Is age just a number? The art of the algorithm


The abundance of product and shopper information offers retailers the chance to remodel their companies, and the retail business. Wealthy linked information supplies alternatives to create true one-to-one personalised experiences. It’s clear that particular shopper attributes, comparable to age, and geography, have an effect on private preferences. In flip, producing related experiences based mostly on these insights about particular person shopper behaviours has change into a rising space of curiosity throughout the retail business.

Offering one-to-one suggestions utilizing primary, generalised shopper info isn’t true personalisation. True personalisation builds upon this primary info, taking the person’s desire under consideration and affords suggestions based mostly on these indicated preferences. Segmenting like-shoppers and providing all of them the identical group of suggestions just isn’t personalisation, however generalisation. Till every shopper’s desire drives the suggestions they obtain, true one-to-one personalisation can’t be achieved.

Take for instance the query of age, does a shopper’s age act as the first driver of non-public preferences, or is it considered one of a number of components that influences desire as an alternative? Right here, Rhonda Textor, Head of Information Science at True Match discusses whether or not age does in actual fact drive a shopper’s private preferences.

Within the age of the patron, practically all selections made by retailers are managed by the consumer’s desires and wishes. As extra manufacturers and retailers bolster their e-commerce expertise, there’s a increased demand to copy the normal brick-and-mortar (B&M) procuring expertise. The patron advantages of B&M procuring, comparable to the flexibility to attempt on garments and sneakers and elegance with different items, are troublesome to copy digitally. Customers demand individualised digital experiences and push retailers to assist them discover suggestions based mostly on their distinctive type, dimension, and match preferences. In doing so, retailers can count on elevated income, AOV, and shopper satisfaction, which results in higher shopper loyalty.

A key subject of curiosity in producing true one-to-one personalised experiences for the patron, and but a controversial one, is age. A typical business strategy is for designers, retailers and suggestion engines to section individuals into averages and make assumptions about what a shopper desires to put on, based mostly on their age.

However are sure attire “attributes” which are historically geared in the direction of older shoppers (for instance, longer hems for ladies and a less-fitted silhouette for males) reflective of older shoppers preferences? Or does age merely affect these preferences, whereas different components really drive them?

Product & shopper information

Utilizing a wide range of totally different shopper demographic information retailers can provide correct product suggestions which are based mostly on the person’s distinctive preferences. Client attributes can embrace age, physique mass index (BMI), and geography, in addition to earlier buy behaviours. The mix of this totally different information is when the advice turns into actually private and related to the person shopper.

No buy historical past

When a shopper outlets with out earlier buy historical past information, retailers can predict which kinds may align with their preferences, based mostly on the patron’s age, peak, weight, and dimension. As soon as buy historical past information from the patron is obtainable, the suggestions generated change into particular to the person.

Suggestions change into extra correct when shoppers make a number of purchases as a result of a retailer is ready to analyse their procuring behaviours. The extra transactions an algorithm can analyse, the higher suggestions they’ll serve, which are distinctive to the person shopper.

With none earlier buy historical past, shoppers within the older demographic are inclined to go for tops with extra sleeve protection, and the youthful shoppers buy extra sleeveless choices, (based mostly on aggregated information from shoppers throughout the identical demographic). 

Nevertheless, as a way to present the very best type, match and dimension suggestions to shoppers, you have to take into account a shopper’s model affinities, buy historical past and match preferences. Taking all of this totally different information under consideration ensures that the person shopper receives suggestions which are related and actually personalised.

Whereas there’s a pronounced desire for sleeveless tops or attire among the many smaller and youthful customers, by incorporating buy historical past information into the equation, many patrons within the high quarter of dimension and age distributions additionally select sleeveless kinds, suggesting that age doesn’t drive a shopper’s desire, however moderately solely influences it.  

With buy historical past

By analysing shopper behaviours with buy historical past information, retailers can higher perceive the consumer’s preferences. If a client’s age is the one issue being thought of, then they might obtain suggestions which are skewed to incorporate gadgets with kind of protection. The additional layer of information, like buy historical past, filters in additional gadgets that match the consumer’s preferences and are prone to be of curiosity to that shopper, moderately than simply assumptions made on their age alone.

There are a number of situations the place older shoppers might present an curiosity for shorter sleeve choices or youthful shoppers begin to buy tops with extra protection. When a shopper’s buy historical past reveals a desire for sleeveless choices, it’s probably she is going to buy extra sleeveless choices sooner or later. Algorithms be taught to prioritise these preferences, expressed by the person shopper, moderately than defaulting to suggestions based mostly solely on age.  With this information, retailers can goal advertising, ads and proposals in the direction of the shoppers who constantly point out their preferences, for instance, shorter or longer sleeve tops.

When suggestions are supplied with out the added information layer round desire, shoppers might really feel underserved or that they’ve been falsely advisable merchandise that don’t fulfil their desires and wishes, inflicting a damaging expertise and weakened loyalty. The added information layer of buy historical past ensures that each shopper receives an individually curated suggestion from the retailer.


When a shopper considers buying a product, particularly within the attire and footwear business, there are a number of things that may have an effect on whether or not the patron in the end makes a purchase order. The type, match, and worth of a product are three principal issues that buyers face when deciding which gadgets greatest match their preferences.

Analysis reveals that whereas a client’s age signifies which merchandise they’ll probably be considering, age alone can not drive actually private procuring experiences. Age is simply a chunk of the bigger puzzle.

Particular person type preferences are nuanced and require greater than age and dimension to mannequin precisely.  Information a couple of shopper’s age is essential as a result of it could possibly assist retailers achieve an summary of the kinds shoppers amongst totally different demographics usually tend to love and hold.

Even with this small quantity of volunteered info from customers, retailers can already present related, personalised procuring suggestions. For customers who lack a gross sales historical past with on-line retailers (because of lack of on-line purchases or privateness issues), offering a restricted quantity of non-public information nonetheless leads to related content material.  Frequent customers who’ve a gross sales historical past can have a extra personalised expertise as a result of machine studying algorithms can seize a extra refined relationship with clothes options.

Retailers ought to look to make use of an algorithm that is ready to curate product suggestions for shoppers, with and with out earlier buy historical past, permits them to energy the very best type suggestions to their shoppers, based mostly on demographic and product information. Age helps retailers information their suggestions however can not exchange the worth in producing suggestions based mostly on every shopper’s expressed preferences.

To search out out extra about how a shopper’s age influences his or her match and elegance preferences obtain The Artwork and the Algorithm: A Client Behaviours Report here.

Rhonda Texter, head of data science, True Fit

Source link

Leave a Reply

Your email address will not be published.

Previous Post

The Value of Big Data to the Pro AV Industry

Next Post

Happy New Year and new decade: Thursday Wake-Up Call

Related Posts