Introduction
Dashboards are still an incredibly popular task given to a data scientist or analyst in most firms. In a perfect world, you might connect to big streaming data pipeline or set up an individual Data Warehouse with OLAP cubes to feed a complicated set of real time analytics. Then build some beautiful flexibility into the app for the business teams to use for ad-hoc numbers. _Maybe _you even add in some flare with a few prediction algorithms augmenting the data. That sounds like a lot of fun. This is not that dashboard.Alternatively, sometimes you need to build a dashboard around an existing set of processes and do very little to disturb the business. That’s where this dashboard lands. The idea with this project is to support a team of field marketers working for a _fictional bicycle company. Each marketer hosts physical events around the country and needs to report on the type, size and frequency to the head of marketing. Additionally, these metrics need to track the progress toward a bonus incentive given to the employees. This f_ictional big company is one with a lot of red-tape, departments inside of departments and an overworked IT team. So, the head of marketing is game for a very simple solution that can completed in a few days that doesn’t require sitting in 3 budget meetings to be approved or training the field staff on a new tool.Enter Google Forms and Sheets. Rather than create another production level data store and web form to ingest data, we will use the simple, free and very familiar tools from Google. This allows for quick adjustment and editing of data by the team. And, as you’ll see, Google has given access to some of the same code they use to build the G-Suite in order to make our own customized solutions.This project will be broken up into 4 parts.
- Design and ETL Process using Google and code.gsConnect the Google Sheets Data to pythonBuild a Plotly Dash appHost it on Heroku [coming soon]
#dash #dashboard #data-science #plotly
