LexisNexis Integration and Project Management Discussion David E.

LexisNexis Integration and Project Management Discussion David E. Glowacki Vice President, Risk Solutions Engineering Jo Ann Sheppard Senior Director, Risk Solutions Insurance Project Management IMF Presentation July 13, 2010 1

Who Are We? LexisNexis Parent company: Reed Elsevier Global reach: customers in more than 100 countries Employees: 18,000 globally Revenue: 4 billion in 2009 Searchable online documents: more than 5 billion Information sources: 40,000 Number of offices worldwide: 110 Risk Solutions Division Approximately 1000 Technology employees across North America, primarily Alpharetta, GA; Dayton; Ohio; Boca Raton, FL All 50 states, 70 percent of local government and almost 80 percent of federal agencies use our services to make the world a safer place. In addition, we support 90 percent of the Fortune 500 companies and the leading non-profits 300 technical projects annually Dave Glowacki – VP Product Engineering – Boca Raton, Florida Dave manages middleware (ESP) engineering, which provides a layer of abstraction between search and retrieval services and customer facing applications as well as centralized authentication and authorization services. In addition, Dave also manages the engineering project office including the technical integration of ChoicePoint. Jo Ann Sheppard – Senior Director Engineering Project Management – Alpharetta, Georgia Jo Ann leads the Engineering Project Management group and oversees the execution of engineering projects for the for the Insurance Data Services and Claims verticals. Prior to this role she led the integration activities for the Screening business unit. 2

Risk Solution Acquisitions Seisint – September 2004 (four year integration target) Better data content, better content licensing than LN. Superior linking technology to LN. HPCC technology offered greater processing at lower price. 4 terabytes of data to migrate, 92,000 bill groups, 20 major products, 200 million in revenue All web, batch and machine to machine applications converted to Seisint backend. Reverse integration, technically and culturally. User experience was a paradigm shift. ChoicePoint – September 2008 (five year integration target) Significant overlap in content and products in Public Records and Screening. Unique Insurance content and products 11 major Insurance systems to integrate, mostly machine to machine, systems not independent. Mainframe to HPCC migration PR products largely web and batch, 10k customers, 11 major products Risk culture and technology driven, seamless migration for Insurance customers desired. 3

Keys to Successful Integration 1. Pay Attention to Culture Specifically understand the resulting culture you want. Don’t ignore culture, it can seriously impede integration on many levels. 2. Manage Customer Migration You generally can’t integrate if you don’t migrate. The best technical solution is useless if not adopted by customers. 3. Clear Technical Direction (Rules and Approach) Technical direction should be clear at the start. Many debates and discussion can be eliminated with this clarity. Project management activities should be designed to support all three! 4

Engineering Culture We believe the desired culture drives many decisions and actions involved in an integration including project management process. What Culture Implies What we value in terms of people and skills. How we think about ourselves and our customers. Culture influences process, commitment, attitude, teamwork– the work environment. Our Target Culture Smart people not smart process. Flat organization, management is close to the work. Innovative and fast moving, HPCC enables us to do the impossible. Build it once and use it over and over again. How Effective pre-close planning. Integration planning began during diligence and continued into pre-close. Day 1 – organization set, leadership bought in, clear communication to employees, clear plan. On going decisions are balanced against culture. 5

Customer Migration We segment our customers not just by market and product but by degree of difficulty to migrate. Machine to machine migrations are often more difficult than Web. The greater the visibility the more difficult the migration. We clearly understand the product overlaps and differences. Customer migration planning begins at the same time as engineering design. Get pricing, authentication, authorization etc out of the way up front. Sales has to buy in! If it’s just about cost savings it can be very difficult to get sales to support they are “coin operated”. Sell internally, new features, better reliability, better results, future ability to enhance rapidly etc. Technology usually has to drive. dedicated business unit ownership for the customer migrations not someone’s partAssign time job. Since technology development comes first it will drive all downstream activities. 6

Customer Migration Use internal parallel testing and beta customer testing for large machine to machine migrations. Clearly defined beginning and end points for this type of testing. Watch for scope creep .customers like to add features they always wanted or not willing to adapt to new features Migrate in waves starting with low difficulty low risk customers first. Count on minor modifications as you go. Align technical deliverables with the waves, this distributes risk. Allow an ability to slow down and speed up. Count on a small handful of customers to be extremely difficult to migrate. Customer Communication Waves of communication aligned with migration waves Inform and generate excitement (may need to defuse angst) about what is coming but don’t box your project in with specific migration dates before you are really sure of the date. 7

Technical Integration Clear Architecture Ground Rules HPCC is the only alternative for large data – two data factories max. One back office system, authentication, authorization, accounting. Single middleware implementation. Two principal axis: migration of content to HPCC and leverage middleware to support component by component replacement. Collective Understanding/End State/Overarching Plan Top 10-15 technicians and managers (core) must understand the system (3 months) Diagram the end state for each major system. Pick the first product to migrate Smallest customer impact Largest infrastructure footprint Build the timeline off of that first product. Find a “Lazarus” so people will believe. Small modeling product that took almost two weeks to build, HPCC built it in two hours! 8

Middleware is the Key In Risk Solutions, practically every component in the system talks via middleware. (Customer products, back end components, tools, back office etc.) ESP – Enterprise Services Protocol is our middleware. It does the following: Translation, Transformation, Authentication, Authorization, Accounting (Log) Insurance CP had something similar called SIDEX protocol combined with a dedicated TSS switching bus. PLAN: - Build an ESP service that translates and transforms SIDEX/TSS By consistently introducing this layer between systems we can effectively abstract components and migrate parts of systems 9

Middleware Example End State Diagram Legacy Billing Modeling Server billing Reports tool - DPPA Inquiry DB Data export (BI for sales) 1 THOR (ET L) ROXIE - Business Rules - Queries - Caching ? - Score - CompID - Reformat - Screen Parse - Activity Files - DHDB/Dup - In-House User DB authentication logging 2 Request Predictor Request Score E MVR Rpt or Score S MVR Request P Vehicle Report Data QA & Load LZ Claims - Activity Files - In-house DB Predictor E 3 S DHDB/Dup Comm Blackbox DMVs or Court Data 10 Business rules and options Purge rules P Rules Engine batch requests authentication Batch R3 Driver Monitoring Admin RDBMS Dups suppressions suppress RDBMS - Hot Dup (1 day) - Suppressions Gateway – MVRs, state-based activity (M2M or screen-scrape) DMV or Third Party or Courts - Customer-specific business rules - State-specific options - Purge rules - Std. Violation Codes - Reference code translations - etc Customer Service Admin Tool 1. Thor and Roxie are the core ETL and real time query engines in Risk Solutions. This represent the single repository for Insurance content. 2. ESP middleware is introduced between all other systems and content 3. ESP middleware sits between external gateways and core content and systems. As we replace systems individual components may be a mix of new and old since ESP abstracts this. This allows us to manage risk.

Project Management’s Role PARTNER/PLAN FOCUS URGENCY Product Roadmaps - simple yet effective in showing progress, forward thinking and driving. Integration roadmap versus normal business roadmap. Aligned with customer migrations. DELIVER Clear Direction Aligns Customer Migration PMO drives urgency Resource Separation - integration resources separate from normal business resources. Focus on Integration Dedication leads to Clear Direction Annual Forecasting – one year roadmap including integration activities, forecast has quarter accuracy. Ties long term and short term together. New Product Request Process – manage requests for new projects that were not identified in annual planning. Priorities, roadmap updates. Consistent Commitment Model – matches the culture desired. Currently, quarter plus a quarter, month plus a month, day plus two weeks. Project Mgmt. is Grease- we make the machine run better, we don’t run the machine. Nobody knows anybody in beginning. Manages short term Balance between Integration and Normal Business. Clear commitments, Room 11 to Maneuver, Innovate, Adjust Matches our Culture

Color KEY: Blue Forecast/Requirements under Development Black Complete Red At Risk Yellow Behind with plan to address Green Active / On Track 12 Example Roadmap

Hurdles and Wins 13