Amber Leaf specializes in quickly breaking down the

Amber. Leaf specializes in quickly breaking down the barriers that stand between your data and action. Software Consolidation and Its Impact on Data Management Presented by: Larry Goldman DAMA–MN April 16, 2008 DATA INFORMATION INSIGHT STRATEGY ACTION

Agenda • Introduction to Amber. Leaf • Business Intelligence Vision • Consolidation History • Customer Benefits and Impacts DATA INFORMATION INSIGHT STRATEGY ACTION

Breaking Down the Barriers Urgency, Speed, Focus, Expertise, Results Information • Data Warehouse/Data Mart strategy, design and development • Data Integration (ETL on Demand) • Data Quality • Customer Information Hygiene DATA Strategy Insight • Reporting Architectures • Management Reporting • Business Performance Management • Analytical Applications • Measurement and metrics INFORMATION • Loyalty Programs • Advanced Analytics • Customer Valuation • Segmentation • Marketing Optimization • Customer Experience Strategy • Sales • Service • Marketing INSIGHT STRATEGY Action • Customer Experience Optimization • Marketing Automation • Personalization • Process Design • Change Agents • Service Optimization • Sales Effectiveness ACTION

Agenda • Introduction to Amber. Leaf • Business Intelligence Vision • Consolidation History • Customer Benefits and Impacts DATA INFORMATION INSIGHT STRATEGY ACTION

ETL Architecture Metadata External Parties ETL Architecture Data Files CMDM_PHYQA_ERR_PARM PHYQA_ERR_KEY INFA_SESS_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE Source/Legacy Systems Support Personnel CMDM_PROCESS_PARM PROC_KEY PROC_NAME PROC_DESC CMDM_INFA_SESS_EXP_PARM INFA_SESS_EXP_KEY INFA_SESS_KEY (FK) EXP_NAME EXP_DESC LKP_TABLE_NAME SOURCE_NAME TARGET_NAME CMDM_INFA_SESS_PARM INFA_SESS_KEY PROC_KEY (FK) INFA_FOLDER_NAME INFA_WORKFLOW_NAME INFA_SESS_NAME CMDM_LOGQA_ERR_PARM LOGQA_ERR_KEY INFA_SESS_EXP_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE CMDM_ERR_LOG_KEY INFA_SESS_KEY (FK) INFA_SESS_EXP_KEY (FK) AUDIT_KEY (FK) ERR_TYPE_CODE ERR_DESC LOGQA_INPUT_VALUE OPER_RESOLUTION OPER_COMMENTS ERR_STATUS_CODE ERR_CALLING_MODULE CMDM_AUDIT_KEY PROC_KEY (FK) PROC_STATUS_CODE PHYQA_STATUS_CODE LOGQA_STATUS_CODE PROC_START_DTS PROC_END_DTS CMDM_INFA_SESS_STAT_KEY PHYQA_ERR_KEY (FK) INFA_SESS_KEY (FK) AUDIT_STAT_DTS NUM_RECS_SUCCESS NUM_RECS_FAILED NUM_RECS_TOTAL PCTG_FAILED_RECS SESS_STATUS_CODE SESS_START_DTS SESS_END_DTS Audit Tables CMDM_PHYQA_ERR_PARM PHYQA_ERR_KEY INFA_SESS_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE Unix Directories Conform Stage (TMP & PS) EDW Data Marts Data Reconciliation Processing / Data Watching Information Consumers Initiate Data Load Validate Data Communicate Issues CMDM_PROCESS_PARM PROC_KEY PROC_NAME PROC_DESC CMDM_INFA_SESS_EXP_PARM INFA_SESS_EXP_KEY INFA_SESS_KEY (FK) EXP_NAME EXP_DESC LKP_TABLE_NAME SOURCE_NAME TARGET_NAME CMDM_INFA_SESS_PARM INFA_SESS_KEY PROC_KEY (FK) INFA_FOLDER_NAME INFA_WORKFLOW_NAME INFA_SESS_NAME CMDM_LOGQA_ERR_PARM LOGQA_ERR_KEY INFA_SESS_EXP_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE CMDM_ERR_LOG_KEY INFA_SESS_KEY (FK) INFA_SESS_EXP_KEY (FK) AUDIT_KEY (FK) ERR_TYPE_CODE ERR_DESC LOGQA_INPUT_VALUE OPER_RESOLUTION OPER_COMMENTS ERR_STATUS_CODE ERR_CALLING_MODULE CMDM_AUDIT_KEY PROC_KEY (FK) PROC_STATUS_CODE PHYQA_STATUS_CODE LOGQA_STATUS_CODE PROC_START_DTS PROC_END_DTS CMDM_INFA_SESS_STAT_KEY PHYQA_ERR_KEY (FK) INFA_SESS_KEY (FK) AUDIT_STAT_DTS NUM_RECS_SUCCESS NUM_RECS_FAILED NUM_RECS_TOTAL PCTG_FAILED_RECS SESS_STATUS_CODE SESS_START_DTS SESS_END_DTS Meta Data Repository Information Metadata • • • DATA Access Types It has been estimated that ETL and data integration make up anywhere from 50% to 80% of all data warehouse efforts ETL technology from leading vendors like Data. Stage (IBM) and Informatica continue to make their applications easier to use, more scalable, more interoperable, and able to capture better meta-data However, for the most part, many users find that these tools have not had any revolutionary new functionality in the last couple releases INFORMATION INSIGHT STRATEGY ACTION

Dynamic Data Warehousing/MDM Order Entry SFA • • • None of the hard parts of ETL go away But some of these technologies buffer the end result (data mart) from changes in sources, business rules, or business models Some generate BI meta data models Customer Data DATA INFORMATION Customer Service Web Inventory ETL Engine Staging Active Meta Data Product Data INSIGHT Sales Analysis Marketing Performance STRATEGY ACTION

Business Intelligence Architecture Metadata External Parties Source/Legacy Systems Process Metadata ETL Architecture Data Files CMDM_PHYQA_ERR_PARM PHYQA_ERR_KEY INFA_SESS_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE CMDM_PROCESS_PARM PROC_KEY PROC_NAME PROC_DESC CMDM_INFA_SESS_EXP_PARM INFA_SESS_EXP_KEY INFA_SESS_KEY (FK) EXP_NAME EXP_DESC LKP_TABLE_NAME SOURCE_NAME TARGET_NAME CMDM_INFA_SESS_PARM INFA_SESS_KEY PROC_KEY (FK) INFA_FOLDER_NAME INFA_WORKFLOW_NAME INFA_SESS_NAME CMDM_LOGQA_ERR_PARM LOGQA_ERR_KEY INFA_SESS_EXP_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE CMDM_ERR_LOG_KEY INFA_SESS_KEY (FK) INFA_SESS_EXP_KEY (FK) AUDIT_KEY (FK) ERR_TYPE_CODE ERR_DESC LOGQA_INPUT_VALUE OPER_RESOLUTION OPER_COMMENTS ERR_STATUS_CODE ERR_CALLING_MODULE CMDM_AUDIT_KEY PROC_KEY (FK) PROC_STATUS_CODE PHYQA_STATUS_CODE LOGQA_STATUS_CODE PROC_START_DTS PROC_END_DTS CMDM_INFA_SESS_STAT_KEY PHYQA_ERR_KEY (FK) INFA_SESS_KEY (FK) AUDIT_STAT_DTS NUM_RECS_SUCCESS NUM_RECS_FAILED NUM_RECS_TOTAL PCTG_FAILED_RECS SESS_STATUS_CODE SESS_START_DTS SESS_END_DTS Audit Tables Support Personnel CMDM_PHYQA_ERR_PARM PHYQA_ERR_KEY INFA_SESS_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE Unix Directories Conform Stage (TMP & PS) EDW Data Marts Data Reconciliation Processing / Data Watching Information Consumers Initiate Data Load Validate Data Communicate Issues CMDM_PROCESS_PARM PROC_KEY PROC_NAME PROC_DESC CMDM_INFA_SESS_EXP_PARM INFA_SESS_EXP_KEY INFA_SESS_KEY (FK) EXP_NAME EXP_DESC LKP_TABLE_NAME SOURCE_NAME TARGET_NAME CMDM_INFA_SESS_PARM INFA_SESS_KEY PROC_KEY (FK) INFA_FOLDER_NAME INFA_WORKFLOW_NAME INFA_SESS_NAME CMDM_LOGQA_ERR_PARM LOGQA_ERR_KEY INFA_SESS_EXP_KEY (FK) TOLERANCE_TYPE_CODE NUM_RECS_TOLERANCE PCTG_TOLERANCE CMDM_ERR_LOG_KEY INFA_SESS_KEY (FK) INFA_SESS_EXP_KEY (FK) AUDIT_KEY (FK) ERR_TYPE_CODE ERR_DESC LOGQA_INPUT_VALUE OPER_RESOLUTION OPER_COMMENTS ERR_STATUS_CODE ERR_CALLING_MODULE CMDM_AUDIT_KEY PROC_KEY (FK) PROC_STATUS_CODE PHYQA_STATUS_CODE LOGQA_STATUS_CODE PROC_START_DTS PROC_END_DTS CMDM_INFA_SESS_STAT_KEY PHYQA_ERR_KEY (FK) INFA_SESS_KEY (FK) AUDIT_STAT_DTS NUM_RECS_SUCCESS NUM_RECS_FAILED NUM_RECS_TOTAL PCTG_FAILED_RECS SESS_STATUS_CODE SESS_START_DTS SESS_END_DTS Meta Data Repository Information Metadata Decision Support Framework Access Types Data Dictionary Data Extracts Ad-Hoc Reporting Statistics Standard Reporting Dashboards Delivery Methods RATE PLAN HISTORY SUBSCRIPTION RATE PLAN HISTORY RATE PLAN CODE BILL ACCOUNT HISTORY RATE PLAN DESC SUB ID SITE NUM CODE BILL ACCT BILL FNAME ACCT ID MKT RATE PLANID CODE BILL SALES AREA CURRENT PRODUCT CODE DESC SUBSCRIPTION RATE PLAN PRD CODE BILL MNAME Bill Account PRODUCT RATE PLAN PRD SUB GRP CODE BILL LNAME -- Subscription PRODUCT CODE TYPE SUB ID RATE BILL IN CARE OF NAME SVC TYPE BILL DAY ACCT ID VALUESHARE FLAG ACCT FNAME SITE NAME CODE PYMT TYPECODE BILL SVC TYPE ACCT SITE VARIANCE 1 CODE SUB FNAME PYMT TYPECODE ACCT MNAME LNAME SUB SITE CODE VARIANCE 3 2 SUB LNAME MNAME EFF DTS ACCT IN CARE OF NAME CONSOLIDATED VARIANCE CO NAME SUB END DTS SAC NAME VARIANCE 4 SUB IN CARE OF NAME ESTABLISH DATE TIER NUM ACCESS CHARGE COST SUB ADDR LINE 2 LINE 1 SUBSCRIPTION STATUS HISTORY HRISCOMPANY MKT NUM NAME ACCT LEVEL IND BIRTH CENTER DATE SUB JDE COMMISSION FLAG SUB ADDR LINE 3 SOCIAL SECURITY NUM JDE WLN WLS CO NUM PACKAGE MINS CREDIT BUREAU SCORE JDE RISK CO NUM SUB CITY SUB ID BILL SYSTEM CODE SUB BILL ACCT ID DEPOSIT JDE FLAG LD CO NUM SRC SYSTEM CODE SUB ZIP STATE EFF DTS DEPOSIT REQUIRED JDE AMTPREPAY CO NUM DISTRIBUTION EFF CHANNEL DEPOSIT TAKEN AMT JDE INT COCO NUM SUB COUNTRY END DTS DTS GENDER PICTURE ID STATE JDE PAGE DIST CHAN CODE OFFPEAK ESTABLISH DATE SUB STATUS ID DIST PICTURE ID VPGM FNAME NUMMIN NAME MIN PEAK GEO CODE VPGM MNAME DIST CHAN SUBANYTIME GRP MIN CODE MIN MDN LAST HOTLINED VPGM LNAME DIST GRP NAME M 2 M MIN LAST NPD DATE VPGMFNAME EMPL POSTN ESN NUM PHONE DIST CHAN SUB GRPCALL CODE DELIVERY CPNI MAP WORK DIST CHAN GRPPOST NAME PKG PEAKRATE HOME PHONE DATE CREDIT REFERENCE MAP NUM MNAME POST PKG M 2 M OFFPEAK DECISION CODE MAP LNAMEPOSTN CANCELLATION VOL DISC FLAG POST RATE NEXT BILL DATE MAP EMPL NUM IN CONTRACT REV ACCT CODE CALENDAR MAJOR ACCT FLAG RP RP FNAME CDPDSTATUS NEI ROAMING MINS MAJOR ACCT CODE MNAME SALES REP HISTORY BILL CALENDAR DATE MAJOR ACCT ID RP EMPL LNAME Sub Handset MAKE MODELCODE ID -- Subscription DAY OF WEEK NAME ACCT STATUS ID RP POSTN NUM RATE PLAN CODE DAY OF SHORT ACCT TYPE ID COO FNAME MAKE MODEL RATE PLAN PRD CODE DAY IN IN MONTH WEEK CREDIT CODE CLASSID ID COO MKT CODE SLSCODE REP ID RATE PLAN PRD GRP DAY CREDIT MNAME MAKE MODEL ID EMPL DISPLAY NAME RATE GRP CODE DAY IN IN YEAR QUARTER CREDIT CODE DESC MKT NUM MAKE HRIS MKT NUM CREDIT CLASS IDSUB DAY BILL COO LNAME MODEL BILL SYSTEM CODE CREDIT CODE DESC ID REP WEEK IN QUARTER MONTH BILL DAY CYCLE IDCODE COO MKT NAME MANUFACTURER DCID CREDIT WEEK IN BILL SYSTEM EMPL POSTN TYPIST NUM ID LIFE CYCLE CODE EFF DTS WEEK IN YEAR TECHNOLOGY BILL HOLD CEO FNAME TYPE CODE END DTS SLS REP ID MONTH BILL CT RDM MKT CODE COMPARABLE MAKE DIST STATUS CHAN CODE MONTH IN NAME BILL IN CONTROL ARREARS FLAG MKT AREA CODE SUB COMPARABLE MODEL ID MONTH QUARTER TREAT CEO MNAME PHONE WEIGHT CURR BAL DUE MKT AREA NUM CODE QUARTER SOFTWARE VER MKT AREA NAME MKT AMT 30 YEAR HISTORY WUW SOFTWARE CEO LNAME REGION CODE SUBSCRIPTION HANDSET AMT 60 HOLIDAY FLAG DIG STANDBY TIME REGION CODE AMT 90 SITE WEEKEND FLAG DIG TALK TIME CEO EMPL POSTN SUB NUM CODE AMT 120 SUB ID LAST DAY IN MONTH BATTERY FLAG TYPE CODE REGION NAME SVC SITE TYPECODEBILL ACCT ID AMT 150 SALES DAY OF VIB WEEK ALERT FLAG SRC SYSTEM CODE AMT 180 BILL SYSTEM CODE PYMTTYPEIDCODE MAKE SALES WEEK RING TONE FLAG MODEL SUB EFF DTS MID MTH COMMISSION RING TONE MTH DOWNLOAD FLAG HS ID BILL SYSTEM CODE END DTS MID MTH COMMISSION DATA FAX QTR FLAG HS BLUETOOTH FLAG ESNSLS DATE VOICE DIAL FLAG MAKE MODEL ID CODE SPEAKER BILL ACCOUNT ADDRESS HS CHG REASON WUW FLAGPHONE FLAG ITEM NUM TTY FLAG BILL ITEM COST QTY GPS FLAG BILL ACCT ADDR ID LINE 1 UNIT AXCESS SHOP FLAG BILL ADDR LINE 2 RETAIL AMT ONE TOUCH DIAL BILL ADDR LINE 3 SUBSIDY MEMORY CT FLAG BILL CITY ORDERCPE IDAMT CALLER IDLOC FLAG BILL ZIP COAM FLAG NAME DISPLAY FLAG BILL COUNTRY STATE SUBSCRIPTION CONTRACT HISTORY SHIP FEE AMT PRL FLAG BILL SHIP COST AMT ACCT ADDR LINE 2 LINE 1 SUB ID FULLFILLMT ACCT CONTRACT START DATE SLS REP ID COST ACCT ADDR LINE 3 HANDSET BILL ACCT ID RETURN FLAGCODE TYPIST CREDIT CLASS ACCT ZIP CITY CONTRACT END DATESRC Sub Handset HS ID CONTRACT LENGTH BILL SYSTEM CODE -- Subscription BILL SYSTEM CODE ACCT STATE ITEM DESC CONTRACT TYPE EFF DTS CREDIT CLASS DESC IDHOME ACCT COUNTRYEARLY RETAIL TYPISTDISPLAY ID CREDIT MAKE PRICE DTSDISC PENALTYEND DTS EMPL NAME WORK PHONE EFF MODEL END DTS BILL SYSTEM CODE Data Model Web DATA INFORMATION Email INSIGHT FTP Client App STRATEGY ACTION

Brett Hively – BRM Agent Desktop Home Page My Lounge My Rankings Search Windows Help Logout Brett Hively’s Home Page AE Home Page Scorecard Summary RM Scorecard Detail Key Metrics Funded Volume Funded Units Broker Penetration Weighted Deviation Funding Ratio Share of Wallet Cost per 1 st Lien Units Loans per AE Fast. Qual % Brokers Funded in 90 Days Yesterday $119, 830, 703 527 14. 5% . 53 66. 8% 34% $2, 680 17. 4 33. 3% 178 Same Day Last Month $120, 543, 987 400 15% . 60 60. 4% 30% $2, 589 16. 5 27% 165 Quarter to Date $119, 830, 703 527 14. 5% . 53 66. 8% 34% $2, 680 17. 4 33. 3% 178 RM Home Page Internal Performance Loan Performance Internal Performance Loan Count Average Days To Next Stage Underwriting 40 2 Potential FPD’s 8 Account Manager 250 8 30 Day 1 Doc’s Out 157 4 60 Day 1 Doc’s Out 140 5 90 Day 0 REO’s 1 Investor Rejects 2 Total Days Sub to Close 13. 5 RM Alert Summary RM Lead Summary Total Alerts Status 185 New 110 Attention 25% Submission Drop 100 60 40 % Below Funding Ratio 50 30 New Loan Officers 25 15 Docs in Past 3 Days 10 5 DATA Internal Performance INFORMATION Total Leads 185 110 Status Assigned Unassigned Approved 100 60 Pending 50 30 Qualified 25 15 Validated 10 5 Terminated 5 - INSIGHT Loan Count RM Territory Summary Broker Swap Candidates Period Funded 60 Days 100 90 Days 50 120 Days 25 180 Days 10 STRATEGY Never Funded 60 30 15 5 ACTION

Lapse Likelihood by Sales Deciles Lapse Likelihood Index by Sales Decile (100 = Average Likelihood) BEST CUSTOMERS ARE LESS LIKELY TO BE LAPSED CUSTOMERS. HOWEVER, SIGNIFICANT PROPORTIONS OF BEST CUSTOMERS ARE STILL LAPSED. DATA INFORMATION INSIGHT STRATEGY ACTION

Seamless Transition From… • • DATA Strategy Tactical Plans Dashboards Management Reporting Ad-hoc Reporting Multi-dimensional Analysis Statistical Analysis INFORMATION INSIGHT STRATEGY ACTION

Real-Time Coordination SFA • • • Order Entry Web Some architectures have chosen a distributed methodology Others have chosen ODS Others have chosen hybrids Customer Service Inventory Distribution Architecture Operational Data Store ETL Decision Support DATA INFORMATION INSIGHT STRATEGY ACTION

Operational Business Intelligence Continuous visibility – For critical, intra-day measurements – Aggregated across multiple systems Flexible user interface – Self-service model – User-defined thresholds and alerts – Graphical watch points Role-based delivery – Role based data level security – Customization by end users Collaboration – Built-in workflow/collaboration – Complete loop from discovery to action DATA INFORMATION INSIGHT STRATEGY ACTION

Brett Hively – BRM Agent Desktop Home Page My Lounge My Rankings Search Windows Help Logout Assign Broker Brett Hively’s Broker List All Funded Never Funded 60 Days Swap Candidate Broker List Select Disp. All ute. Broker Name Broker Stage Code Assigned AE # of Loan Officers # of Funded Units Broker Temp Last Assign Date Broker ABC At Risk Frank Morley 10 24 Un Happy 1/2/06 Broker DEF At Risk Jim Smith 5 3 Un Happy 2/8/06 Broker GHI At Risk Brian Slucki 25 39 Un Happy 1/12/06 Assign Broker Name. Broker JKL City Broker MNO Broker ABC Irvine Broker PQR Broker STU Assignment History At Risk # of Loan Assigned. Pete AE Berchich Officers Mike Masters Frank Morley 10 Todd Walsh At Risk Joe Widlowski Broker State. At Risk CA At Risk Broker VWXAssigned Jr. AE At Risk Last Assign Dustin Lagar Assigned AE Date Broker YZ At Risk Bob Purchase Frank Morley 1/2/06 Broker ABC At Risk Frank Morley Jim Smith 2/8/06 Broker DEF At Risk Jim Smith Brian Slucki 1/12/06 Dennis Ugolini Jr. Broker GHI At Risk Brian Slucki Pete Berchich 2/11/06 Broker JKL At Risk Pete Berchich 20 Assignment 4 Un Happy 2/8/06 Assignee 50 6 Un Happy 2/4/06 Frank Morley 10 14 • Assign Un Happy 1/11/06 10 24 5 3 AE Assignment 25 22 Mike Masters 12 At Risk Todd Walsh 15 33 Un Happy 1/31/06 Broker STU At Risk Joe Widlowski 20 4 Un Happy 2/8/06 Broker VWX At Risk Dustin Lagar 50 6 Un Happy 2/4/06 Broker YZ At Risk Bob Purchase 10 14 Un Happy 1/11/06 Broker PQR 2/3/06 INFORMATION INSIGHT 39 • Assign • AE Assignment History • Broker Scorecard • AE Scorecard • Broker Relationships Un Happy 1/2/06 Scorecard Un Happy • AE Scorecard 2/8/06 30 Tim Smith 20 Eddie Olson At Risk Assigned Jr. AE • Undispute • Broker Un Happy 1/12/06 • Broker Relationships Un Happy 2/11/06 • Cancel Un Happy 2/3/06 Mike Masters Broker MNO DATA Broker Last Assign 2/11/06 30 # of Funded 22 Satisfaction Un Happy Units Date Reason 20 12 Un Happy 2/3/06 24 Un Happy 1/2/06 15 33 Un Happy 1/31/06 • Dispute STRATEGY ACTION

Brett Hively – BRM Agent Desktop Home Page My Lounge My Rankings Search Windows Help Logout AE Ranking Frank Morley’s Rankings Frank Morley’s Ranking for Funded Units as of April 4 th, 2008 Ranking Same Day Last Month Metric Value Current Ranking Current Metric Value Scott Soga 1 23 15 9 Brian Pavur 2 22 14 10 Tony Bafano 3 21 13 11 Rick Harris 4 20 12 12 Mark Liskewicz 5 19 11 13 Frank Morley 6 18 10 14 Frank Melella 7 17 9 15 Joanna Basil 8 16 Robin Budz 9 15 7 17 Kristi Volger 10 14 6 18 Steve Tangenhorse 11 13 5 19 Ron Chester 12 12 4 20 Heather Thompson 13 11 3 21 Anthony Astrowski 14 10 2 22 Terra Arthur 15 9 1 23 Conversion Ratio DATA INFORMATION INSIGHT STRATEGY ACTION

What Do We Want? • • • DATA Save us time? Save us money? Easier maintenance? Faster development? Better integration between ETL and Business Intelligence? Easier impact analysis across the ecosystem? Sharing business rules between real-time and ETL data integration? Ability to seamlessly work between BI technologies? Ability to embed BI in our CRM/ERP applications? INFORMATION INSIGHT STRATEGY ACTION

Agenda • Introduction to Amber. Leaf • Business Intelligence Vision • Consolidation History • Customer Benefits and Impacts DATA INFORMATION INSIGHT STRATEGY ACTION

SAP Acquisitions Business Objects, Outlook. Soft, Pilot Software, Callixa, A 2 i Business Objects Crystal, Infommersion, Cartesis, Nsite, SRC, OLAP@work, Acta, Medience, First. Logic, Inxight, Informatik Indicates duplicate product sets due to acquisition DATA INFORMATION INSIGHT STRATEGY ACTION

Oracle Acquisitions Siebel, Hyperion, IRI, Thinking Machines, Sunopsis, One Meaning, Times. Ten, Stellent, Inno. DB, People. Soft Siebel Hyperion Nquire Upstream, Brio, Razza, Decisioneering Indicates duplicate product sets due to acquisition DATA INFORMATION INSIGHT STRATEGY ACTION

Microsoft Acquisitions Pro. Clarity, Stratature DATA INFORMATION INSIGHT STRATEGY ACTION

IBM Acquisitions Cognos, Informix, Ascential, Alpha. Blox, DWL, File. Net, Data. Mirror, Venetica, Trigo, Green Pasture, SRD Cognos Celequest, Applix, Adaytum, Decision. Stream, Frango Indicates duplicate product sets due to acquisition DATA INFORMATION INSIGHT STRATEGY ACTION

Marketing Automation Aprimo Double. Click Unica Marketing Central, Market. Soft, Sane Solutions Indicates duplicate product sets due to acquisition DATA INFORMATION INSIGHT STRATEGY ACTION

Marketing Automation (cont’d) Lyris, Email. Labs, Click. Track, Hot Banana DATA INFORMATION INSIGHT STRATEGY ACTION

Agenda • Introduction to Amber. Leaf • Business Intelligence Vision • Consolidation History • Customer Benefits and Impacts DATA INFORMATION INSIGHT STRATEGY ACTION

Did It Make Sense? - BI ETL Oracle SAP MDM Analytical Applications Reporting OLAP Stats Dashboard X X (major duplication) X (not well known) X (major duplication) X(du plicat ion) X X X (duplication ) X X Microsoft X IBM X DATA X X INFORMATION X X INSIGHT STRATEGY Operation -al BI X ACTION

Did It Make Sense? – Marketing Campaign Marketing Resource Management Statistics Bidirectional Reporting Real-Time Offer/Event Detection Lead Management Web Analytics Unica X X (duplication) X X X Aprimo X X X Lyris X DATA INFORMATION X INSIGHT STRATEGY ACTION

Conclusions • Most of the acquisitions over the past few years have propelled the “aquirers” into marketplaces where they were lacking or had no previous capabilities • Some of these companies have moved into fairly new areas – IBM in user facing tools – SAP as an independent tool vendor • Though this helps the software company, it has rarely had, or is having, a positive impact on the user base • The most immediate impact would be one-stop shopping and pricing • Except for consolidating the number of vendors you need to call, no real new capability has come (or has even been announced) from these organizations • Examples of successful BI acquisitions: – – DATA Business Objects and Acta: Provides a full data warehouse suite for some BO clients Oracle and IRI: Provides OLAP and relational database from one company Unica and Marketsoft: Provides the bridge between sales and marketing Siebel and Paragren and n. Quire: Provides the most powerful CRM suite, analytical and marketing capabilities INFORMATION INSIGHT STRATEGY ACTION

What May Be Next • Hopefully, we will see some innovation down the road: – True operational BI from Oracle by combining Oracle ERP, Siebel CRM, and Siebel Analytics (and possibly Thinking Machines). – True operational BI from Microsoft by combining Business Intelligence Services and Microsoft Dynamics. The key for Microsoft will be to make analytics as simple as Microsoft Word for the masses: Guided Analytics, pre-canned solutions, and vertical integration. – Pre-packaged, right-time BI engine from IBM by combining Websphere, Ascential, and Cognos. DATA INFORMATION INSIGHT STRATEGY ACTION

Thank you. Larry Goldman, President 773. 456. 3996 larry@amberleaf. net DATA INFORMATION INSIGHT STRATEGY ACTION

Amber. Leaf specializes in quickly breaking down the barriers that stand between your data and action. Social Networking and Customer Feedback – It’s all about the Data Presented by: Larry Goldman DAMA–MN April 16, 2008 DATA INFORMATION INSIGHT STRATEGY ACTION

Agenda • The New Data Hunger • Players • New Structures • Example DATA INFORMATION INSIGHT STRATEGY ACTION

360 Degree of the Customer - Example Organizations (UCM/MDR) Web Activity (Web Trends) Geography (Excel) Responses (CISPUB/PICK/ FRS/ARGI/SJO/ ATG, MPS, Unica, SPIN) Books (UPM/SPIN) Customer Adoptions (ATG, CISPUB/PICK/FRS) Journals (UPM) Alerts (ATG, SJO) Course (MDR/ FRS/PICK/ATG) DATA INFORMATION INSIGHT Orders/Quotes/ Usage (CISPUB/ PICK/ARGI/ MS CRM, MPS) STRATEGY ACTION

Internal Web Tracking • Software like Omniture, Unica Net. Insight, and Web. Trends provide metrics about your site: – Site Usage and Optimization – Standard metrics: page views, visits, and unique visitors – Marketing Conversion – Referral sites – Repeat visitor analysis DATA INFORMATION • Traditional software does not tell you: – Are your metrics good compared to your competitors? – Where are your customers when they are not on your site? – What are clients saying about you when they are not filling out your surveys? – What are your customers interested in beyond your products and services? – What are the major changes in the behavior of your industry’s on-line customers? – What are the major changes in the perspective of your customer base regarding your brand, new products, or service? INSIGHT STRATEGY ACTION

DW’s Start Chasing The Dream Again Look-a-like Prospects Non-vertical Web Behavior Business to Business Contacts Competitive Web Behavior Reviewed Products Customer Product/Company Ratings Credible Reviewers Cross Brand Analysis Net Promoter DATA INFORMATION INSIGHT Brand Attributes STRATEGY ACTION

Agenda • The New Data Hunger • Players • New Structures and Challenges • Example DATA INFORMATION INSIGHT STRATEGY ACTION

Competitive Web Metrics Element Definition Sources of Information ISP’s, Corporate licenses, proprietary toolbars/panels Panel size Vendors range between 500 K to over 3 million users Outliers Does the vendor correct for non-standard activity or coupon grabbing? Algorithm to understand overall audience size Typically, random dialing and surveys to understand the general population of the internet Projection Algorithm How does the vendor forecast actual web metrics based on their sample data? Search How detailed does the vendor capture search referrals and results? Auxiliary information Does the vendor supplement the click stream with attitudinal or demographic data? Regional organizations Is the information national in perspective or does it have enough representation locally? Vertical expertise Ability to compare functionality across different sites in the same vertical? DATA INFORMATION INSIGHT STRATEGY ACTION

Vendors • Nielsen: This advertising rating company collects information on the Web mostly focused on Media buyers to help organizations understand reach of different sites and where to place their media. Query tools are not great and custom reports are expensive. Nielsen contains B-to-B data as well as home PC user data. • Comscore: Comscore offers almost the exact same services as Nielsen. The focus is on media buying and reach rankings. Comscore contains international and domestic panels. • Hitwise: Hitwise tries to commoditize this marketplace by offering a large amount of canned reports across many different industries. Brittle and not much customization outside of the canned reports. • Compete: Boasts the largest panel of all the vendors and provides deep analytics in a subset of verticals across national ISP’s. Compete has indepth analysis on Search data as well as provides free general metrics on their web site. DATA INFORMATION INSIGHT STRATEGY ACTION

Web 2. 0 Metrics Element Definition Sources of Information How many blogs, how many sites, how far back? Scraping Technology How does the vendor scrape sites and can they extract individual fields, ratings, and recommendations? How does the technology react to site changes? Language Parser What is the origin of the text parser? Does it come from the off-line world? How was it trained? Sentiment Analysis Can the technology understand the difference between a positive and negative statement? Categorization Can you categorize different key words together as a specific concept? Can you customize the categories? Influencers Can you identify who is influencing the crowd as opposed to following the flock? Access to Detail Do you have access to the detail and ad-hoc query? Custom Roll Ups How does the analysis group roll up products and services? DATA INFORMATION INSIGHT STRATEGY ACTION

Vendors • Umbria: Does not provide direct access to the data, but provides deep dive analysis on a per project basis. • Chatter Guard: Focuses on the travel industry by reading all blog and feedback sites and manually scoring, rating, and categorizing the different items. • Clarabridge: Less a service than an ETL tool for unstructured information. Combines sentiment and language parser in order to provide key performance indicators on social networking sites. DATA INFORMATION INSIGHT STRATEGY ACTION

Agenda • The New Data Hunger • Players • New Structures and Challenges • Example DATA INFORMATION INSIGHT STRATEGY ACTION

New Data Sources/ETL Internal Systems • • • Third Party Demographics New outside vendor for competitive web analysis Some vendors will send extracts and others will only allow access New technology for interpreting unstructured Internet data DATA INFORMATION Outsourced Systems Competitive Web Blogs, Community Scraping, Language Parsing Standard ETL EDW Business intelligence INSIGHT STRATEGY ACTION

New Structures • New ETL challenges arise on how to match up the various customer level information – Internal data is at a very low level – External data will be aggregated – Screen names and e-mails may be available to consolidate in some cases • Data Modeling Efforts – Competitor. Need to normalize against your organization structure – Web behavior. Need to normalize against your site sections or taxonomy – Competitive Products. Need to normalize against your product hierarchy – New Customer Transactions. Ratings, reviews, pageviews, visits DATA INFORMATION INSIGHT STRATEGY ACTION

Data Quality Concerns • New subject areas – Competitive organizations – Competitive products – Competitive sites – Competitive behavior – Social Networking transactions DATA INFORMATION INSIGHT STRATEGY ACTION

Agenda • The New Data Hunger • Players • New Structures and Challenges • Example DATA INFORMATION INSIGHT STRATEGY ACTION

Available Insight • • • DATA Hotel – City – Name – Chain (where applicable) – Owner (where applicable) Author – Channel Score – how many postings and author makes – Credibility – other on-line members’ ratings on how helpful the author’s comments are – Velocity – how often an author posts – Influence – identifies if the author is leading or following the discussion Reviews – Author – Hotel – Date – Overall Rating for the hotel – Would the author recommend the hotel? – Rating for how good of a value the hotel was INFORMATION INSIGHT STRATEGY ACTION

Available Insight (Cont’d) • Hotel Metrics – Awareness – How much is being discussed – Favorability – Score that is a proxy for loyalty and how often the Hotel would be recommended by authors – Satisfaction - Based on overall ratings from reviews – Value – Based on value ratings from review – Net Promoter – based on promoters and detractors from the last year – Metrics can be sliced for influencers only or the entire Author base • Key Words – Score of how influential the key word is for that hotel – Metrics can be sliced for influencers only or the entire Author base DATA INFORMATION INSIGHT STRATEGY ACTION