like they are perfect will still have insecurities.

Analysis of data from the internet of things reveals that the top 3 tips for getting a ‘Bikini Body’ are:

1. Choosing the right bikini: Wear the bikini that fits well and feels good. This will instantly boost your confidence and comfort. Include accessories like scarfs, hats, earrings, sunglasses, sarongs, etc.

2. Developing Confidence: Feeling confident in a bikini isn’t really about losing that extra weight or having a supermodel body. Confidence comes from wearing clothes that fit you, and feeling good about who you are. Believe it or not, one of the best ways to look taller, leaner and more confident instantly is to stand up tall. Focusing on what you need to improve before you’ll allow yourself to feel confident is a losing game. Once you realize that your body is perfect, just the way it is, you’ll feel a lot better about putting on any kind of bikini. You cannot possibly be thin enough, toned enough or strong enough.

3. Treating your body well: This hints towards mindful eating which means that you eat when you’re hungry and stop when you’re full because a lot of weight gain boils down to eating when you’re not hungry. Give yourself permission to eat as much as you want and figure out what makes your body feel good. Doctors don’t know of any approach that leads to a significant weight loss in a lot of people. If diets worked, we’d all be thin already.

Fareys are aware of this information and use it to promote confidence in its targeted shoppers.  Julia receives a series of follow-up e-mails from Fareys after showing interest in their first mail. Every successive email comes with a quotation of the day which is directed towards making her feel better about her body. This is derived from research and the data mined after Julia’s choice of articles browsed.

She is now really looking forward to her trip, has bought the saved shopping basket after making a few alterations including the bikini type that suits her body, some fancy accessories and lots of SPF. Fareys has been a blessing in disguise and she is certainly one of their loyal customers.  Julia finally achieved a ‘Bikini Body’, with a little help from Fareys.

The fact of the matter is, a ‘Bikini Body’ is a body with a bikini on. It’s about realistic women wearing revealing clothing, and loving every inch of their skin.

Although there is the potential for our personal data to be used in unwanted ways, it can also be used to make our lives better. We are all time-constrained, yet we expect personalised services. Are we ready to forego a little from a data privacy perspective, to achieve that? Should we choose to do this, we will be aiming for a utopian future where PA will help decide what the customer wants, based on concrete data about what they need; where the experience and gut of market experts will be questioned constantly, ensuring that a new product will never be a hit-and-miss.

[1] Fareys is a fictitious name used for demonstration purposes.  It is not a real on-line trader.

More Analytics Blog Posts