Datary

Scan the QR code and open the Dashboard (Need to have an Microsoft office account to open it))
Datary Brand Logo
Datary-Home
Datary-Overview
Anomaly Detection
GIF
Interactivity Between Graphs
GIF
Trend in Sales through years, quarters, months, and days
GIF
Question and Answer with Free Text Field

Inspiration

We plan to promote the concept of citizen developers in the company. This helps non-tech employees automate their daily needs using low code platforms such as PowerBI.

What it does

The tool or solution that we have developed supports data analysis by providing predictions, detecting anomalies, suggesting rationale for this anomaly, answering questions through a free text field ,as well as leveraging AI algorithms to generate decomposition trees to give a broader overview. An overview of the solution with gif images are also attached.

Based on this overview on the dashboard itself, a short story of these statistics can be created and shared with other employees within the company. This helps save time and makes the process faster.

How we built it

We use PowerBI and Power Automate to do data analysis, anomaly detection, as well as automation of the data pipeline. For the video, we used an open-source editor on the dashboard which can create videos, as well as edit them. Once the video has been created it can be shared and stored securely with Sharepoint, Google Drive, and other data sharing services. It can also be published to other video sharing solutions such as Vimeo and Youtube.

Challenges we ran into

We did not have a professional license, so we couldn't develop the video editor tool on the dashboard. Also, there were challenges analyzing the data because it is not clear what value x, y, and z from the columns meant. As an example, we had to translated the city id into an actual city name for our heat map.

Accomplishments that we're proud of

We were able to draw insights to make actionable recommendations and make the video and dashboard easily consumable by upper management. A non-tech person can also easily add whatever additional feature they would like. The dashboard can be easily customizable to add further data on the retail sales. All these points address the main areas of the challenge that was provided to us by SAP.

What we learned

We were targeting user-friendliness and maintainability for this project which is why we have chosen a low-code citizen developer technology. Since it's a hackathon, only a proof of concept needs to be developed in a short amount of time. We thought of using a solution that already exists and solves all the challenges, instead of reinventing the wheel. Therefore, we used PowerBI. In this process, we learned this tool in a short amount of time and how easy it is to develop with everything available. The concepts such as summary narratives, Q&A, AI decomposition trees, and a range of visualizations show or cover everything desired to understand the retail data.