WHITE PAPER

Business Intelligence 101

The Basics for Business
Professionals
 WHITE PAPER

Business Intelligence 101

Table of Contents

Executive Summary ................................................................................................................................................................ 3

Definition and Goal of Business Intelligence Software ...................................................................................................... 3

Questions Business intelligence (Bl) is Designed to Answer ............................................................................................. 3

Examples of Bl in Action - Sales Executives Using salesforce.com ............................................................................... 3

Why Other Solutions Cannot Match Bl for Analysis and insight ......................................................................................... 3

Traditional Enterprise Bl Software .................................................................................................................................... 3

Software as a Service (SaaS) Bl ...................................................................................................................................... 4

Overview of Business Intelligence Solutions ............................................................................................................................ 5

5 Stages of Business intelligence ..................................................................................................................................... 5

Turning Data into insight .................................................................................................................................................. 5

Why Bl is Necessary for Analysis and insight .................................................................................................................... 7

Examples of the Benefits of Shifting from Operational Application Reporting to Bl ........................................................... 10

Traditional Enterprise Bl Software Versus Software as a Service (SaaS) Bl..................................................................... 11

The 5 Stages of Business intelligence ............................................................................................................................ 14

The Benefits of Software as a Service Versus Traditional Enterprise Software ................................................................ 16

Time to insight................................................................................................................................................................ 18

Summary ....................................................................................................................................................................... 19

Evaluating Business Intelligence Solutions and Vendors ................................................................................................................ 20

The Criteria to Use in Evaluating a Bl Solution................................................................................................................ 20

The Criteria to Use in Evaluating the Bl Vendors Themselves ......................................................................................... 21

The Business Intelligence Glossary ....................................................................................................................................... 22

2
 WHITE PAPER

Business Intelligence 101

Executive Summary faster; it is also easier to follow up on interesting data

trends. For example, a BI solution for sales data like the
The Definition and Goal of Business Intelligence information found in salesforce.com might include:
Software
• Interactive Dashboards with graphs showing at-a-
Business Intelligence (BI) software takes the mass glance trends and key findings such as: Leads to
of information generated by a company’s various Pipeline to Revenue
operational data sources, integrates it, analyzes it, and • Reports that cover such key details as: Rep Activities
presents timely findings in the form of reports, alerts, vs. Pipeline Generated

and dashboards. These reports, alerts, and dashboards • Alerts that notify the executive on key changes to the
provide a more accurate, actionable view of a business pipeline or upcoming events/actions

than is possible without BI. Why Other Solutions Cannot Match BI for
The success of a BI solution ultimately is determined Analysis and Insight
by how well it helps both business and technical users Important information may be found in every operational
throughout an organization meet their mission critical area of an enterprise: sales, finance, customer service,
goals such as: engineering, marketing, and more. Enterprises have
• Achieving or surpassing revenue numbers
traditionally tried to use operational or spreadsheet
applications to keep track of the information, analyze the
• Finding opportunities to reduce costs throughout the
organization data, and present findings. Examples include:

• Maximizing profitability by identifying the most • Using an operational software program designed for
profitable customers, products, services, or programs a specific functional area (e.g., sales automation
software, marketing software, or inventory
Questions Business Intelligence (BI) is Designed management software)
to Answer • Exporting the operational data to Excel or a database
for manual analysis
A BI solution, with the right data and features, should be
able to take operational data and enable users to answer Operational programs and spreadsheets, however, were
specific questions such as: not designed to integrate, analyze, or present data in
large volumes or from multiple operational sources. As
• Which customers should I target? a result, organizations have been struggling with time-
• Which are my most profitable and least profitable consuming and costly analysis approaches that are
marketing and sales campaigns per region? fundamentally limited and don’t answer the full range
• Where are the delays in my production process? of critical business questions required to manage a
• How are my company’s key financial indicators business effectively. This is similar to trying to drive a car
trending? at high speeds with only half a headlight working.

Example of BI in Action - Sales Executive

Traditional Enterprise BI Software
Using salesforce.com
Business Intelligence solutions were originally delivered as
Many operational solutions have some form of reporting traditional enterprise software – that is, software solutions
in them. However, that reporting functionality is often that are installed on your company premises, by your IT
fairly limited - offering only a few dimensions of inquiry team or by 3rd party consultants, and managed by your
for only the data within that one system. In contrast, a IT team or consultants over time. The software
business intelligence solution allows an executive to is expensive, often requiring hundreds of thousands or
quickly and easily analyze that system’s data across millions of dollars in upfront costs. Implementing the
multiple dimensions, as well as supplement that data with software is also expensive; in addition to the cost of the
information from other key sources, such as marketing people, it often requires additional hardware purchases
and finance. Analysis and reporting is more robust and for the software to work. The implementation period

3
 WHITE PAPER

Business Intelligence 101

could be long and complicated. Maintenance and Advantages of SaaS (on-demand) BI include:
support costs are lower than upfront costs, but also can
• Shorter Time-to-Insight
be expensive.
• Less Time to Recoup costs = Faster ROI
In addition to this, a single BI implementation might • Lower Upfront Costs
actually require the purchase and integration of multiple • Lower On-going Costs
software components -- software for extracting data,
• Less Reliance and Drain on IT Resources
integrating data, storing it, analyzing it, and then
• Scalability
reporting on it. This complexity is the driver of the time
and cost. • Flexibility

Businesses highly valued the analysis that they could get

from traditional BI, so they were willing to invest in the
time, software, hardware, and human resources required
to make it happen. But these high costs also made it
diffcult for anyone but the largest companies to buy it,
and even then, those companies had diffculty expanding
the solution to realize more value.

Software-as-a-Service (SaaS) BI
In order to overcome the high cost, complexity, and time
requirements of traditional BI solutions, new Software-
as-a-Service (known as SaaS or on-demand) solutions
have emerged. Instead of purchasing and implementing
software themselves, SaaS BI customers purchase BI
capabilities as a "service" from a vendor. The software
no longer has to be implemented and maintained on
the premises, but instead runs on servers in a remote
location maintained by the vendor; the company
accesses the service via secure internet connection.
Salesforce.com is an example of a successful SaaS
solution for salesforce automation.

SaaS solutions offer signifcant benefts over traditional

software. Time to value is reduced, since SaaS requires
weeks or months to implement, instead of years. Upfront
costs are dramatically reduced, removing a lot of risk
from software purchase decision-making – there is no
software to buy, no extra hardware to buy, and far less
implementation effort involved. SaaS is typically priced
on a subscription (usually a monthly fee) or pay-for-
use basis.

4
 WHITE PAPER

Business Intelligence 101

Overview of Business Intelligence contrast, newer SaaS Business Intelligence solutions can
Solutions provide equivalent power and capability in faster
timeframes and at lower cost, with ease of use benefts
The 5 Stages of Business Intelligence for business users. By lowering the hurdles to achieving
Both traditional enterprise BI and SaaS BI involve the powerful business intelligence, SaaS solutions make it
same fve stages of taking raw data and presenting it as possible to bring business insight to a greater number of
relevant, actionable insight to users. However, traditional users and to small and medium sized organizations that
enterprise BI software and SaaS BI solutions can differ could not afford BI before.
signifcantly at each stage in terms of time and effort
required. The fve stages are: Turning Data into Insight
Business Intelligence (BI) is the process of taking
1. The Data: defning which data will be loaded into the
operational data and putting it to use in making informed
system and analyzed.
business decisions. BI software takes the mass of
2. The ETL (Extract, Transform, and Load) Engine: information kept in a company’s operational data sources,
moving the source data to the Data Warehouse. integrates it, analyzes it, and presents timely fndings in
This can be a complex step involving modifcations the form of reports, alerts, and dashboards. These
and calculations on the data itself. If this step doesn’t reports, alerts and dashboards provide a more
work properly, the BI solution simply cannot be accurate, actionable view of a business than is possible
effective. without BI. In fact, a BI solution should be able to provide
insights that are simply unknowable in a timely manner by
3. Data Warehousing: connects electronic data from any other means.
different operational systems so that the data can
be queried and analyzed over time for business
decision making.

4. Analytic Engine: analyzes multidimensional data

sets found in a data warehouse to identify trends,
outliers, and patterns.

5. Presentation Layer: the dashboards, reports and

alerts that present fndings from the analysis.

Conclusion BI takes information kept in

Business Intelligence solutions make it possible for
groups within organizations to gain actionable insight from
operational data sources,
business data, and to leverage these insights to meet
critical goals. Business intelligence solutions offer
integrates it, analyzes it,
business-focused analysis at a scale, complexity, and
speed that is not achievable with basic operational
and presents timely findings
systems reporting or spreadsheet analysis, thereby
delivering signifcant value.
in the form of reports
While traditional approaches to business intelligence and dashboards.
can provide great value, they require a very manual,
lengthy, and expensive process to implement. In

5
 WHITE PAPER

Business Intelligence 101

The Goals of BI • Financial questions

The success of a BI solution ultimately is determined by • What is the fully loaded cost of new products?
how well it helps users throughout an organization meet • What is the expected annual proft/loss based on current
their mission critical goals, such as: marketing and sales forecasts?

• Achieving or surpassing revenue numbers • How are forecasts trending against the annual plan?

• Maximizing profitability by identifying the most • What are the current trends in cash low, accounts
profitable programs payable and accounts receivable and how do they
compare with plan?
• Finding opportunities to reduce costs throughout
• Overall business performance questions
the organization
• What are the most important risk factors impacting the
• Preventing overload on IT resources by making business
company's ability to meet annual proft goals?
teams more self-reliant
• Should we expand internationally and, if so, which
• Enabling IT to service internal business customers faster
and more efficiently geographic areas should we frst target?

• Creating a 360 degree view of customers Example of BI in Action - Sales Executive Using
salesforce.com:
A Business Intelligence solution enables business teams
Many operational solutions have some form of reporting
to understand the interrelationship of data generated
in them. However, that reporting functionality is often
throughout the organization, synthesize this into
fairly limited - offering only a few dimensions of inquiry
comprehensive analysis on results, trends, and likely
for only the data within that one system.
outcomes, and understand the implications of this insight
on business objectives. BI users gain an understanding of In contrast, a business intelligence solution allows an
what has happened, what is happening and what actions executive to quickly and easily analyze that system’s
need to be taken next. data across multiple dimensions, as well as supplement
that data with information from other key sources, such
Questions BI is Designed to Answer
as marketing and fnance. Analysis and reporting is
A BI solution, with the right data and features, should be
more robust and faster; it is also easier to follow up on
able to take operational data and enable users to answer
specifc questions such as:

• Sales and marketing questions

• Which customers should l target?
• What has caused the change in my pipeline? A BI solution helps users
• Which are my most proftable campaigns per region?
• Did store sales spike when we advertised in the local paper answer specific questions
or launched an email campaign?
• What is the most proftable source of sales leads and how
has that changed over time?
and understand what
• Operational questions has happened, what
• Which vendors are best at delivering on time and on budget?
• How many additional personnel do we need to add per store is happening and what
during the holidays?
• Which order processing processes are most ineffcient? actions need to be taken.

6
 WHITE PAPER

Business Intelligence 101

interesting data trends. For example, a BI solution for 2. The Review: The sales executive reviews customized
sales data like the information found in salesforce.com reports on the region and fnds that the reps who
might include: are behind on reaching quota are focusing on
products and customer segments that have longer
• Sales Dashboards with graphs showing at-a-glance what
is happening so that the proper action can be taken.
sales cycles and lower probabilities of closing. The
Dashboards can include such crucial data as: Sales to executive drills down in the reports to determine the
Goal, Sales by Region vs. Plan, Pipeline Trend to Goal, segments and products which have the shortest sales
Products Sold by Customer Segment, and Time to Sell
cycles and highest probabilities of closing, to
by Product Line.
compensate for the shortfall.
• Actionable Sales Reports that are generated
automatically and cover such key issues as: Actual vs.
3. The Action: The executive and the regional manager
Forecast by Product, Rep, and Region; Rep Activities vs.
Pipeline Generated; Top 30 Customers (who have not review the data and refocus the reps. This puts the
purchased a specific product) by Rep, Demographic reps in alignment with the more successful reps, as
Profile of Closed Accounts vs. Accounts with The Most well as places the region in alignment with the more
Activity.
successful regions.
• Sales Alerts that notify the executive on key changes or
upcoming events/actions. Why BI is Necessary for Analysis and Insight
Here is an example of how sales executives might use In this age of rapidly expanding electronic data,
a Business Intelligence solution to help them meet enterprises have found it increasingly diffcult to effciently
their goals: or effectively leverage their data to meet business
goals. Important information may be found in every
1. The Warning: A sales executive is alerted to a key operational area of a company: sales, fnance, customer
change in the pipeline, goes to the dashboard for an service, engineering, marketing and more. Companies
at-a-glance understanding of what is causing the have traditionally tried to use their operational software
change, and notices that a region is suddenly behind programs to both run operations and analyze the resulting
on hitting its revenue numbers. data. Enterprises have tried to use operational software
programs designed for a specifc functional area (e.g.,
sales automation, marketing automation, inventory
management, ERP, or fnancial services) or export the
data to Excel or a database for manual analysis.

A BI solution helps users Operational programs, however, were not designed to

integrate, analyze, or present data in large volumes or
answer specific questions from multiple operational sources. Each of these
approaches has drawbacks, summarized in the
and understand what following table:

has happened, what

is happening and what
actions need to be taken.

7
 WHITE PAPER

Business Intelligence 101

Approach Drawbacks for Business Analysis

Reporting Primary purpose of operational software is for transactions and conducting operations, not
available within analysis of the resulting data. Reporting capability tends to be very limited in terms of the
operational variables and dimensions of analysis, requiring signifcant manual work (such as exporting data
software programs to a spreadsheet) to do analysis that business users really want.
(CRM, ERP, HR,
accounting, etc.)
Reporting functionality within the individual program is only designed for the data captured by that
program, so the user gets just a partial picture of:

• How this business function impacts the rest of the business

• How what is happening in other operational areas, in turn, impacts them

Analyzing data in an operational system can slow down the capability for which the system was
originally designed - running operations. This problem grows worse as data volume increases.

• Operational applications are designed for simply queries and to make

insert/update operations quickly
• Business Intelligence requires complex queries that span many tables
and operate on large row sets
• Making large-scale analysis demands on an operating system can crash
the system or slow it significantly

May require expensive and time-consuming integration with other operational groups’ programs
in order to achieve desired cross-functional analysis.

Excel Highly manual. Extremely cumbersome for data analysis, report presentation, and for ongoing
data updating and maintenance. Updating an analysis with new data often requires re-running the
analysis from scratch. The spreadsheet is often only easily understood by its original owner.

Smaller scale data sets only. Does not handle

the larger data volumes which BI solutions can easily handle.

Insuffcient analytic capability:

• Does not provide typical BI capabilities like slice & dice, drill down capabilities, pixelperfect
report layout
• Requires an analyst mindset and advanced Excel proficiency
• Not usable for advanced analysis by the typical business user

Integrating data from multiple operational sources requires signifcant time and skill.

Prone to error when updated.

8
 WHITE PAPER

Business Intelligence 101

Approach Drawbacks for Business Analysis

Databases Databases such as Oracle or SOL are good places for storing data, but do not provide the
breadth of analysis and presentation tools required for business analytics.

Presentation tools are lacking:

• Lack charting, pivot table, dashboard functionality

• Requires technical knowledge to craft SQL statements and present the results as a plain table
• Doesn’t address how insights are shared in the organization

Integration of data from multiple systems may require custom work.

Signifcantly slowed down by BI's large cross-functional data sets:

• Can crash the system

• As with operational applications, OLTP-type databases are designed for simply queries and to
make insert/update operations quickly
• Not well suited for BI type queries

9
 WHITE PAPER

Business Intelligence 101

Examples of the Benefits of Shifting from • After purchasing, do store customers visit the website
for add-on purchases? Which programs best incent
Operational Application Reporting to BI them to visit the site and purchase?
Even if all of the technical hurdles of operational system For example, through using website and email analytics
reporting were solved, operational teams simply need a company may fnd that low price products are
access to data from other operational areas in order to easiest to promote on the website and generate the most
fully understand what has happened, what is happening, completed orders. So the web marketing team is
and what they should do next. Here are some examples enthusiastic about promoting low cost online deals and
of how combining information from multiple applications petitions the VP of Marketing for more investment along
and functional areas can drive greater overall success. these lines.

Combining Marketing and Sales Automation Data to However, when looking at a more comprehensive picture
Increase Sales by adding in Point of Sale (POS) store data, a Business
A sales team with access only to sales force automation Intelligence system’s dashboards and reports may
(SFA) application data does not have a clear picture of demonstrate that customers’ desire to purchase seems to
the visits their customers make to the corporate website; be satisfed by buying only the promoted low-cost
the website data is owned by the marketing team. Adding product online. As a result, customers do not visit a store,
the website information into the analysis could potentially where they are more likely also to buy higher-price,
alert sales to new opportunities or to the possibility that a higher-margin products. The end result: lower enterprise-
customer is closer to making a purchase decision. wide revenue and proft.

Using this same example, the sales team may also learn Further, analysis of reports may inform marketing that
how accounts they close generate sales on the website the company maximizes revenue and proft if marketing
and whether the multi-channel activities generate more promotes higher price items on the site and then uses
or less revenue from their accounts. The more sources email marketing to promote online add-on products to
of data with which the sales team integrates their SFA customers who made their initial purchase in the store.
data (e.g., fnance, support), the greater the insight they By looking at the entirety of information available with a
receive into account behavior and potential opportunities. business intelligence solution, the VP of Marketing can

Combining Point of Sale (POS) with Email and Website

Analytic Data to Maximize Revenue and Profitability
Customers who buy in multiple channels (e.g., retail and
online) are worth more than single channel customers, Operational teams need
according to The Nielsen Company. By analyzing only
online data (e.g., email and website visit statistics), a brick access to data from other
and mortar retailer’s marketing team will not be able to
understand how to drive multichannel purchases so as to departments to see the
maximize revenue and proftability. Using only their own
data, marketing will not be able to answer such basic full picture of what is
questions as:

• What is the optimal mix of online and retail store sales

happening and what
that provide the company with the greatest revenue and
profitability? to do.
• Which marketing programs most cost-effectively drive
customers into stores?

10
 WHITE PAPER

Business Intelligence 101

make smarter decisions about the types of promotions Traditional Enterprise Software Business Intelligence
and marketing campaigns that boost enterprise-wide Traditional business intelligence solutions are delivered
sales and proftability. as enterprise software– that is, software solutions that are
installed on your company premises, by your IT team or
Traditional Enterprise BI Software Versus by 3rd party consultants, and managed by your IT
Software as a Service (SaaS) BI team or consultants over time. The software is expensive,
often requiring hundreds of thousands or millions of
The traditional Business Intelligence solutions are
dollars in upfront costs. Implementing the software is also
delivered as enterprise software. While these solutions
expensive; in addition to the cost of the people, it often
can be effective, they require high upfront costs, intensive
requires additional hardware purchases for the software
use of IT resources or 3rd party consulting services,
to work, such as hardware servers. The implementation
and hardware investment to get started. Maintaining the
period could be long and complicated. Maintenance and
system is also costly and resource intensive, thereby
support costs are lower than upfront costs, but also can
limiting the solutions only to the largest companies be expensive.
and, even then, only certain groups within those
companies. Extending or expanding the solutions is often In addition to this, a single BI “implementation” might
prohibitively expensive. In recent years, SaaS, or on- actually require thepurchase and integration of multiple
demand, Business Intelligence solutions have emerged to software components — software forextracting data,
address the cost and time requirements, as well as some integrating data, storing it, analyzing it, and then
of the feature limitations, of traditional solutions. For these reporting on it. This complexity is the driver of the time
reasons, SaaS solutions are gaining customers in both and cost.
enterprises and SMBs alike.
Here is a diagram of a typical traditional business
intelligence solution implementation:

Extract, Transform, Load Business Intelligence Users

Finance Data Operational Dashboards

Data Store
7+ Servers
CRM Data OLAP 2+ Vendors
Logical Model Ad Hoc 9+ Months
Operations Data
Data Mart/
Warehouse
More Data Reports

11
 WHITE PAPER

Business Intelligence 101

Businesses highly valued the analysis that they could as a "service" from a vendor. The software no longer has
get from traditional BI, so they were willing to invest in to be implemented and maintained on the premises, but
the time, software, hardware, and human resources instead runs on servers in a remote location maintained by
required to make it happen. But these high costs also the vendor; the company accesses the service via secure
made it diffcult for anyone but the largest companies internet connection. Salesforce.com, NetSuite, and
to buy it,and even then, those companies had diffculty RightNow are examples of successful SaaS solutions for
expanding the solution to realize more value. Customer Relationship Management (CRM).

Some recent analysis conducted by independent analysts Lower Upfront Costs and Resource Requirements
shows the challenges faced by traditional enterprise SaaS solutions offer signifcant benefts over traditional
software BI implementation: software. Time to value is reduced, since SaaS requires
weeks or months to implement, instead of years. Upfront
• Average implementation time1:
costs are dramatically reduced, removing a lot of risk from
• 17 months in total; 5 months to deploy the frst usable software purchase decision-making – there is no software
analytic application
to buy, no extra hardware to buy, and far less
• Mean annual expenditure on BI software2 : implementation and ongoing effort involved. SaaS is
• $1.1 million for companies with >1,000 employees typically priced on a subscription (usually a monthly fee)
• Project success rate3 : or pay-for-use basis. For this reason, SaaS solutions
• 31% success rate, at best often are paid for as an operating expense instead of a
capital expense, making it easier for SaaS BI projects to
• Ability to meet needs – getting the right data to the
right person4 : get approval.
• Only 36% are confdent that reports and dashboards
Integrated, Automated Solutions – Faster and
deliver the right data, to the right person, at the right time
Easier Implementation
Software as a Service (SaaS) or On-demand One of the keys to the lower cost of SaaS BI is the level of
Business Intelligence integration and automation that a SaaS BI vendor users.
Software as a Service business intelligence (known as A fully integrated and automated vendor has already done
SaaS BI or on-demand BI) emerged in response to the the BI integration for you, and automated some of the
complexity, resource drain, lengthy timeframes, and most challenging and time consuming components. Here
prohibitive cost of traditional enterprise software. Instead is an example of how an integrated solution would look, in
of purchasing, integrating, and implementing software contrast with the traditional diagram from the prior section:
themselves, SaaS BI customers purchase BI capabilities

Users

Finance
Dashboards
Data

CRM Data
Operational
Data Store Unified 1 App Server
Operations
Logical 1 Vendor
Data Model Ad Hoc 1+ Months
Data Mart
More Data
Warehouse

Reports

1 DM Review & lDC Business lntelligence Survey, October 2006, 2004 (Material on implementation time not available in 2006 survey) 2 lnfoWorld & lDC Business lntelligence
Survey, October 2007 3 DM Review & lDC Business lntelligence Survey, October 2006, 2004 (Material on implementation time not available in 2006 survey) 4 lnfoWorld & lDC
Business lntelligence Survey, October 2007
12
 WHITE PAPER

Business Intelligence 101

Lower Ongoing Costs and Resource Requirements

SaaS also offers ongoing cost and value benefts. SaaS
can be rolled out to a limited project at frst, so that
value and effectiveness can be quickly demonstrated.
Then, it can be rolled out on an iterative basis, allowing
other data sources and users to be added easily later. In
this way, companies can prove to themselves the value
before making extensive commitments, and high priority
projects are easily tackled frst, with less additional
effort required to scale or extend a deployment. SaaS
customers are often upgraded automatically by the
vendor, so that individual companies no longer have to
concern themselves with the time and cost of rolling out
software upgrades themselves.

Flexible and Easier to Use – Business Users Can Serve

Themselves
SaaS solutions also tend to be more user friendly than
traditional BI solutions, which were built expressly for the
use of analysis experts within an IT organization. While
the traditional approach allowed highly complex analysis
to be done, it made IT into a critical bottleneck for
achieving business reporting, since the number of trained
users were limited in number and reported through the
IT organization instead of a business function. SaaS
solutions are often created to serve both business users
and analysis experts, so that business users can receive
the robust reporting and data exploration that they need,
while experts can tackle more complex and involved
inquiries.

13
 WHITE PAPER

Business Intelligence 101

The 5 Stages of Business Intelligence stage of moving the source data to the Data Warehouse.
Both traditional enterprise BI software and SaaS BI If this doesn’t work properly, the BI solution simply
solutions involve the same fve stages of taking raw data cannot be effective.
and presenting it as relevant, actionable data to users.
Traditional method: Technical teams manually map
However, traditional enterprise BI software and SaaS BI
source data and create defnitions, map to tables, and
solutions can differ signifcantly in each of the fve stages then map to a data warehouse production environment.
in regards to time, cost, complexity, and resources
This process is repeated for each BI initiative within the
required.
organization.
The fve stages are:
Integrated SaaS method: The entire process can be
1. The Data automated to create staging and warehouse tables.
2. The ETL (Extract, Transform, and Load) Engine
There are two other key issues to address at this stage:
3. Data Warehousing
1. What type of data can be inputted into the system?
4. Analytic Engine
The types of data BI solutions can utilize should
5. Presentation Layer
encompass any data that can be captured and a
The frst four stages in the BI process comprise the user wants to track. A BI solution should enable
“back end” processes and are traditionally driven by the enterprises or groups to upload fles of large size
technical team(s). Most business users primarily interact and integrate with any solution that can export data
with the front end or Presentation Layer. such as:

The Backend of BI– Inputting, Integrating, and Analyz- • Data extracts from operational applications such as
CRM or ERP
ing the Data
applications
Traditional enterprise software and SaaS solutions take
• SQL, Oracle, or other relational databases
different approaches to the backend processes and
• Flat file databases
require different levels of resources and timeframes.
• Comma-delimited files (the standard “generic” file
Stage 1: The Data format)
The frst step in the BI process is to defne which data • Excel or other spreadsheets
will be loaded into the system and analyzed. A key issue
2. Does the solution have automated data refresh?
to address at this stage is whether to defne all of the
data upfront or identify it on an iterative basis. Now it is Optimally the solution has automated data refresh.
possible for an enterprise or group to start with a subset of Otherwise, the data will need to be uploaded
their overall data and add more in as they gain access to it manually. A Business Intelligence vendor should be
or simply decide to include it. Traditional method: The able to help develop an integration plan and, directly
overall data structure must be defned up front, or through partners, offer integrations services.
requiring companies to fully defne which data they want
to use before they can get started with the rest of the
backend process. Integrated SaaS method: Companies
can start with available data and without defning up front
all of the data sources. Then, additional data sources can
be added on an on-going basis.

Stage 2: The ETL Engine

The ETL (Extract, Transform, and Load) Engine is the

14
 WHITE PAPER

Business Intelligence 101

Stage 3: Data Warehousing The front end of BI - tools for acting on data analysis
Data Warehouses connect and keep electronic data - Stage 5: Presentation Layer
either in one large system or multiple connected systems BI also includes tools for presenting fndings from the
— so that the data can be retrieved and analyzed. analyzed data:

Traditional method: Traditionally, this is a very frustrating • Interactive Dashboards. A dashboard is a set of high-
level reports on key metrics, typically for managers. There
stage for companies because they have been forced to
may be multiple reports on a single dashboard, much the
choose between two sub-optimal approaches: same way that a car’s dashboard has multiple gauges and
displays on it. With a dashboard, users can gain an at-a-
1. Multiple data marts and warehouses fed by multiple glance understanding of key trends and metrics.
ETL processes, limiting the scope of any one Dashboards can be customizable to work for anyone in
an organization, from a sales rep or frontline operations
application OR manager to a middle manager or senior executive.
2. One Enterprise Data Warehouse (EDW) that An “interactive” dashboard allows users to take those
provides scale but severely limits the flexibility of dashboard reports and flter information to more deeply
analyze trends and results, or to “drill down” into deeper
delivering new analytical applications.
and more detailed analysis of the data. That is, by clicking
Integrated SaaS method: Data warehouses created on on the particular reports or results, they can explore more
detailed information to fnd root causes
the fly for each business intelligence project/application.
of results.
No reliance on a single EDW approach.
• Customizable Reports which can present high-level
findings as well as enable a user to drill down to find
Stage 4: Analytic Engine
specific details. Most BI systems either come with report
Once data is in the system, the next step is to conduct templates and/or provide the capability to create and
analysis on it. Traditionally, this stage has required customize reports.
signifcant engineering resources to make the analysis • Alerts notifying users to changes selected as key to
work for the company's data warehouse. Now, SaaS meeting user goals. Alerts can be set to warn users on an
imminent event, changes to data, or that new data needs to
BI solutions can offer an automated approach to the
be entered into the system.
analysis stage and move it online with an Integrated
OLAP (On-Line Analytical Processing) Engine. Traditional method: Reporting packages need to be
implemented from multiple solutions, since only some of
Traditional method: IT develops another set of metadata them provide pixel perfect reporting, while another
to navigate universes, subject areas, and cubes or to provides banded reporting, and so on. IT needs to build
build custom cubes for analysis. and publish unique dashboards and reports for each
different type of user in the organization.
Integrated SaaS method: Metric and dimensional
metadata can be automatically generated and Integrated SaaS method: A fully integrated reporting
maintained. package is ideal, so that IT does not have to integrate
multiple systems. Advanced reporting techniques such as
banded reporting, drag and drop formatting, and pixel-
perfect reporting are all included in one solution.

15
 WHITE PAPER

Business Intelligence 101

The Benefits of Software as a Service Versus

Benefts of SaaS vs. Traditional BI Traditional Enterprise Software
Software include:
Less Time to Recoup costs = Faster ROI
• Quicker deployment, faster changes, A faster “time to value” allows for a quicker return on investment.
and less ongoing maintenance and
customization SaaS BI solutions offer quick deployment and easy changes.
• Pricing that is easier on budgets Unlike traditional enterprise BI software implementation, which
due to far lower upfront costs and can take 12 to 18 months or longer, SaaS Business Intelligence
subscription pricing
solutions can typically be up and running in just a few weeks.
• Lower on-going costs
This is because there is no additional hardware to install and no
• Low risk and high reward thanks to database to set up. Also, on-demand solutions can start with just
lower costs, limited commitments,
the available data, with other data added in as available.
and lower resource requirements
With the solution up and running sooner, companies can start
• Easier budget approval for a monthly,
predictable expense
reaping a return on their investment more quickly.

• Time to insight is shorter Ongoing maintenance and customization is also faster and
• Ease of sharing reports and easier. Since all of the hardware and infrastructure is maintained
dashboards by the vendor, all software upgrades and architecture changes
• Less impact on IT resources, so that are handled by the vendor and delivered to the customer
IT can provide greater and faster
automatically. Customization and changes to the reporting and
service
analysis tools can be handled by end users with limited, if any,
• Companies can easily scale usage
use of IT resources.
from group to enterprise level
• Greater flexibility for business users Lower Upfront Costs
and IT
On-demand pricing models and lower implementation costs
SaaS BI solutions usually require a mean dramatically lower upfront costs.
few weeks to get started rather than
traditional BI software’s many months Pricing is easier on most companies’ budgets because on-
to years.
demand solutions are usually sold as a monthly subscription.
This reduces upfront costs and often allows companies to pay
for the solution from the operating budget, which is easier to get
approval on, than from the capital expenditures budget.

SaaS BI (also known as on-demand BI) providers manage all of

the backend systems for their service as well as host their
applications, so customers are spared the upfront hardware and
setup costs associated with having a BI solution deployed
- there is no hardware to buy, no software servers to purchase
and set up. Low cost also makes these solutions less risky to
implement, generating a higher ROI.

Lower On-going Costs

No fxed investment, lower maintenance, and fewer hardware
requirements.

16
 WHITE PAPER

Business Intelligence 101

The SaaS subscription fee provides an all-in cost for Time to Insight is Shorter
the right to use the solution, ongoing maintenance, Faster implementation, built-in templates, and interactive
and support. This subscription is usually based on dashboards enable users to reap the benefts much
the number of users who will interact with the system, more quickly.
the volume of data analyzed, and the level of support
Less time is needed to be operational with a SaaS
required. This pricing approach ensures that customers
(on-demand) solution. Not only is it easier to generate
pay only for what they need, as they need it – so the
templates with on-demand solutions, they may come with
customer retains fnancial control of the project and
automatically generated report and dashboard templates
maintains the flexibility to scale up as their needs expand.
so that users can immediately gain insight before setting
Because the customer’s solution is running on a shared up a single report or dashboard; reports and dashboards
infrastructure, this increased fnancial control and can then be customized after users have had a chance
flexibility comes with lower ongoing costs. For most to use the system. Interactive dashboards and reports
organizations, the cost to deploy a SaaS BI solution with can also lower the reliance on technical staff and make
subscription licensing will be far lower than the cost of business teams more self-reliant without turning them
buying a conventional on-premise solution; with SaaS into programmers.
there are no servers to maintain, no patches to install, no
360- degree Communication
ongoing software maintenance to pay and minimal
Easily share data and reports, even with customers
IT resources required, according to IDC. Additionally,
and partners.
ongoing maintenance and upgrades are handled by the
vendor, minimizing cost and risk for the customer. SaaS solutions are deployed over the internet, so that
users can easily share data with others, both inside and
Low Risk – High Reward
outside the customer organizations. This makes it easy
The upside of insight without the risks of traditional
for users to share insight with remote offces, suppliers,
enterprise software.
feld sales organizations, partners, and customers.
As a monthly subscription service, SaaS (on-demand)
In addition, since SaaS solutions can leverage data from
solutions offer a low-risk opportunity to acquire business
anywhere in the world, users can integrate data from
intelligence without high costs, onerous commitments, or
multiple sources in different parts of the world, from
high resource requirements. The rewards of greater
other internal business units, and also from suppliers and
business insight, improved effciency, higher productivity
partners in any company's extended value chain. This
and lower costs make this a winning strategy for
is especially important for any frm with multiple sites in
businesses of all sizes.
different physical locations.
Easier Budget Approval
Less Reliance and Drain on IT Resources
Lower entry costs and predictable ongoing expenses.
Everyone wins when business teams can conduct most
On-demand solutions are paid for as subscriptions; their of their own analysis.
license cost is a monthly, predictable expense and not a
Business teams require far fewer IT resources to integrate
large, up-front payment for licenses as with conventional
data, confgure the system, or generate templates. IT
software. Therefore, on-demand solutions can be
can, in turn, provide greater service to business teams
typically purchased as an ongoing operational expense
by being able to help them faster without signifcantly
(OpEx) rather than a capital expense (CapEx). This makes
impacting IT resources. Because SaaS BI solutions can
these solutions particularly attractive in years, like this one,
be set up in a fraction of the time of traditional solutions,
where CapEx budgets are under immense scrutiny.
the time and resources required by IT to deploy a fnished

17
 WHITE PAPER

Business Intelligence 101

solution are dramatically reduced. Since SaaS BI simply not ready access to it, the implementation period
vendors maintain and upgrade the system and hosting will be lengthened.
servers, IT is also freed from much of the normal on-
going workload associated with traditional enterprise Most business teams, fortunately, fnd that they have
software. ready access to enough data for a SaaS solution to give
them meaningful insight. Solution providers should be
Scalability able to help enterprises prioritize their data needs to
Easily scale from group to enterprise level solutions. determine whether the readily available data is suffcient
for business purposes.
Software as a Service (SaaS) solutions are designed to
support a large number of customers simultaneously, with The amount of automation of back-end processes
capacity to spare. This means that any individual Much of the back-end processes (ETL, OLAP, and Data
customer can quickly expand their on-demand solution by Warehousing) can be automated to signifcantly reduce
simply requesting a larger account size or more users. implementation time, costs, and resource drain.

Unlike on-premise solutions, customers can expand their The solution’s Extract, Transform and Load
on-demand solution without having to buy and deploy (ETL) capabilities
more hardware or install different, larger software Advanced ETL solutions make it easier to pull multiple
servers. Since the vendor is the one responsible for types of data into the system and run the necessary
capacity, organizations can begin with a small number of pre-calculations (transformations) that might be required.
users and a small set of data, easily scaling to an If the solution does not readily accept all necessary
enterprise level platform, or anything in between. data formats, then expensive and time-consuming
customization may be required.
Flexibility
Ouickly adapt to changing business needs. The ease of generating and customizing reports
and dashboards
Unlike traditional solutions, SaaS BI solutions can be Reports and dashboards that are automatically created
changed easily, so non-technical users can quickly add can save signifcant time and enable users to start
new reports and dashboards, new data sources, and new gaining insight immediately; over time, dashboards and
analyses. With traditional BI solutions, such changes reports can be customized to better ft needs. Also, some
could take weeks or months, and involve signifcant IT solutions provide the ability to create templates, which are
resources. easily reused and shared, dramatically reducing report and
dashboard creation time.
Time to Insight
With traditional enterprise software applications,
A SaaS BI solution will usually require just a few weeks
customized dashboards and reports have required
to get started, rather than the many months to years
signifcant IT resources to produce.
involved with getting traditional BI software live.
Responsiveness of the solution provider
Key factors that determine how long it will take a
As with any other product or service, the responsiveness
Business Intelligence solution to become operational
of the vendor is crucial to decreasing time to insight, as
include:
well as helping an enterprise optimize its use of BI. A
Access to the data series of factors have to be considered such as: planning
With SaaS, any size of entity can start analyzing its capabilities; timeliness of response to fxes and requests;
available data and, over time, input data from additional and quality of work. Please see "The criteria to use in
sources. However, if specifc data is needed and there is evaluating SaaS BI vendors” for more information.

18
 WHITE PAPER

Business Intelligence 101

Summary
Business Intelligence solutions help organizations gain
actionable insight from their business data in order to
meet business goals. Most operational groups already
have some way to utilize some information within their
domain, typically through spreadsheets, databases, or the
limited reporting available in an operational solution like a
CRM solution. However, business intelligence
solutions help to provide the greatest business insight,
since they can easily analyze larger amounts of data, can
integrate data from multiple areas of the business, are
easier to use and update for complex calculations, and
have a way of easily distributing compelling reports and
dashboards.

Business intelligence solutions are available either in the

traditional manner, as an on-premise software solution,
or as an integrated SaaS (or on-demand) solution. While
powerful, traditional business intelligence can require
signifcant upfront costs, lengthy implementations, and
high resource demands for hardware and IT services.
SaaS Business Intelligence solutions are a more modern
and flexible approach, with signifcant value benefts.
SaaS BI can be implemented quickly, in weeks rather
than months, has lower upfront costs, and signifcantly
lower resource requirements.

Business intelligence solutions are only growing

in demand as available data rapidly expands and
companies seek analysis and insight that can give them a
competitive edge in a diffcult economic environment.
SaaS BI in particular is seeing signifcant traction among
SMBs and enterprises alike, due to their dramatically
improved value proposition.

19
 WHITE PAPER

Business Intelligence 101

Reporting and Dashboards

Evaluating Business Intelligence
• Does the solution offer a full range of analysis and
Solutions and Vendors reporting such as: operational reporting/banded
reporting, dashboards, OLAP/slice-dice/pivot, and
The Criteria to Use in Evaluating a BI Solution alerts?
There are many offerings that fall under the banner of • How customizable are reports and are there report
business intelligence, and so it can be confusing for a templates to help you get started?
company to determine which particular solution will best • How easily can dashboards be customized? If so, do the
meet their needs, whether that solution is a traditional, dashboards display at-a-glance the information you
need to track?
on-premise one or a SaaS BI solution.
• How actionable is the data? Another way to look at this
Here are some questions to ask in evaluating BI solutions: is – Is the analysis easily understandable and do you
readily understand what action you need to take based
on the findings?
Back-End Process Automation
• Does the solution automate much of the back-end Data and Application Security
processes (ETL, OLAP, and Data Warehousing) to • Are the applications providing access to your data
reduce time-to-insight, costs and resource drain? secure? Make sure they are designed from the ground
up to protect the security of your information.
Data Upload and Integration
• From which types of data sources can you analyze
information: flat files, Excel documents, SQL databases,
operational application data from CRM and web analytic
solutions, etc?
• Are there connectors to popular business applications
- such as salesforce. com and Google Analytics – that
make it easier to connect to data sources and control
which types of data get included in analysis?
• How easy is it to connect the SaaS or on-demand
solution with your data sources? How easy is it to
combine data from multiple sources together?

Usability and Manageability

• Is the solution a fully integrated solution or a patchwork
quilt of legacy and newer solutions?
• How much configuration does the system require to
publish complete, accurate, and current reports?
• Is the overall interface intuitive and easily usable - with
training - by nontechnical members of the business
team? A well-designed BI solution will provide left brain
understanding to right brained users.
• Can you control with whom you share your information
and what they can see?
• Once you’ve uploaded data, does the system
automatically generate some reports to get you
started?
• Can you invite your colleagues to have access to your
most recent reports whenever they want; can they
monitor data trends over time without having to locate
old emailed reports?

20
 WHITE PAPER

Business Intelligence 101

The Criteria to Use in Evaluating the BI

Vendors Themselves
Besides evaluating a vendor’s solution(s), research the
vendor’s:

References
• Are there reference customers who are willing to
talk with you? If so, ask them about the vendor’s
responsiveness and commitment to customer service.

Professional Services
• Can the vendor understand your specific situation?
• Does the vendor offer a full-range of services to help
you optimize your use of Business Intelligence?
Services can include: strategic planning; project
management; data integration; and analytic guidance.
• Does the vendor provide training that enables you and
your team to use the solution as a knowledgeable user
who can use it to achieve the goals for which you
purchased the system?

Pricing
• What pricing plans does the vendor offer?
• Are there any upfront fees?
• What are the ongoing costs? Is it an all-in cost for the
right to use the application service, maintenance, and
support? If not, what are the outlying costs?
• How do changes in data volume or number of users
impact the cost?

Security
• Physical Security: How secure is the hardware
containing the customer data?
• System Security: How secure are the networks and
hardware containing the data? Are they hardened and
tested against attack?
• Operational Security: Are the physical security and
network environments operated in a secure manner?
• Reliability: Is the data available when you need it?

21
 WHITE PAPER

Business Intelligence 101

The Business Intelligence Glossary Dimensions

Alerts The natural groupings for measurements such as: Years,
Automatic notifcations and/or reports that are triggered Months, Product Categories, Sales Regions, etc. For
by the data itself, such as a sudden shortfall in a sales example, one might want to know the monthly sales
forecast or an inventory stockout. Alerts can be set to by product.
warn users on an imminent event, changes to data, or
ETL Engine
that new data needs to be entered into the system.
ETL stands for Extract, Transform, and Load. The ETL
ASP Engine moves the source data to the Data Warehouse,
ASP stands for Application Service Provider– an sometimes performing calculations or “transformations”
approach to software rental that gained in initial on the way.
popularity in the 1990’s and has faltered since then,
Hierarchies
due to inherent limitations. ASPs may be thought of
Inherent in dimensions is the notion of hierarchy.
as a vendor that hosts a particular implementation of
Consider Time for example.There are many units of Time
software for you. While this may sound like SaaS, it is in
that one can be used to measure sales. One could
fact dramatically different, since it does not take away
the issues with upfront costs, lack of integration, time look at yearly sales, quarterly sales, monthly sales,
intensive customization, and other issues associated with weekly sales, daily sales or even hourly sales. It is
implementing your own software. SaaS has signifcant convenient, therefore, to think of all of these units as
cost and feature benefts over ASPs. levels in a hierarchy of Time.

Cloud Computing Integrated OLAP (On-Line Analytical Processing)

The use of the web to access technology-enabled Engine
services of any type. A key requirement, according The solution component that analyzes data across
to research and analysis giant Gartner, is that it be multiple dimensions, allowing users to ask complex
“massively scalable.” A cloud computing customer can business questions and get rapid answers.
easily scale up (or down) and change its monthly costs to
meet changes in demand without having to invest in or Interactive Reports
manage new servers (or sell excess servers.) Interactive reports enable users to customize the
presentation data. Interactive features include the abilities
A SaaS/on-demand solution is one of the types of cloud to sort, apply flters, defne computations, create charts
computing providers. and select columns and rows. Interactive reports can also
enable users to control fle formats.
Dashboards
Graphical presentations of data analysis that give users an Measures/Facts
at-a-glance understanding of key trends and metrics. Measures or facts are measurements of business
These usually take the form of multiple reports and charts data. Measures can include things like Revenue, Sales,
on a single web page or series of web pages. Assets, Number of Orders, etc. Measures, in turn, can be
analyzed across Dimensions.
Data Warehousing
A repository of a group or organization’s electronically OLAP
stored data, typically designed to facilitate reporting See "Integrated OLAP (On-Line Analytical Processing)
and analysis. Engine”

22
 WHITE PAPER

Business Intelligence 101

On-demand
See “Software as a Service”

Presentation Layer
Business Intelligence solutions provide users with
a set of tools that present fndings from the mass of
cross-functional data loaded into and analyzed by the
BI system. The tools include: interactive dashboards,
customizable reports, and alerts.

Software as a Service (SaaS)

Software as a Service normally is referred to by its
acronym, SaaS. The terms SaaS and "on-demand"
are normally used interchangeably. Either term, SaaS
or on-demand, refers to a solution built to be hosted
and delivered to customers securely over the
internet, removing the burdens of on-premise software
implementation and maintenance expense, Customers
securely access their solution and information from
their internet browsers. Salesforce.com, NetSuite,
and RightNow are examples of well-known SaaS
solution providers. This approach enables the sharing
of application processing and storage resources in a
“one-to-many” model where a variety of customers can
share the processing and storage, thus reducing the cost.
SaaS or on-demand solutions are typically paid for on a
subscription basis.

23