Data Dashboards for AI Chatbot Product
Creating activity reports to better understand the efficiency of chatbots.
Project Overview
Taiger is an Artificial Intelligence company focused on hybrid AI solutions to unlock your company's operational efficiency by streamlining cognitive tasks and document processing automation.
CLIENT
Taiger
Team Model
Back-end Developer
Front-end Developer
Product Owner
PLATFORM
Web
TECHNOLOGY
Netlify
Java
VueJS
Anychart
Challenge
How can we view reports on activity to better understand the efficiency of our chatbots?
Request
Taiger hired CodeLink to create a visual dashboard that would display reporting from the chatbox platform. The dashboard includes information such as the number of users served per day, user satisfaction, error status, and more.
Engagement Model
CodeLink worked as an autonomous team to build the product from scratch. On the backend, we collaborated with Taiger developers to request and aggregate data from various customer sources. On the frontend, we worked directly with the client's design team to research different chart types and select the appropriate ones for each report.
Engagement Length and Scale
We released the entire platform within the client's 3-month budget window.
Project Outcome
Due to the client's time constraints and desire to get dashboards live as soon as possible, we recommended working with VueJS. It is simple to organize, easy to implement, and suitable for the size of the project. There was no need to divide it into many modules or use many libraries like ReactJS. VueJS helped us produce a solid and rapid web application. Once Taiger approved the technology, we worked to iteratively release the product. Throughout the project, we faced some difficulties in sourcing the different customer data points. Currently, there is a lack of the needed customer data to render the requested dashboards. To overcome this, we built many virtual data sources to simulate waiting until the real data is available. Using this mock data ensured that we could keep the development process uninterrupted and then replace the data with real data once the data points were prepared. We were able to release all of the desired functionality to production with all requested reports within the client's time window, meeting the desired release date.