Transcript Document

Business Intelligence
At
Insight
SUBHEAD OR DATELINE CAN GO HERE
About Insight

Established in 1988

Premier single source provider of IT products and services

Fortune 1000 company with 4,500 employees

Headquartered in Tempe, AZ with many locations throughout
the U.S.

International presence in Canada and the United Kingdom

Over 200,000 name brand products from over 1700
Manufacturers

Specialize in complete IT lifecycle services
Goals of Presentation
On Completion you will:

Know what Insight used prior to BW (Business Warehouse)

Understand the existing architecture in the BWDSS (Business
Warehouse & Decision Support System) environment

Understand the main reports and tools contained in BWDSS
environment

Be able to create a simple query through BW
Prior to BW

MAX – IMS BASIC based system
–

Data mining tools
–
–

Data stored in Flat files
Excel
Access
Problems
–
–
Reports were saved locally
Reports were not consistent
Current System Architecture
SAP R/3
System
Master Data &
Transaction Data
Updates
SAP BW
System
Current System Architecture –
contd
SAP BW ( Business Warehouse)




Used as the extraction point for external reporting data
Used as an On-Line Analytical Processing (OLAP) server
for analysis
Not real-time; feeds done nightly
Data extracted, cleaned, and transformed into queryable
objects
Current System Architecture - contd
–
DSS ( Decision Support System)



Relational database fed from BW, Symposium, and
other sources
Not real-time; same schedule as BW
Back-end for Reportzone
Current System Architecture - contd
–
Reportzone



An Active Server Pages web application for visibility into
DSS system
Provides quick, canned reports company-wide
Transparent security model with no need for logging in
Rationale behind architecture
–
Technical

Needed to distribute load due to 1500 users

Each environment serves a different purpose:
BW – Ability to analyze large amounts of data without having
to know how to write SQL
– DSS – Ability to manipulate data and run complex queries
against multiple data sets without BWDSS team to model
data in advance
– Reportzone – Quickly display data for ease of use
–
Rationale behind architecture
Business

Supports three distinct types of users:
Technical Users – adhoc reporting, somewhat complex
queries
– Power Users – analysis, easy access to large volumes of
data
– Casual Users – spoon feed users information that is deemed
appropriate
–
Data Extraction Procedure

Setup the extraction structures
– A structure is a definition of the layout of the Data elements
pertaining to a specific Transaction ( like sales Order)

Make Structures available in both the Source and the Target systems

Setup periodic jobs to send Data from R/3 to BW.

Also setup change pointers – for change documents
Data Available

Sales Orders
–

Deliveries
–

Customer master details
Materials
–

Invoice Header and Line item details
Customers
–

Order Delivery details
Invoices
–

Order Header and Line item details
Material Master and Virtual Sourcing Data
Symposium / Avotus ( Phone System)
–
–
Only through DSS
Call summary data and call detail data
Reports Available
–
Commission Recap – Quick access to commissionable invoices by
day.
–
Performance Recap – Rep scorecard, includes variety of metrics in
addition to commissionable GP.
–
Customer Purchase Reports – Quote / Order / Invoice history for a
customer.
–
Top 50 Reports – Displays top 50 customers, materials, material
groups, material pricing groups, and manufacturers for both MTD and
yesterday.
–
Flash – Product Manager scorecard for determining revenue and GP on
a monthly basis, relative to budgets they assign themselves
Reports Available - contd
–
Contest Reports – Simplified reporting wizard to query data for contest purposes
–
Sales Query – Simplified query tool for reps to search for sales in their owned
accounts
–
Call Lists – Tool for Analytics team to build a call list out of a rep’s book and get
that list back to sales
–
Cost Adjustments – Tool to allow the Sales floor to put in cost adjustment
requests that work their way through an approval process up to Accounting
–
Inter-Company Referral Tool –it allows the COE (Center of Excellence) groups to
communicate with the Sales floor
User Profiles

Three types of users in BW:
– Technical Team
– Power Users
– Casual Users

Technical Team will create and save queries that are not editable by anyone
else. Used to control information for certain types of reports (such as 10Q/K)

Power Users have the ability to create queries and modify any query other
than those created by the Technical Team. This will be most of the ISBAs,
and ultimately will include FBAs and Department Power Users.

Casual Users only have the ability to run queries that someone else has
created. This will be people that are not trained on the system, but may need
access to do some things adhoc if necessary.
Objects to Query
 ODS
–
Stands for Operational Data Store, which really means it is a
flat structure, i.e. a database table
–
These are the foundations for a cube, and ultimately provide
the granularity behind the cube data
–
Fully queryable, just like a cube
Objects to Query - contd

Cubes
– Multi-dimensional structures created to allow fast access to mass
amounts of information.
– Structured around two key concepts: Key Figures and Characteristics
 Examples of Key Figures: Revenue, Cost, Quantity, Number of
Invoices
 Examples of Characteristics: Sales Organization, Sales Office,
Payment Term, Sales Person
– Characteristics in BW are known as Dimensions in other Data
Warehousing tools
– In BW, a group of characteristics comprise a “Dimension”
 Example: We group the following characteristics into a Dimension
called “Organization”: Company code, Sales Office, Sales
Organization, Sales Group
Demo of a Query from BW
 BEX
–
(Business Explorer Analyzer )Analyzer
Can be run through a variety of mediums, Like Excel/Access
– User-friendly drag and drop tool, doesn’t require knowledge
of underlying structure of the data, or any coding
Demo of a Query from BW - contd

The BEX runs via Excel . The only difference initially is the BEX
toolbar:
Demo of a Query from BW - contd

Click on the far left icon (the open folder) and select Queries:
Demo of a Query from BW - contd


We will then see things based on the access level of the user (in this example, we have
logged in as an SD ISBA (Sales & Distribution IS business Analyst).
Click on the white paper in the tool bar at the top right to create a Query
Demo of a Query from BW - contd


We get a choice to what we want to query. As an example, if we want to see
invoiced sales (revenue) by sales organization since 1/1/04.
To get there, we will go to the Sales and Services InfoArea (selected below).
Demo of a Query from BW - contd

We navigate to the Sales and Dist folder off the Sales and Services area, and it
gives us the available queryable objects in this folder.
Demo of a Query from BW - contd


Since our example was regarding Invoices, we know to navigate to the “Cube”
labeled “Invoices” and double-click.
Note the different icons on each type of object.
Demo of a Query from BW - contd

We are now presented with the actual query designer. This is the drag-and-drop
tool used to query BW data.
Demo of a Query from BW - contd

Areas of the query
designer:
–
Left Window Pane
includes the objects
available to select in your
query. It is organized by
Key Figures (numerical
values) and Dimensions
(areas of characteristics).
–
Filter: allows you to reduce
the records returned, using
any field you would like,
except for the ones you
select in the other areas.
Demo of a Query from BW - contd

Areas of the query designer:
–
Free Characteristics allow us
to select characteristics that
we might not want to include
in our initial query, but might
need to in the future, allowing
us to not have to go back to
the designer to do so.
–
Rows and Columns: defines
the x and y axes of your query.
They will behave the same.
Note, we must include at least
one Key Figure and one
characteristic for a query to
work.
Demo of a Query from BW - contd

In our example, we want
invoiced revenue by sales
organization since 1/1/04.
Thus, we will need sales
organization characteristic,
the product revenue key
figure (what we’re trying to
measure), and a filter on
billing date.

As listed on the right, we
can drag and drop from the
left window pane to the
appropriate areas on the
right.
Demo of a Query from BW - contd

To set the date filter, the
calendar day characteristic
can be dragged and
dropped under the filter.
Then right-clicked and
choose “Restrict”.
Demo of a Query from BW - contd

This window is the filter
selection window whether
it’s calendar date or any
other characteristic.

We choose a date and use
the blue arrow in the middle
to add it to the right
selection window.

Then hit OK, to add it to the
query.
Demo of a Query from BW - contd

Now, we are ready to save
and run the query.

Click on the green execute
button in the tool bar:
Demo of a Query from BW - contd

If only we want ourselves to
be able to see this query
going forward, we can save
it in favorites.

Otherwise, we can click on
the Roles button on the left.
We will then be presented
with the Roles areas we
have access to save things
to. In this example, we are
SD ISBA’s (aka Power
Users), thus we only see
that Role area.
Demo of a Query from BW - contd

At the bottom, enter both a
Description and a Technical
Name. The Description can
be whatever we want.

The technical name,
however, needs to ALWAYS
start with a “Z”.

Enter the values at the
bottom and click the SAVE
button.

Once we save it, the query
will execute by launching us
back into excel.
Demo of a Query from BW - contd




This gives our results window,
which will look like a regular
excel formatted spreadsheet.
(It isn’t, but it looks like one!)
As our query specified the
Sales Organization in the
Rows pane of the query
designer, this will be located on
the x-axis.
Since we only specified one
key figure (Product Revenue)
on the Columns pane (the yaxis), these will show as such:
THAT’S THE ENTIRE
QUERY!!!!
Demo of a Query from BW - contd

We could add additional
characteristics, you could format
it into a chart to present, etc.

Let’s say we want first by Sales
Organization then by Sales
Office. We could modify the
query appropriately (adding the
Sales Office characteristic to the
columns pane via the query
designer), and this would be the
result:

Notice the sales offices across
the top of the results.
Usage of the new Data Mining tools






Marketing Data Analysis
Manager’s Cockpit
Sales Out reporting
Order Flow reporting
Order Fulfillment reporting
Commission reports
Reports to be developed in the Future







Gross Profit analysis
Market Analysis
Market Segment Sales Analysis
Sales Forecasting
Budgeting
Strategic decision Making
Center Excellence performance analysis