Five Good And Really Bad Ideas For Bi Iniatives

50 %
50 %
Information about Five Good And Really Bad Ideas For Bi Iniatives

Published on January 21, 2009

Author: MrBlockHaus

Source: slideshare.net

Description

Presented at the ITA Business Intelligence Roundtable in January, 2009, this presentation discusses best practices and anti-patterns in launching a successful BI / DW initiative.

ITA BI Roundtable The Top 5 Good and Very Bad Ideas in Implementing your BI Initiative Jeff Block Capstone Consulting Tuesday, January 13, 2009 © Copyright 2008 Capstone Consulting, Inc. Capstone Consulting, Inc

What are we talking about? Today’s Agenda  Why are we here?  Who’s in the room?  Presentation – Five Good and Very Bad Ideas  Discussion / Networking © Copyright 2008 Capstone Consulting, Inc. 2 Capstone Consulting, Inc

What are we talking about? Today’s Agenda  Why are we here?  Who’s in the room?  Presentation – Five Good and Very Bad Ideas  Discussion / Networking © Copyright 2008 Capstone Consulting, Inc. 3 Capstone Consulting, Inc

Where’s “here”? Why are we here? Welcome to the first meeting of the ITA Business Intelligence Roundtable © Copyright 2008 Capstone Consulting, Inc. 4 Capstone Consulting, Inc

Who am I? Why are we here? Jeff Block, Capstone Consulting BI Roundtable Chairman © Copyright 2008 Capstone Consulting, Inc. 5 Capstone Consulting, Inc

Logistics Why are we here? • 2nd Tuesday of every month; 8-10 AM – Here at the ITA TechNexus unless there’s a good reason to change venues • Sometimes a presentation – My ideas, your ideas, case studies, best practices, new developments, etc • Sometimes an outside speaker – Love to have some of you step up to the plate – Don’t want to hear myself talk to much • Always discussion – Collaboration is the whole point of this group • Always networking – Meet people who will be valuable connections © Copyright 2008 Capstone Consulting, Inc. 6 Capstone Consulting, Inc

Topics and Target Audience Why are we here? • Business and technology leaders – Not going to spend much time deep in the technical weeds • Those who want to – Learn from each other – Collaborate on solutions – Network in the BI space • ITA members and their friends and their friends and … © Copyright 2008 Capstone Consulting, Inc. 7 Capstone Consulting, Inc

In Scope Why are we here? • Business Intelligence – Vision and strategy – Planning and implementing BI initiatives – High-level architecture – Best practices / Anti-patterns – Case studies – Etc • What about data warehousing? – It’s in! (part of BI, in my world) © Copyright 2008 Capstone Consulting, Inc. 8 Capstone Consulting, Inc

What is Business Intelligence? Why are we here? BI is not about delivering access to a massive repository of data, often unconstrained and overwhelming. BI is about delivering highly relevant, highly valuable applications, reports, and dashboards, designed to maximize the user’s ability to gain specific, actionable knowledge from corporate data. Of course – and this can’t be overstated – this requires a culturally-relevant, well-supported governance model and a highly-specialized, flexible data architecture, both optimized for this purpose. © Copyright 2008 Capstone Consulting, Inc. 9 Capstone Consulting, Inc

What is Business Intelligence? Why are we here? BI is not about delivering access to a massive repository of data, often unconstrained and overwhelming. BI is about delivering highly relevant, highly valuable applications, reports, and dashboards, designed to maximize the user’s ability to gain specific, actionable knowledge from corporate data. Of course – and this can’t be overstated – this requires a culturally-relevant, well-supported governance model and a highly-specialized, flexible data architecture, both optimized for this purpose. © Copyright 2008 Capstone Consulting, Inc. 10 Capstone Consulting, Inc

Classic BI Architecture Why are we here? Our focus is the stuff in this picture and the practices and processes that get it there effectively. BI Presentation Components Data Data Data Data Mart Mart Mart Mart OLAP Services Source Source Systems ETL Data Warehouse ETL Systems © Copyright 2008 Capstone Consulting, Inc. 11 Capstone Consulting, Inc

Out of Scope Why are we here? • Other stuff – How’s Aunt Ruth’s cat? • Building the tech together • Arguing over low-level tech details • If we talk about – Project management / SDLC – Architecture and design – Business processes – Etc then it will be in the context of BI / DW © Copyright 2008 Capstone Consulting, Inc. 12 Capstone Consulting, Inc

Some Quick Feedback Why are we here? How does this line up with your expectations? © Copyright 2008 Capstone Consulting, Inc. 13 Capstone Consulting, Inc

What are we talking about? Today’s Agenda  Why are we here?  Who’s in the room?  Presentation – Five Good and Very Bad Ideas  Discussion / Networking © Copyright 2008 Capstone Consulting, Inc. 14 Capstone Consulting, Inc

Brief Introductions Who’s in the room? Please share with us… • Name • Company • Role • What you want to get out of this group? © Copyright 2008 Capstone Consulting, Inc. 15 Capstone Consulting, Inc

What are we talking about? Today’s Agenda  Why are we here?  Who’s in the room?  Presentation – Five Good and Very Bad Ideas  Discussion / Networking © Copyright 2008 Capstone Consulting, Inc. 16 Capstone Consulting, Inc

Five Best Practices in Implementing your BI Initiative © Copyright 2008 Capstone Consulting, Inc. 17 Capstone Consulting, Inc

Five Best Practices 1. Build a solid value statement © Copyright 2008 Capstone Consulting, Inc. 18 Capstone Consulting, Inc

Solid Value Statement Best Practices #1 Axiom: Every complex, expensive initiative requires a strong business case describing cost and benefits © Copyright 2008 Capstone Consulting, Inc. 19 Capstone Consulting, Inc

Solid Value Statement Best Practices #1 • Make an air-tight case – For the right solution – To real needs – And measurable pain – Which key people are feeling – And passionate about fixing • Be straightforward about cost – Be able to justify that cost with projected ROI – They’ll likely be qualitative, but quantitative is better • Speak in the language of the business • Think out objections and know how to handle them before you start © Copyright 2008 Capstone Consulting, Inc. 20 Capstone Consulting, Inc

Five Best Practices 1. Build a solid value statement 2. Make sure the right person sponsors the initiative © Copyright 2008 Capstone Consulting, Inc. 21 Capstone Consulting, Inc

Initiative Sponsored Best Practices #2 by the Right Person Axiom: Strength of character, leadership, political influence, and personality are required © Copyright 2008 Capstone Consulting, Inc. 22 Capstone Consulting, Inc

Initiative Sponsored Best Practices #2 by the Right Person • One person has to be responsible – Advocacy: The initiative’s top champion – Authority: The influence to make it happen – Accountability: Where the buck stops • Typically the CIO – Or someone he trusts and appoints himself • Must first be a business person – Sponsor has to know the business – Cannot be (even perceived as) an IT initiative • Must have cross-functional clout – Cannot be (even perceived as) about a single business unit • Must be a diplomat – Will have to broker countless peace agreements © Copyright 2008 Capstone Consulting, Inc. 23 Capstone Consulting, Inc

Five Best Practices 1. Build a solid value statement 2. Make sure the right person sponsors the initiative 3. Warehouse no more or less data than you need right now © Copyright 2008 Capstone Consulting, Inc. 24 Capstone Consulting, Inc

No More or Less Data Best Practices #3 Than you Need Myth: Everything has to be in the data warehouse © Copyright 2008 Capstone Consulting, Inc. 25 Capstone Consulting, Inc

No More or Less Data Best Practices #3 Than you Need • Understand strategy and organization’s KPIs • Gather and prioritize requirements from key data stakeholders • Build EXACTLY what they need – No more • If you can’t “justify” why a piece of data is in the data warehouse, then it shouldn’t be there • “Justify” means relating back to the business case / strategy / top-level KPI’s • Everything you do should have that traceability © Copyright 2008 Capstone Consulting, Inc. 26 Capstone Consulting, Inc

No More or Less Data Best Practices #3 Than you Need • Build EXACTLY what they need – No less • Make sure you get everything you need to delivery value • Don’t worry about “complete” requirements – There’s no such thing • Don’t fudge, clean up, assume or create data in your ETL • Data quality issues are fixed in the source system • Mature iteratively – Release new versions (with concrete value) frequently • Presentation layer – every 30 days • Data marts – every 60 days • Data warehouse / ETL – every 90 days © Copyright 2008 Capstone Consulting, Inc. 27 Capstone Consulting, Inc

Five Best Practices 1. Build a solid value statement 2. Make sure the right person sponsors the initiative 3. Warehouse no more or less data than you need right now 4. Identify and build on bedrock data © Copyright 2008 Capstone Consulting, Inc. 28 Capstone Consulting, Inc

Bedrock Data Best Practices #4 Axiom: Drive a stake in the “bedrock data” in your organization, and only graft onto it when expanding © Copyright 2008 Capstone Consulting, Inc. 29 Capstone Consulting, Inc

Bedrock Data Best Practices #4 • Bedrock data is – Data you have (typically) – Data structures that won’t change easily or any time soon • Don’t build on shifting sand – Data most important people in the organization can agree on (conformed dimensions) – Likely related to your business’ most important processes • Which makes it simultaneously very important and very difficult to lock down • Work hard to avoid silos in your BI/DW solution – Hard to work with silos in transactional systems – Nearly impossible to consolidate silo’d DWs © Copyright 2008 Capstone Consulting, Inc. 30 Capstone Consulting, Inc

Five Best Practices 1. Build a solid value statement 2. Make sure the right person sponsors the initiative 3. Warehouse no more or less data than you need right now 4. Identify and build on bedrock data 5. Rely on dimensional modeling to structure your data © Copyright 2008 Capstone Consulting, Inc. 31 Capstone Consulting, Inc

Dimensional Data Modeling Best Practices #5 Axiom: Transactional systems and BI systems require completely different data models © Copyright 2008 Capstone Consulting, Inc. 32 Capstone Consulting, Inc

Dimensional Data Modeling Best Practices #5 • Operational systems – Where we put data in – Users make the wheels of the organization turn – Users affect one row at a time • Data model optimized for the current transaction; writing a single new record • No need to know history (so shouldn’t) – Users perform the same tasks over and over © Copyright 2008 Capstone Consulting, Inc. 33 Capstone Consulting, Inc

Dimensional Data Modeling Best Practices #5 • BI systems / Data warehouse – Where we get data out – Users watch the wheels of the organization turning – Users read and perform domain aggregate functions on many, many rows at a time • Almost never deal with one row at a time • Optimized for reading and comparing many rows, aggregating and disaggregating data, drilling up/down/through, etc • Ideal for consistent historical and comparative trending – Users continuously change the kinds of questions they ask © Copyright 2008 Capstone Consulting, Inc. 34 Capstone Consulting, Inc

Dimensional Data Modeling Best Practices #5 Dimensions Dimensions • Two core elements: Facts Customers Clerks “facts” and “dimensions” – Facts Customer • Core data of a Items Purchased Dates business event Item • The “verb” in the sentence describing Store Times the event – Dimensions • Context in which the event occurred • The “nouns” in the sentence • Sometimes called “cubes” – Because comparisons are made by pivoting on dimension tables joined by fact tables • Sometimes called a “star schema” – Because of the star-like shape of a fact table surrounded by dimension tables © Copyright 2008 Capstone Consulting, Inc. Capstone Consulting, Inc

Five Best Practices: Questions? 1. Build a solid value statement 2. Make sure the right person sponsors the initiative 3. Warehouse no more or less data than you need right now 4. Identify and build on bedrock data 5. Rely on dimensional modeling to structure your data © Copyright 2008 Capstone Consulting, Inc. 36 Capstone Consulting, Inc

Five Worst Practices in Implementing your BI Initiative © Copyright 2008 Capstone Consulting, Inc. 37 Capstone Consulting, Inc

Five Worst Practices 1. Pick the technology first © Copyright 2008 Capstone Consulting, Inc. 38 Capstone Consulting, Inc

Pick the Technology First Worst Practices #1 Myth: The technology is the hard part © Copyright 2008 Capstone Consulting, Inc. 39 Capstone Consulting, Inc

Pick the Technology First Worst Practices #1 • Many see BI / DW as a technology problem – So, they purchase tools in an attempt to “buy BI” • Tech is actually the (comparatively) easy part • Much harder – Securing executive commitment (including funding) – Building consensus / conforming dimensions – Establishing governance – Controlling scope and growth – Accurately defining requirements – Supporting the system – Etc, etc, etc © Copyright 2008 Capstone Consulting, Inc. 40 Capstone Consulting, Inc

Pick the Technology First Worst Practices #1 A quick, rough order of priority… 1. Plan 2. People 3. Process 4. P… Technology © Copyright 2008 Capstone Consulting, Inc. 41 Capstone Consulting, Inc

Five Worst Practices 1. Pick the technology first 2. We’re agile; we should just start building © Copyright 2008 Capstone Consulting, Inc. 42 Capstone Consulting, Inc

Just Start Building Worst Practices #2 Myth: We can figure that out later (whatever “that” is) © Copyright 2008 Capstone Consulting, Inc. 43 Capstone Consulting, Inc

Just Start Building Worst Practices #2 • News flash: The reason you’re putting it off is because it’s hard • Hard stuff has to get done first, not last • Always flesh out your risk as soon as possible • Agile / iterative is good, even essential – Applies to iterative construction of tools, delivering on requirements – Implies phased rollout, short sprints, frequent releases of value, etc – Does NOT imply that we jump over or scream by the vision and strategy stages © Copyright 2008 Capstone Consulting, Inc. 44 Capstone Consulting, Inc

Five Worst Practices 1. Pick the technology first 2. We’re agile; we should just start building 3. Ignore the political landscape © Copyright 2008 Capstone Consulting, Inc. 45 Capstone Consulting, Inc

Ignore the Political Landscape Worst Practices #3 Myth: Good planning, smart people, and hard work can overcome anything © Copyright 2008 Capstone Consulting, Inc. 46 Capstone Consulting, Inc

Ignore the Political Landscape Worst Practices #3 • Do invest too heavily in your BI initiative until you have… – Plentiful executive support • Demonstrated, not just verbal – Some skeptics won over – Staff bribed in key positions all over the organization – Excellent business case – Well-conceived objection handling plan – Outside help on speed-dial for areas which aren’t your strengths • Much of this implies quick wins up front © Copyright 2008 Capstone Consulting, Inc. 47 Capstone Consulting, Inc

Five Worst Practices 1. Pick the technology first 2. We’re agile; we should just start building 3. Ignore the political landscape 4. Leave out critical support functions © Copyright 2008 Capstone Consulting, Inc. 48 Capstone Consulting, Inc

Critical Support Functions Worst Practices #4 Myth: Build it and they will come © Copyright 2008 Capstone Consulting, Inc. 49 Capstone Consulting, Inc

Critical Support Functions Worst Practices #4 • You need governance – How do we decide how to decide? – People, policies, processes, permissions, etc – Key policy needed: Limit Jeff’s alliteration of “P” • You need marketing – It’s not true that “if we just have this awesome data warehouse, then everyone will fall over themselves to use it” – Build a communication plan – Expect most people to be skeptics, especially most of those who helped you get your funding © Copyright 2008 Capstone Consulting, Inc. 50 Capstone Consulting, Inc

Critical Support Functions Worst Practices #4 • You need user training – Fact: Business users won’t know how to use BI unless the presentation is mind-numbingly simple • If you need paragraphs of explanation, rethink your design – Spend time in the classroom, not generating help files • You need IT support – The DW won’t take care of itself – Needs more care and feeding than other apps, not less – Consider external managed support © Copyright 2008 Capstone Consulting, Inc. 51 Capstone Consulting, Inc

Five Worst Practices 1. Pick the technology first 2. We’re agile; we should just start building 3. Ignore the political landscape 4. Leave out critical support functions 5. Don’t address the question of data ownership © Copyright 2008 Capstone Consulting, Inc. 52 Capstone Consulting, Inc

Data Ownership Worst Practices #5 Myth: IT owns the data © Copyright 2008 Capstone Consulting, Inc. 53 Capstone Consulting, Inc

Data Ownership Worst Practices #5 • IT should NOT own the data – They own the machinery that holds the data – Difference between “water” and “plumbing” • No one business unit should own the data either – The more you can make it the whole organization’s data, the better off you’ll be – After all, the goal is cross-functional analysis • Governance council is the representative government of the data warehouse / BI © Copyright 2008 Capstone Consulting, Inc. 54 Capstone Consulting, Inc

Five Worst Practices: Questions? 1. Pick the technology first 2. We’re agile; we should just start building 3. Ignore the political landscape 4. Leave out critical support functions 5. Don’t address the question of data ownership © Copyright 2008 Capstone Consulting, Inc. 55 Capstone Consulting, Inc

Capstone BI Preparedness Index • Six key dimensions of readiness – Passion – Knowledge – Need – Vision – Influence – Credibility • Capstone meets with key personnel – In the context of a BI assessment – Organization has to collectively score well enough on all these dimensions to justify firing the start gun • Offer simple quantitative indicator of readiness – Offer gap analysis if not ready – Recent example: © Copyright 2008 Capstone Consulting, Inc. 56 Capstone Consulting, Inc

Q&A Group Discussion © Copyright 2008 Capstone Consulting, Inc. 57 Capstone Consulting, Inc

w w w. c a p s t o n e c . c o m © Copyright 2008 Capstone Consulting, Inc. Capstone Consulting, Inc

Add a comment

Related presentations

Related pages

Big Five Personality Test - Queendom: personality tests ...

Big five personality factors and personality type ... One of on our most popular personality tests, the Big Five should really get you thinking about what ...
Read more

Funny T Shirts | BadIdeaTShirts.com

At Bad Idea T Shirts, we ... we really do offer hundreds of different ... It’s all in good fun, after all! Our funny t shirts range from squeaky clean ...
Read more

Good - Idioms by The Free Dictionary

Definition of good in the Idioms Dictionary. good phrase. ... take the bad with the good; Take the goods the gods provide; ... What’s really good witcha?
Read more

Idea Practices | Idea Practices

This is a concept we’ve all seen take off in a really big way over these past few years and it shows no sign of slowing down. ... The Good And The Bad
Read more

Ideas for Business and Excel - ExcelUser Home Page

... it's a good idea to take stock of your business to ... Excel can be the key component of a powerful and inexpensive Business Intelligence (BI) ... Five ...
Read more

Birthday Gifts & Ideas - Gifts.com

Find birthday gift ideas for everyone you care about. ... Whether they are turning five or ... Cookies and other treats also make good birthday gifts for ...
Read more

How to Be Good: 11 Steps (with Pictures) - wikiHow

How to Be Good. Community Q&A ... etc.), then that means you are really good ... greater good. What might seem like a good idea in your situation might not ...
Read more

Writing: What is Good Writing? Six Traits Assessment Quality

The Five Facts of Fiction ... Good writing has: Ideas that are interesting and important. Ideas are the ... Not too bad, ...
Read more