Powering the final solution

The summarized solution

We went out on a mission to elevate the businesses of Diagon Alley by improving the customer experience and creating a solution to help the business automate their day-to-day tasks and improve their efficiency and revenue. As a way of simplifying the development, we have chosen to focus on the local pet store as proof of concept solution, but always with plans of expanding in the back of our heads

In our attempt to move the store lines to the cloud, we have a given the customers a Power Portal website where they can sign in to order pets, food and cages, select a preferred estimated delivery time in addition to the cost of individual items and in total. To help the customers use this site, we have also developed a “magic” mirror that gives recommendation for a pet and provides a QR-link to the registration form with the suggestion automatically set.

The customers can also get an easy access to the website using our “magical” smart mirror known as “Mirror of Noitceles”. While standing in front of this mirror, a camera will take their picture and, using Azure Open AI, will recognize facial features and come up with a suggestion for a pet. They will also be given a QR-code they can scan that provides them a link to the registration website with the recommended pet already set in the registration form.

By using the Power Platform to create the solution, we have created an easily accessible solution that could be implemented into Teams and therefore by available for business owners that works on the floor all day. This way, we also can provide an integrated connection with SharePoint and their documents handling and simplify communication by implementing textmessage, phonecalls and pre-made Power App components.

This also gives an easily maintained database in the shape of Dataverse, and by synchronizing this to Fabric we also have solid foundation for adding future modules that can improve the experience by sharing data between different businesses and develop complex AI-solutions that will improve their experience and return of interest further.

We also recommend further reading at https://www.behance.net/gallery/217701961/Mirror-of-Noitceles-Making-every-selection-magical

Category: Fabric Fera Verto

As the business using our solution, and their customers, are registering data into Dataverse, we are synching this into Fabric in the backend part of the solution. To maximize the value of Fabric, we also have connected an Power BI report that displays the data we need to improve the solution in the best way possible.

By doing this we are creating a massive data collection that can be used to improve the solution between business and train AI-models to create an even better solution and modules to add on in the future.

Category: Low-Code Charms

Our solution is mainly taking advantage of the magical powers within the low-code Power platform. We are using Power Pages to maintain a website for customers where they can register their order to the store. This is also connected to our smart mirror that recognized customers, suggest a pet based on their appearance and provides them an QR-link to this website with the suggested pet already selected.

We are also using a model-driven Power App to maintain the data within the Dataverse, and allows the employees to handle the customers order. They are able to see all orders, pets, and type of food in a entity-view (tables) – as well as an Admin-form with a kanban-board (thank you Resco) for updating Order-statuses and see the sales-stream, and a PowerBi-report (connected to Fabric). The store owners also have an “Admin”-page set up with an overview of available pets, their aisle location and more, with the possibility to register new Pets as they become available.

The solution is further improved by using Power Automate to automate tasks that that usually takes unnecessary time, like sending and receiving text message or generate supplier agreements in SharePoint and sending supply orders.

Category: Pro-Code Potions

As a lot the customer functionality is handled by low-code solutions, the heavier aspects of the solution is handled in the backend using more traditionally coded functionality.

We have a Raspberry Pi configurated with the “Magic mirror” node.js solution (https://magicmirror.builders/), and have improved this by adding and adjusting modules, including a list view of a Dataverse table and the generation of QR-code that send customers to the ordering website. We also have adjusted the Power Pages website by using jQuery to implement URL parameters so the suggested pet shows up automatically.

We have set up a scraping solution to the Harry Potter Wiki-page to integrate data from this universe, and are using the external python package “Cv2” to handle API’s for facial recognition. Oth er technologies we have used are msal, axios, Flask and Websockets.

Category: Digital transformation

As a lot the customer functionality is handled by low-code solutions, the heavier aspects of the solution is handled in the backend using more traditionally coded functionality.

We have a Raspberry Pi configurated with the “Magic mirror” node.js solution (https://magicmirror.builders/), and have improved this by adding and adjusting modules, including a list view of a Dataverse table and the generation of QR-code that send customers to the ordering website. We also have adjusted the Power Pages website by using jQuery to implement URL parameters so the suggested pet shows up automatically.

We have set up a scraping solution to the Harry Potter Wiki-page to integrate data from this universe, and are using the external python package “Cv2” to handle API’s for facial recognition. Oth er technologies we have used are msal, axios, Flask and Websockets.

Category: Magic Matrix

We are using a combination of Teams and Power Platform to create a magical solution for businesses and customers. We have document handling in SharePoint, Customer communication with phone calling and text messages in Teams as well as a good integration with Power platform

Leave a Reply