Analytics, choosing it

We observe many BI Project Sponsors clearly asking for an Analytics Package implementation to meet business needs; the benefit is that it saves time. By deciding on an analytics package we can get the application up quickly and comes with all typical benefits of a ‘buy’ solution against a ‘build’ solution.

So what are the key parameters that we need to look for in choosing an Analytics Package. The following would be the points to consider in choosing an Analytics Package, in the order of importance.

1.The effort to arrive at the right data model for a BI system is huge and as well quite tedious, so a comprehensive ‘Data Model & Metrics, Calculations’ from the package is very important.

2.The flexibility and the openness in managing Data Model is also very critical, some of tools to manage the data model elements that can be looked for are

• Ability to browse the data elements and its definitions
• Support for customization of the data model without getting back to the database syntax
• Auto Source System profiling and field mapping from the source systems to the data model
• Enabling validation of data type, data length of the data model against the source system field definitions
• Means to ensure that customization of the data model in terms of field addition doesn’t happen when a similar element exists
• Availability of standard code data as applicable to the functional area
• Supporting country specific needs in terms of data representation

3. ETL process for a BI system is also a major effort. Though the absolute effort of pulling the data and making it available for the package in the required format cannot be avoided, availability of plug-ins that can understand the data structure from typical systems like ERP would save good amount of effort.

4. Availability of ETL process for typical data validation as part of ETL is also a must; integration with any data quality product would be valuable

5. Ability to support audit and compliance requirements for data usage and reporting

6. Integration of the package with industry specific research data from vendors like D&B, IMS etc to enable benchmarking the performance metrics against industry peers/competitors

7. Customizable Security Framework

8. Semantic layer definition with formulas, hierarchies etc

9. Ready to use Score Cards and dashboard layouts

10. Pre built reports and portal

Often all the pre delivered reports under go changes and are almost completely customized when implemented. So availability of a larger list of reports itself doesn’t mean a lot since most of the reports would be minor variations from one other. Certain compliance reports would be useful when it comes along with the package; these would be published industry standard report formats.

Definitely an evaluation phase to test the Analytics products capability on a sample of the data before choosing it is a must, the above ten points would the evaluation criteria during this exercise.

Business Process for BI Practitioners – A Primer

Business Intelligence has a fairly wide scope but at the fundamental level it is all about “Business Processes”. Let me explain a bit here.
BI, without the bells and whistles, is about understanding an organization’s business model, its business processes and ultimately find the reason (analytics) and way to optimize the processes. The actions are carried out based on informed judgments (aided by BI), to make the organization better in whatever endeavor it has set itself to accomplish.

Assuming that BI practitioners are convinced that understanding business process is critical to their work, let me delve a bit into the basics of it.

1) What is a business process? (As a side note, one of the best explanation for business models is given by Joan Magretta in her book ‘What Management Is”)

Business processes are set of activities involved within or outside an organization that work together to produce a business outcome for a customer or to an organization. The fact is that for an organization to function, there are many outcomes that are required to happen on a daily basis.

2) What are BPM Tools?

Business Process Management (BPM) tools are used to create an application that is helpful in designing business process models, process flow models, data flow models, rules and also helpful in simulating, optimizing, monitoring and maintaining various processes that occur within an organization.

3) The Mechanics of Business Modeling

Business Process Modeling is the first step, followed by Process Flow Modeling and Data Flow diagrams. All these 3 diagrams and associated documentation will help in getting the complete picture of an organization’s business processes. Brief explanation of these 3 types are given below:

a) In Business Process Modeling, an organization’s functions are represented by using boxes and arrows. Boxes represent activities and arrows represent information associated with that activity. Input, Output, Control and Mechanism are the 4 types of arrows. A box and arrows combination that describes one activity is called a context diagram and obviously there would be many context diagrams to explain all the activities within the enterprise.

b) Process Flow Modeling is a model that is a collection of several activities of the business. IDEF3 is the process description capture method and this workflow model explains the activity dependencies, timing, branching and merging of process flows, choice, looping and parallelism in much greater detail.

c) Data Flow Diagrams (DFD) are used to capture the flow of data between various business processes. DFD’s describe data sources, destinations, flows, data storage and transformations. DFDs contains five basic constructs namely: activities (processes), data flows, data stores, external references and physical resources.

Just like the data modeler goes thro’ conceptual, logical and physical modeling steps, a business process modeler creates the Business Process Models, Process Flow Models and Data Flow Diagrams to get a feel for the business processes that take place within an enterprise.

Thoughts for BI Practitioners:

1. Consider viewing BI from the point of optimizing business processes
2. Might be worthwhile to learn about Business Process Modeling, Process Flow Modeling and Data Flow Diagrams
3.  
4. Understand the working of BPM tools and its usage in the enterprise BI landscape
5. Beware of the acronym BPM. BPM is Business Process Management but can also be peddled as Business Performance Management.
6.  
7. My view is that Performance Management is at a higher level, in the sense, that it is a collective (synergistic) view of the performance of individual business processes. A strong performance management framework can help you drill-down to specific business processes that can be optimized to increase performance.

What is “Safe to Bet On” in Business Intelligence?

While the phrase “Safe to Bet On” is an oxymoron of sorts, it is that time of the year where we first look at the past, derive some insights and look forward to what the future has in store for us. I have no doubts that 2009 will be doubly interesting for BI practitioners as compared to 2008.
Having said that, I decided to do a bit of introspection to figure out what skills (can also be read as competencies) should I be looking at to stay relevant in the Business Intelligence world far into the future, say at 2020. Hopefully that resonates with some of you.

Let me first try and get down to defining the skills required for Business Intelligence and Analytics. The trick here is to stay “high-level” as any BI person will acknowledge the fact that one we get down to look at the trees (rather than the forest), the sheer number of skills required for enterprise level BI can get daunting

Taking inspiration from the fact that any business can be condensed into 2 basic functions, viz. Making & Selling, I propose that there are 3 key skills that make for successful BI

Skill 1 – Business Process Understanding: If you are a core industry expert and can still talk about multi-dimensional expressions, that’s great! But most BI practitioners have their formative years rooted on the technology side and have implemented solutions across industries. The ability to understand the value-chain of any industry, map out business processes, identify optimization areas, translating IT benefits to business benefits are the key sub-skills in this area.

Skill 2 – Architecting BI Solutions: This skill is all about answering the question of “What is the blue-print” for building the Business Intelligence Landscape in the organization. Traditionally, we have built data warehouses & data marts either top-down or bottom-up, integrated data from multiple sources into physical repositories, modeled them dimensionally, provided ad-hoc query capability and we are done! – NOT ANYMORE. With ever increasing data volumes, real-time requirements imposed by Operational BI, increased sophistication for end-user analytics, the clamor for leveraging unstructured data on one hand and the advent of On-Demand Analytics, Data Mashups, Data Warehouse appliances, etc., there is no single best way to build a BI infrastructure. So the answer to “What is the blue-print?” is “It depends”. It depends on many factors (some of which are known today and many which aren’t) and the person / organization who appreciates these factors and finds the best fit to a particular situation is bound to succeed.

Skill 3 – BI Tools Expertise: Once a blue-print is defined and optimization areas identified, we need the tools that can turn those ideas into reality. BI practitioners have many tools at their disposal straddling the entire spectrum with excel spreadsheets at one end to high-end data mining tools at the other extreme. If you bring in the ETL & data modeling tools, the number of industry-strength tools gets into the 50s and beyond. With convergence of web technologies, XML, etc. into mainstream BI, it probably makes sense to simplify and say “Anything you imagine can be done with appropriate BI tools”. “Appropriate” is the key word here and it takes good amount of experience (and some luck) to get it right.

In essence, my prescription for BI practitioners to stay relevant in 2020 is to be aware of developments on these 3 major areas, develop specific techniques / sub-skills for each one of them and more importantly respect & collaborate with the BI practitioner in the next cubicle (which translates to anywhere across the globe in this flat world) for he/she would bring in complementary strengths.

Read More About  Safe to Bet On

Business Intelligence Challenge – Product Upgrades & Migrations, Validation – 5

Once the code has been moved to the target platform (Moving the Code), whether it’s an upgrade to a newer version or migration to another newer platform, the next step is to validate the objects moved.

Validation Process involves verification or testing of the objects in the target platform to ensure that they deliver the same output as the older objects in the source platform.

Validation is a key process by which the migration or upgrade process is certified as successful, it’s usually laborious and a time consuming process. Let us see how the Validation Process can be broken into different steps and automated for saving time and for improved accuracy. We can look at the Validation process to encompass three steps, they are

• Metadata Validation
• Run Validation
• Output Validation

Metadata Validation involves comparison of the metadata definitions between the existing source environment and the target environment. This requires that the metadata of the source and the target environment be captured for the comparison.

Steps Involved:

• Capture the source metadata into a relational structure, as part of Object Consolidation we would have captured the source metadata
• Capture the target platform metadata in a similar way into a relational structure
• Run SQL queries to automate the metadata comparison process

Metadata Comparison would be done at the level of semantic layer definitions and individual reports. Let us take the case of metadata comparison between two semantic layers, in case of Business Objects; Universe is the semantic layer definition. After an upgrade from an older version of Business Objects to its newer version, the first level of metadata validation between the universes would be to check whether the object counts between the universes match like the classes, the objects, the filters and then further comparison on their definitions.

If there are any differences when comparing the definitions and if they fall within the known differences between the two versions (source & target) then they are good else would require code fixing in the upgraded object.

Since we always try to validate the reports by what it gives as output, the validation process is limited by the data fed in; we could miss scenarios of a filter clause not being tested. Metadata Validation can overcome the limitation in data preparation for different scenarios for testing. If a report passes through a Metadata Validation expectation then we could 100% say that the report has upgraded or migrated effectively.

Benefits:

• Sets up a strong base on the metadata understanding, as the objects between different platforms has to be mapped and the bridges gaps identified to run automated metadata validation
• Improved accuracy in the validation process, overcomes the limitation in data preparation
• Enables determining issues without running the report against the data

Run Validation is to perform a dry run of the reports in an automated way to determine whether the reports run (open) successfully or not.

When we give a report to a tester, the first activity he would perform is to run the report and if it doesn’t go through the problem is reported or analysed further. We try to foresee this problem in an automated way.

Steps Involved:

• Have scripts to invoke the reports in batch mode, as soon as the objects are upgraded invoke(open) all the upgraded reports in the batch mode
• Capture the errors while opening/running the report into a log
• Classify them into two categories ‘reports that ran’ and ‘reports that failed’

Some reports could fail to open because of incorrect connection details, some due to object not found etc. This process of quick run in an automated way enables to locate the failure reports immediately and also help determine the reason for the failures in one go. Limiting the data input should be considered while invoking the report.

Benefits:

• Saves time in determining errors due to report opening or running
• Enables building a common solution for the code fixing team, as the ‘run errors’ are consolidated

Output Validation, is to validate the output delivered by the reports. There are two levels of output validation; they are Format Validation and Data Validation.

Format Validation is to check on the format of the data presented like font size, colour, bold, label location etc which doesn’t relate to the data value.

Data Validation is to check cell by cell the data value content between the two reports.

Steps:

• Run the source report and export the output data to excel/word
• Run the target report and export the output data to excel/word
• Compare the outputs for the format and the data

The best means of comparing the output of two reports is to export them to Excel and then performing a comparison between the two Excel’s. If we can export the reports to a word format then we can leverage the word compare utility, even an export to XML would enable using available utility. In case of excel we would need to build a utility that can compare the two excel sheets.

The above three validations are some of the key aspects in validating the objects of semantics and reports; let me know your thoughts on the other means of validation …

Read More About  Product Upgrades And Migrations

The Esoteric World of Predictive Analytics

Let me start with the defintion of Predictive Analytics as used in literature – “The nontrivial extraction of implicit, previously unknown and potentially useful information from data”. If that doesn’t sound esoteric enough, you are probably more advanced than what this post gives you credit for!

For a BI practitioner, it is important to get an understanding of Predictive Analytics (also known as Data Mining) as this subject definitely deserves a place in the wide spectrum of Business Intelligence disciplines. BI at a broad level is about optimizing business through “Hindsight, Insight and Foresight”. Predictive analytics adds the powerful “Foresight” part to business decision making.

Most BI practitioners tend to equate statistics with predictive analytics and this post explains why such a view is inaccurate. To understand this let’s start at the very beginning (a la Alice in Wonderland). Broadly, this world is divided into 2 types of systems:

• Physical Systems – Has causality and hence can be modeled mathematically with relative ease
• Human Behavioral Systems – Lacks causality and can be modeled only with specialized techniques

Predictive analytics for business decision making is all about modeling human behavioral systems.

Why Traditional Statistics is insufficient?

Though the entry into predictive analytics requires that we understand the implications of traditional statistical analysis, statistics by itself is insufficient in the business context. Traditional statistical analysis allows us to understand the general group behavior and is primarily concerned with common behavior within the group – the central tendencies.

In business we generally develop models to anticipate human behavior of some type. Human behavior is inconsistent, lacks causality and distributions based on human behavior almost always violate the assumptions of traditional statistical analysis (like normal distribution of data, stability of mean and standard deviation etc). The strength of data mining comes from the ability of the associated techniques to deal with the tails of the distributions, rather than the central tendencies, and from the techniques’ ability to deal with the realities of the data in a more precise manner.

In the realm of predictive analytics, we are concerned with modeling human behavior and hence are interested with the tail of our distribution – small percentage of the population that responds to a campaign, commits a fraud, leave our business or purchase the next service.

Though there are specialized techniques used for Predictive Analytics (viz. Non-linear statistics, Induction Algorithms, Cluster Analysis, Neural Networks to name a few), a BI practitioner is only expected to appreciate its usage in different business situations, prepare and model data as required by the tools and interpret the results correctly (a much less daunting task indeed!)

Typically the model development process involves the following steps – a) Define Project, b) Select Data, c) Prepare Data, d) Transform Variables, e) Process Model, f) Validate Model, g) Implement Model. I will explain these steps in more detail in subsequent posts.

Fundamentally, an end-to-end BI view requires the practitioner to learn the concepts around statistics and predictive analytical techniques as available in tools (like say SQL Server Analysis Services) in addition to their technology bag of tricks around data integration, data modeling and OLAP.

Read More About  Predictive Analytics

Business Intelligence Challenge – Product Upgrades & Migrations, Moving the Code – 4

Last time we discussed about Impact Assessment , the next logical step after this is to perform the actual upgrade or migration of the code.

Moving the Code: Performing Upgrade or Migration of the Objects

When we talk about product upgrades, always the product vendor provides tools by which the objects in the earlier version can be upgraded to the latest version. Yes we would see some objects failing through while using such tools; these are the ones that would need rework after the upgrade process.

When we talk about product migration like moving from Cognos to Business Objects or Business Objects to Cognos, there is good scope for us to look for some ways to automate the code migration. Earlier discussions have been on how to leverage the metadata for understanding the environment, now we are looking at an option on how to manipulate or transform the metadata so that an object in platform ‘A’ becomes compliant to platform ‘B’.

Steps involved in building an automated product migration process

Perform metadata level object mapping between the two platforms, determine the gaps. This would actually be a ‘by product’ of ‘Step 2’ in Impact Assessment

Build individual components that would

• Read the metadata from the source platform and prepare a repository
• Have the knowledge of the match & gap between the platforms, could be reference tables
• Transform the ‘source’ metadata and write out as understood by the ‘target’ platform by using the reference tables

Benefits of Automated Migration

• Helps avoid creation of objects from scratch
• Ensures availability of time for testing (core task) than code development
• Enables team to have a flexible skillset
• A faster way of delivering things when a ‘one to one’ migration from the source platform is seen as a must

Automated Migration Challenges

Transforming the source metadata to the target platform would be a challenge, especially with data manipulation functions. Having a good understanding of the gaps will help; a reference table mapping the functions between the platforms would be useful. In scenarios where a function cannot be converted to the target platform, a comment can be written into a log file enabling quicker attention.

Have seen good success in writing such automated migration components though not 100%. With almost every products providing good SDK kits for reading and as well writing metadata and as well with the support for XML structures, writing such bridges for object migration are getting easier.

Whether the objects in a product are migrated/upgraded in an automated way or not, the following activity of ‘Validation’ plays a key role in ensuring the final quality, next time let us discuss on some of the means for effective validation ….

Read More About  Business Intelligence Challenge

Zachman Framework for BI Assessments

The Zachman Framework for Enterprise Architecture has become the model around which major organizations view and communicate their enterprise information infrastructure. Enterprise Architecture provides the blueprint, or architecture, for the organization’s information infrastructure. More information on the Zachman Framework can be obtained at www.zifa.com.

For BI practitioners, the Zachman Framework provides a way of articulating the current state of the BI infrastructure in the organization. Ralph Kimball in his eminently readable book “The Data Warehouse Lifecycle Toolkit” illustrates how the Zachman Framework can be adapted to the Business Intelligence context.

Given below is a version of the Zachman Framework that I have used in some of my consulting engagements. This is just one way of using this framework but does illustrate the power of this model in some measure.

Some Salient Points with respect to the above diagram are:

• The framework answers the basic questions of “What”, “How”, “Who” and “Where” across 4 important dimensions – Business Requirements, Conceptual Model, Logical/Physical Model and Actual Implementation.
• Zachman Framework reinforces the fact that a successful enterprise system combines the ingredients of business, process, people and technology in proper measure.
• It is typically used to assess the current state of the BI infrastructure in any organization
• Each of the cells that lies at the intersection of the rows and columns (Ex: Information Requirements of Business) has to be documented in detail as part of the assessment document
• Information on each cell is gathered through subjective and objective questionnaires.
• Scoring Models can be developed to provide an assessment score for each of the cells. Based on the scores, a set of recommendations can be provided to achieve the intended goals.
• Another interesting thought is to create a As-Is Zachman framework and overlay that with To-Be one in situations where re-engineering of a BI environment is undertaken. This will help us provide a transition path from the current state to the future.

Thanks for reading. If you have used the Zachman framework differently in your environment, please do share your thoughts.