1 Ease into this relationship
As shoppers begin using our applications, the data we collect is difficult to confidently act on. It’s general and ambiguous. Even information explicitly provided by shoppers early on isn’t enough to inform long-term personalizations. Remember, ‘actionable context’ is persistent and unambiguous. One-time events aren’t enough to accurately define relevant personalizations.
As a customer interacts with the environment and application over time, our inferences and ‘best guesses’ increase in accuracy, or at least they better. The more information we have, the more explicit and direct our personalizations can be. Don’t jump the gun. There is far greater risk that shoppers will abandon your application if it is overly presumptive and inaccurate in its recommendations and personalizations.
2 Understand the goals of the shopper to meet the needs of the shopper
Understanding the goals of the shopper is extremely powerful information when driving personalization and customization. As we build a more complete understanding of the shoppers purchase behaviors, shopping cadence, average transaction amount, time spent shopping, etc., we can assume what their goals are at a high level. The accuracy of our assumptions is not guaranteed so it’s important to seek opportunity to get explicit validation from the shopper. For example, if a shopper consistently purchases reduced-fat products and they are spending a long time in close proximity to the yogurt section, ask the shopper if they would like to compare the nutritional values of yogurt currently available on the shelf in front of them.
Timing is critical. Don’t be afraid to directly ask the shopper what they need to accomplish or want from the shopping experience. But be sure to ask them at a time that maximizes their benefit while minimizing the inconvenience of providing an answer.
3 Provide hyper-relevant recommendations
Hyper-relevant recommendations are the byproduct of intimately understanding the customer. This class of personalization is most successful when we leverage a diverse mix of quantitative and qualitative information of a particular shopper.
4 Understand, then support their personality, preferences and processes
For the application to seamlessly weave into shoppers’ normal interaction with the environment, we must support them as they plan (create lists, collaborate), shop, complete their purchase (checkout) and reflect on their experience. The method and extent of support we need to provide depends on shoppers’ personalities as well as the speed or cadence with which they prefer to shop. A task-driven or goal-oriented shopper just wants to get in, buy their stuff, then get out. Recreational shoppers can take their time, collaborate or engage with friends or family. The latter might find an application pushy if the experience is optimized for the fast-paced nature of the task-oriented shopper. If you don’t have enough information to accurately support a shopper, strike a balance between utility and sport.
5 Support Natural Language
Provide a robust vocabulary of terms related to the products and environment to allow shoppers to communicate with the application in a more natural and personal way. For instance, some shoppers might be looking for soda, while others want pop. We know the meaning of these two words are the same and so should your application. Understanding the personal subtleties of shopping is one thing, communicating it is another.
6 Leverage what people can do (focus on abilities, not disabilities)
Our applications should be designed with the understanding that shoppers have different levels of experience at shopping and potentially, physical and cognitive impairments that can limit their ability to complete their tasks. This includes situational-induced impairments like loud noise or inadequate lighting. Our applications should sense when a shopper is having difficulty and intervene when necessary. It should then use this information to improve their experience in the future.
As a general rule, design for flexibility, personalization and inclusion and the shoppers’ experience will be uniquely rewarding.
Measuring the cadence of the shopper can be valuable in determining the degree of mobility of the shopper (walking with a basket, pushing a cart, riding a wheelchair).
16 references informed this principle
 Adomavicius, Gediminas & Tuzhilin, Alexander, Context-aware recommender systems, Proceedings of the 2008 ACM conference on Recommender systems - RecSys '08, 335, 2008.
 Berry, Leonard L.; Bolton, Ruth N.; Bridges, Cheryl H.; Meyer, Jeffrey; Parasuraman, A. & Seiders, Kathleen, Opportunities for Innovation in the Delivery of Interactive Retail Services Journal of Interactive Marketing, Vol. 24, No. 2, 155-167, May 2010.
 Bhattacharya, Sourav & Floren, P, Ma \$ \$ iv-An Intelligent Mobile Grocery Assistant Intelligent, 2012.
 Camara, A, Natural user interfaces, Proceedings of the 13th IFIP TC 13 international, 2011.
 COOPER, D, User and design perspectives of mobile augmented reality, 2011.
 Dobson, S & Nixon, P, More principled design of pervasive computing systems, Human Computer Interaction and Interactive Systems, 2-7, 2005.
 Kourouthanassis, P & Roussos, G, Developing consumer-friendly pervasive retail systems, IEEE Pervasive Computing, 32-39, 2003.
 Kourouthanassis, Panos E.; Giaglis, George M. & Vrechopoulos, Adam P., Enhancing user experience through pervasive information systems: The case of pervasive retailing, International Journal of Information Management, Vol. 27, No. 5, 319-335, October 2007.
 Li, Ian; Dey, AK & Forlizzi, Jodi, Understanding my data, myself: supporting self-reflection with ubicomp technologies, Conference on Ubiquitous computing, 2011.
 Malizia, Alessio & Bellucci, Andrea, The artificiality of natural user interfaces, Communications of the ACM, Vol. 55, No. 3, 36, March 2012.
 Savage, Norma Saiph & Baranski, Maciej, I’m feeling LoCo : A Location Based Context Aware Recommendation System in Location-Based Recommendation System, Springer Berlin Heidelberg, 2012.
 Tamura, Hiroshi; Sugasaka, Tamami & Ueda, Kazuhiro, Enhancing grocery shopping experiences, Proceedings of the 12th international conference on Entertainment and media in the ubiquitous era - MindTrek '08, 114, 2008.
 Wobbrock, Jacob O.; Kane, Shaun K.; Gajos, Krzysztof Z.; Harada, Susumu & Froehlich, Jon, Ability-Based Design, ACM Transactions on Accessible Computing, Vol. 3, No. 3, 1-27, April 2011.
 Xu, Jie; Chen, PY & Uglow, Scott, Design Challenges and Guidelines for Persuasive Technologies that Facilitate Healthy Lifestyles, Computer Technology, Vol. 3, 140--147, 2012.
 Zhang, Xiaohui & Uruchurtu, Elizabeth, A User Experience Perspective of Design for Context-Aware Adaption, 2011 Seventh International Conference on Intelligent Environments, 1989, 322-325, July 2011.
 Consumer Choice Behavior in Online and Traditional Supermarkets: The Effects of Brand Name, Price, and other Search Attributes, A. Degeratu, A. Rangaswamy, J. Wu, International Journal of Research in Marketing, Vol. 17, No 1, P. 55-78. 2001.