CartHack

Home Screen and Shopping Suggestion
Results page

Inspiration

We have seen how our shopping bills have been rising each week and wanted to find a smarter way to shop to find deals across different supermarkets and keep the price of our meals down when prices are going up.

What it does

You input your shopping list into the app and it searches stores that deliver to your location to find the best prices for the things on your shopping list. This takes into account delivery costs when ordering from different places and finds the most cost effective ways to make your up the quantity of each item you want. The app also provides suggestions for similar items that could save you money on your shop. When we find the best shopping list for you we also send a text to your mobile phone containing the shopping list.

How we built it

Price data was obtained by scraping the relevant websites on demand in the NodeJS backend. Then we wrote algorithms to find the cheapest version of each item at the correct quantity from each store. We then took theses lists and compared how combining them by taking the cheapest items from each list and considering the extra delivery cost in different ways. We then took the best result which gives us the cheapest shop. We used Twilio to send a text message of the shopping list to the users phone.

Challenges we ran into

Each shop had different formats on the website so we had to use bespoke algorithms for each one. Time was also a big factor in not getting all the extra features we would have wanted.

Accomplishments that we're proud of

We got a working program that does a good job at finding the most efficient shopping list from different shops.

What's next for CartHack

We would like to allow you to place the order inside the app from different shops and be able to choose convenient delivery times.

Built With

Submitted to

HackTheMidlands 7.0

Created by

Worked on the algorithm to compared prices, stores and combined shopping lists. Worked on twilio implementation and the suggestions feature.

Daniel Jennings
I worked on the backend webscraping and API, as well as contributing to the frontend application.

Will R
I designed and created the logo and icons, and formulated the primary templates for the site. I also did some front end work in React which is a framework I've been interested in trying for a while.

Sragnet