The organization I am working for is working towards improving it's business intelligence capabilities. This opened me the doors to learn a brand new subject which I was never exposed to in my career anywhere before.
This is an innocent attempt to collect all my learning to one location which may help any business analyst who is a total novice like to to the huge business intelligence arena.
Just as any other project the very first step of a business intelligence project is a Request For Proposal (RFP). It is a document that solicits proposal, often made through a bidding process , by an organization interested in procurement of a commodity, service or valuable asset, to potential suppliers to submit business proposals.
As per the request the proposal was made and submitted which went through many discussions and finally we got the project after many iterations of proposal reviews and negotiation sessions.
So then the Statement Of Work (SOW) which again prepared together with contribution of the legal team.
All done and it is closer to start the requirements gathering phase of the project. This is my very first project and I wanted to learn some DW/BI before going to the client as I need to satisfy the customer and make a good impressions while making sure that the necessary details are gathered and brought back to the office for the team to implement the solution.
Since the beginning of a well designed Business Intelligence solution is a data warehouse, my first approach was to learn bits and pieces on data warehousing concepts. The learning outcomes until now on basic data warehousing concepts are listed below and definitely this added certain level of confidence to me before walking to a client.
Know Little DW/BI
Know the needful technical and business aspects - jargon/basic concepts/methodologies always plays handy.For most of the non technical business analysts in the IT industry find this challenging though worth giving a try as I guarantee the level of value it can add to you.
What is a Data Warehouse? A data warehouse is constructed and integrating data from multiple data sources. It can support analytical reporting and decision making. Further a data warehouse contain more generalized and consolidated data in a multi dimensional way.
Difference between a dataware house and a data mart is that while a data warehouse serves an organizations data mart is focused on on particular business segment. So as a business analyst we might need to collect requirements for a BI solution which supply analytics for the whole organization of a department.
For me a data ware house is also another database with more processed data which has pre-populated data in it.It removes the effort of querying and joining multiple tables in the database but have the data readily available and anyone interested can view data faster than in an operational database.
The type of data ware house needed for this assignment is a analytical type as the requirement is to slice- dice and drill down data to evaluate sales and marketing performance of the organization.
So me being the business analyst for the BI solution implementation for the XYZ organization, my first assignment is to understand the facts and related dimensions . Now what is facts and what is dimensions?
Fact table contains the numerical and historical data. Ex: Number of Orders Placed, Total Revenue, Quantity goes in to fact list as they are all numerical and organizations use these figures to take decisions on their business. Data warehouse consultants collect the facts to create fact tables.
Dimensions are categorized by which summarized data can be viewed.Ex: Number of Orders placed can be viewed against the location, time/date/month or even by sales agent. Further Quantity can be viewed against the time by customer by location or even by the sales personnel. The dimensions can create dimension tables.A dimension has its attributes.Ex: Customer dimension has name, address and country as its attributes.
Fact tables and dimension tables together create sachems.
STAR Schema
The fact table is connected to its dimensions in the shape of a star which is the reason why this is called a star schema. Dimension tables are not duplicated/repeated.
Each dimension table alone with its attributes is linked to its fact table and all the data in the schema are de-normalized.
SNOWFLAKE Schema
In the snowflake scheme there can exist tables which are normalized. Further the table structure is more complected and may not be able to provide the expected level of fast data retrials to the end solution.
The star schema is advised to be used in the projects.
When deriving the dimensions for a DW it is advised to use the most granular level as possible as it ease the analysis process.
I will be write on a BUS MATRIX - the first deliverable of requirements gathering in a DW/BI project as my next post.
Stay tuned. Next Post is Coming Soon.
P.S: Please feel free to comment on the items which I have not written correct as if will be a good learning to me and everyone else who is following me.
This is an innocent attempt to collect all my learning to one location which may help any business analyst who is a total novice like to to the huge business intelligence arena.
Just as any other project the very first step of a business intelligence project is a Request For Proposal (RFP). It is a document that solicits proposal, often made through a bidding process , by an organization interested in procurement of a commodity, service or valuable asset, to potential suppliers to submit business proposals.
As per the request the proposal was made and submitted which went through many discussions and finally we got the project after many iterations of proposal reviews and negotiation sessions.
So then the Statement Of Work (SOW) which again prepared together with contribution of the legal team.
All done and it is closer to start the requirements gathering phase of the project. This is my very first project and I wanted to learn some DW/BI before going to the client as I need to satisfy the customer and make a good impressions while making sure that the necessary details are gathered and brought back to the office for the team to implement the solution.
Since the beginning of a well designed Business Intelligence solution is a data warehouse, my first approach was to learn bits and pieces on data warehousing concepts. The learning outcomes until now on basic data warehousing concepts are listed below and definitely this added certain level of confidence to me before walking to a client.
Know Little DW/BI
Know the needful technical and business aspects - jargon/basic concepts/methodologies always plays handy.For most of the non technical business analysts in the IT industry find this challenging though worth giving a try as I guarantee the level of value it can add to you.
What is a Data Warehouse? A data warehouse is constructed and integrating data from multiple data sources. It can support analytical reporting and decision making. Further a data warehouse contain more generalized and consolidated data in a multi dimensional way.
Difference between a dataware house and a data mart is that while a data warehouse serves an organizations data mart is focused on on particular business segment. So as a business analyst we might need to collect requirements for a BI solution which supply analytics for the whole organization of a department.
For me a data ware house is also another database with more processed data which has pre-populated data in it.It removes the effort of querying and joining multiple tables in the database but have the data readily available and anyone interested can view data faster than in an operational database.
The type of data ware house needed for this assignment is a analytical type as the requirement is to slice- dice and drill down data to evaluate sales and marketing performance of the organization.
So me being the business analyst for the BI solution implementation for the XYZ organization, my first assignment is to understand the facts and related dimensions . Now what is facts and what is dimensions?
Fact table contains the numerical and historical data. Ex: Number of Orders Placed, Total Revenue, Quantity goes in to fact list as they are all numerical and organizations use these figures to take decisions on their business. Data warehouse consultants collect the facts to create fact tables.
Dimensions are categorized by which summarized data can be viewed.Ex: Number of Orders placed can be viewed against the location, time/date/month or even by sales agent. Further Quantity can be viewed against the time by customer by location or even by the sales personnel. The dimensions can create dimension tables.A dimension has its attributes.Ex: Customer dimension has name, address and country as its attributes.
Fact tables and dimension tables together create sachems.
STAR Schema
The fact table is connected to its dimensions in the shape of a star which is the reason why this is called a star schema. Dimension tables are not duplicated/repeated.
Each dimension table alone with its attributes is linked to its fact table and all the data in the schema are de-normalized.
SNOWFLAKE Schema
In the snowflake scheme there can exist tables which are normalized. Further the table structure is more complected and may not be able to provide the expected level of fast data retrials to the end solution.
The star schema is advised to be used in the projects.
When deriving the dimensions for a DW it is advised to use the most granular level as possible as it ease the analysis process.
I will be write on a BUS MATRIX - the first deliverable of requirements gathering in a DW/BI project as my next post.
Stay tuned. Next Post is Coming Soon.
P.S: Please feel free to comment on the items which I have not written correct as if will be a good learning to me and everyone else who is following me.
