Inspiration
Did you know people spend more than 36% of their salary on purchase of essentials? 77% of people buy grocery every week. 95% of the time, people make similar purchases. An average buyer spends at least 30mins per week on making the right shopping list. There are many items that users purchase periodically. For e.g. if its bread, they probably need it everyday, groceries probably every 10 days and for people who have little kids, they probably need diapers on a monthly basis. Therefore, just curating the list for buyers exactly when they need it presents a huge opportunity for sales conversion. Leveraging the power of AI and NCR apps, we bring to you, your own personalized instant shopping cart. Our machine learning model takes into account your purchase history e.g. frequency of purchase of an item, last time of purchase, average duration between purchases and predicts what you need and when you need it. We are offering functionality for the future of retail, none of the bigger retail brands currently provide this feature.
What it does
You receive the notification that your instant cart is ready. Instead of scrolling through so many items, you go to the instant cart by clicking here. You can see, the items have been pre-populated as per your purchase trends. You can modify the quantity, select/de-select items and then with a single click, place your order. Voila! Shopping all done. As a business, using speedmart gets you increased sales conversion and customer retention. Not only that, our predictive model can also help businesses optimize their supply chain through better inventory planning. For customers, our solution saves their time and provides a seamless experience.
How we built it
We used Javascript, HTML, CSS, Python for ML, NCR catalog and order APIs
Challenges we ran into
We ran into numerous problems with getting the NCR API work correctly. We dived into documentation and experimented with APIs to understand their functionalities. We are new grad students who worked together for the first time we are happy we were able to coordinate successfully.
Accomplishments that we're proud of
We started without a pre defined idea. The way we transitioned from ideation phase to working product is something we are really proud of.
What we learned
Working in a team. We got good understanding of use-cases for NCR APIs and motivation for their design.
What's next for SPEEDMART
Obtain more data after deployment and improve prediction algorithm by including more parameters as we work on larger data set. Extend this POC to retail stores.
Log in or sign up for Devpost to join the conversation.