Define Your Data

Before you start any PhixFlow project, it is important to take time to understand your data and what you need to do with it.

For this example, a sales team is using an Excel spreadsheet to manage a list of Business contacts. The data represents business customers (location, turnover) and people (names and addresses). The spreadsheet has the following tabs.

Businesses tab:

Contacts tab:

The data in the two tabs is connected by the the business name.

You want to make it accessible and maintainable by more than one person. You also want to add some simple CRM functions, so that you can add descriptions about the business.

Explore the possibilities

Remember that you can:

PhixFlow can connect to data stored in external databases (list or link?) or data files in a range of formats (list or link?)
PhixFlow analysis models are a powerful tool that you can use on any data project. Analysis models can help you to improve your data integrity, by
- identifying missing data, duplication and variation
- reconciling data in different files or systems.
The Introduction to Modelling Course shows how to do this
set up your own styles and palettes to create completely customised layouts. The Building Applications Course goes through this process.

You can represent your data and requirements in an Entity-Relationship Diagram (ERD); see Wikipedia. It can help you to understand the data entities, their attributes and relationships. An ERD shows you the data architecture. As well as clarifying the data you already have and what you need to calculate, an ERD can help you identify:

static data: once added is unlikely to need updating. For example, the list of possible salutations:
- Mr, Ms, Dr, Sir, Lord
control data, that PhixFlow uses for screen input, for example data that is displayed in a drop-down list.
CRUD data, that PhixFlow will be able to create, read, update and delete.

Within PhixFLow it is a good idea to be consistent, at a minimum within an application. You can use prefixes to distinguish CRUD or control data from static data. During the planning phase you can set up a naming policy.

Move out

Before you start any PhixFlow project, it is important to take time to understand your data.

What is your data? Do you need to represent people, places, products, money, time? These are the data entities.
How does the data connect? These connections are the data relationships.
How will you represent the data in PhixFlow? This depends on the attributes the data has.
Is the data "clean" or do you need to tidy it up?
What data is missing? PhixFlow can combine existing data from different sources and use it to calculate additional data.
What volume of data do you need to handle? For very large data volumes, you will be able to use modelling strategies to process data efficiently.
What do you need to do with the data? What is the business problem you want to solve?