Why :: Choosing a Data Model approach¶
Toucan Toco is not a data integration tool. Toucan Toco’s characteristic is to retrieve the data in the most actionable way possible. We offer a simple and easy to use tool.
Our approach is data storytelling. We take the user by the hand; we guide him to rediscover his data and help him to make decisions. We write data scenarios. This approach leads us to define in a pragmatic and precise way what data we need. The objective is to prepare only the data necessary for the business users.
Creating a data model is part of our Agile Design First methodology¶
The setup phase of the Toucan Toco application is supported by an agile project management method, divided into 2 big phases (check it out in detail in How to :: Lead a Toucan Project.
- 1st step is about Design, made with fake data
- 2nd step concerns Data Integration, it’s when your real data is integrated into the app
This Agile Design First method allows us to target precisely the needs in terms of data modelization. The goal is to end your project with 2 deliverables for the client :
- A perfectly designed application for the end-users
- An associated data model so conceptors know exactly what data they need to extract, in which format
Concretely, how does it work? Once the design has been validated with the business stakeholder, we build and send to the client project team a Data Model that will give the structure and drive the preparation of the data. Either the client will need to fill it (if flat files are used) or to reproduce it (if connexion to client database is used) to integrate his data into the app.
It starts the industrialization phase of the project - that it is to say the data integration and the automatization of data updates.
What are the advantages for the client ?¶
- The client has full control over the indicators and calculation formulas. It can update the data completely independently
- The client is autonomous regarding his business rules (e.g. changing training cost formula)
- Data integration into the application is fast. Because data are pre-aggregated and the format is adapted to the tool
- Data check is quick and efficient
- We have more time to connect the application directly to the client’s etl tools thanks to our connectors —> App industrialization is faster
Prerequisites¶
- You need someone to talk to : check that there are people in charge of the data in the project team, and make sure to involve them at the very beginning of the project.
- You need someone to work with : check that resources are available for industrialization. These people will be committed to generate data extracts and iterate with you on the data format.