Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Empowering the future of legal decision-making
© LexPredict 2012-2016
@lexpredict
www.lexpredict.com
contact@lexpredict.co...
Assessing capabilities and planning improvements
Data Strategy Maturity Model
Stage 2
the awakening
Stage 1
proto-(data)-c...
Stage 5
golden years
Capability Continuum
reporting to analytics
Time for discussion
Questions
What is a data strategy?
Statement and Framework
Data Strategy: Defined
A data strategy is a top-down mission statement acknowledging
the value of ...
D > I > K > W
From Data Strategy to Wisdom
Data
Information
Knowledge
Wisdom
Direct record of fact,
signal, symbol
Indirec...
Data Strategy Maturity Model
assessing capabilities and planning improvements
Data Strategy Maturity Model
Data Availability
Measurement Strategy
Reporting
Actionability
Maturity Assessment Dimensions
Stage One
proto-(data)-culture
There is no directory or
publishing of data
Data Availability
Recognition
Not recognized as data
Accessibility
Storage
Pub...
Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
Measurement Strategy
• Typically not present
• Typically not refined or tested
Reporting
• Anecdotal, story-based only
• Little to no ability to validate
Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
Actionability
Reporting and Assessment
No integration of data into
department reporting and
assessment
Early Warning Capab...
Stage Two
the awakening
• Little to no directory of
publishing
• Existing data is
unknown to others
Data Availability
Recognition
Increasing recog...
Measurement Strategy
• First iterations of measurement
and quality improvement begin
• Conception of metric-based
thinking...
Reporting
• First reports begin to emerge
• Reports are generally constructed
in Excel
• Reports are generally not availab...
Actionability
Reporting and Assessment
No integration of data into
department reporting and
assessment
Early Warning Capab...
Stage Three
the teenage years
• Little to no directory of
publishing
• Existing data is
unknown to others
Data Availability
Recognition
Increasing recog...
Measurement Strategy
• Master data management ideas begin
to emerge around measurement
standards and QC/QA
• Unstructured ...
Reporting
• Begins to take on a “designated”
role, with first reporting hires
• Experiments with reporting
frameworks or B...
Actionability
Reporting and Assessment
Data begins to be integrated
into department reporting and
assessment
Early Warning...
Stage Four
terra firma
Beginning of data directory
and inventory discussions
Data Availability
Recognition
Data is reliably
recognized
Accessibil...
Measurement Strategy
• Master data management standards
and roles emerge
• QA and QC begin to be applied to data
• Key met...
Reporting
• Dedicated reporting resources are
available
• Reporting frameworks and BI tools
receive increased investment a...
Actionability
Reporting and Assessment
Department reporting and
assessment incorporates data
Early Warning Capabilities
No...
Stage Five
the golden years
Data directory and
inventory are published
Data Availability
Recognition
Data is reliably
recognized
Accessibility
Storage...
Measurement Strategy
• Dedicated master data management
roles and standards are present
• QA and QC are reliably applied t...
Reporting
• Dedicated reporting resources are
available
• Reporting frameworks and BI tools
are available
• Reports integr...
Actionability
Reporting and Assessment
Department reporting and
assessment is driven by data
Early Warning Capabilities
Ea...
Tell you only what happened in the past
Historical Metrics
• Typically based on averages or aggregates, unconditioned on s...
Can tell you why or how something happened in the past
Historical Analytics
Identify Outliers
Maybe unique, maybe areas fo...
Can tell you what may happen in the future
Predictive Analytics
Predictive Capability
Both YES/NO and
probability/range
Pr...
Historical reporting
Examples from Legal
Question: What did we
spend on settlements and
legal expenses last quarter?
$1.2M...
Historical Analytics
Examples from Legal
Question: What factors
drove time to close
lease negotiations last
quarter?
Quest...
Examples from Legal
Predictive Analytics
Question: Should we settle this
dispute at outset?
• The counterparty is expected...
Historical reporting vs. historical/predictive analytics
Varying Skillsets
Historical Reporting
• Database administrators
...
Baseline strategic decisions and resources against models
Baseline Comparisons
Decision-makers
• Doesn’t mean autopilot
• ...
Need to baseline or adjust for counterfactuals
Counterfactuals
• Example: We lost negotiations on renewal term provisions ...
LexPredict: Legal Data Strategy Maturity
https://www.lexpredict.com
@lexpredict
Thank you for reading.
Empowering the future of legal decision-making
LexPredict Background
Importing best-in-class processes
and technologies fr...
Spanning the breadth of business
LexPredict Offerings
Consulting Services
Process, technology, data, and strategy
Data Pro...
Empowering the future of legal decision-making
© LexPredict 2012-2016
@lexpredict
www.lexpredict.com
contact@lexpredict.com
Upcoming SlideShare
Loading in …5
×

Legal Data Strategy Maturity: Assessing capabilities and planning improvements

2,167 views

Published on

LexPredict presentation on Legal Data Strategy Maturity Models: Assessing capabilities and planning improvements. The talk focused on assessing current state maturity and planning future state improvements for data-driven capabilities for law firms, corporate legal, risk, and compliance.

Published in: Law
  • Be the first to comment

Legal Data Strategy Maturity: Assessing capabilities and planning improvements

  1. 1. Empowering the future of legal decision-making © LexPredict 2012-2016 @lexpredict www.lexpredict.com contact@lexpredict.com Assessing Legal Data Strategy Maturity prepared for: Public Distribution prepared on: Aug 2016
  2. 2. Assessing capabilities and planning improvements Data Strategy Maturity Model Stage 2 the awakening Stage 1 proto-(data)-culture Stage 3 the teenage years Stage 4 terra firma
  3. 3. Stage 5 golden years Capability Continuum reporting to analytics Time for discussion Questions
  4. 4. What is a data strategy?
  5. 5. Statement and Framework Data Strategy: Defined A data strategy is a top-down mission statement acknowledging the value of an organization’s data combined with a framework for developing data-related capabilities. While data strategies are built on lists of principles and technologies, they address much more: strategic communication and change management, process improvement, knowledge management, and risk management, to name a few.
  6. 6. D > I > K > W From Data Strategy to Wisdom Data Information Knowledge Wisdom Direct record of fact, signal, symbol Indirect record or description Interpretation of information Actionable inference or heuristic Data-Information-Knowledge-Wisdom Data Readings from a temperature sensor in Tahoe. Information The average temperature in the month of December is 32.2F. Knowledge Snow is likely to accumulate in December. Wisdom January is a good month to plan a ski trip to Tahoe.
  7. 7. Data Strategy Maturity Model assessing capabilities and planning improvements
  8. 8. Data Strategy Maturity Model Data Availability Measurement Strategy Reporting Actionability Maturity Assessment Dimensions
  9. 9. Stage One proto-(data)-culture
  10. 10. There is no directory or publishing of data Data Availability Recognition Not recognized as data Accessibility Storage Publication Data is not accessible Data is not stored Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  11. 11. Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Measurement Strategy • Typically not present • Typically not refined or tested
  12. 12. Reporting • Anecdotal, story-based only • Little to no ability to validate Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  13. 13. Actionability Reporting and Assessment No integration of data into department reporting and assessment Early Warning Capabilities No early warning capabilities Predictive Capabilities No data-driven predictive capabilities Project Management No integration of data into project management Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  14. 14. Stage Two the awakening
  15. 15. • Little to no directory of publishing • Existing data is unknown to others Data Availability Recognition Increasing recognition of data Accessibility Storage Publication Data is typically difficult to access Data is stored, but typically in diverse and non-normalized manners Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  16. 16. Measurement Strategy • First iterations of measurement and quality improvement begin • Conception of metric-based thinking begins to emerge Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  17. 17. Reporting • First reports begin to emerge • Reports are generally constructed in Excel • Reports are generally not available to users outside of small requesting group Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  18. 18. Actionability Reporting and Assessment No integration of data into department reporting and assessment Early Warning Capabilities No early warning capabilities Predictive Capabilities No data-driven predictive capabilities Project Management No integration of data into project management Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  19. 19. Stage Three the teenage years
  20. 20. • Little to no directory of publishing • Existing data is unknown to others Data Availability Recognition Increasing recognition of data Accessibility Storage Publication Data is stored, and discussions around normalization and warehousing begin • Companies begin to identify and store unstructured data sources • Data begins to be formally databased, with increasing accessibility to technical users Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  21. 21. Measurement Strategy • Master data management ideas begin to emerge around measurement standards and QC/QA • Unstructured data approaches begin to emerge • Existing metrics begin to crystallize • New metrics begin to emerge an accelerating rate Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  22. 22. Reporting • Begins to take on a “designated” role, with first reporting hires • Experiments with reporting frameworks or BI applications begin • Reports begin to be distributed outside of small groups Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  23. 23. Actionability Reporting and Assessment Data begins to be integrated into department reporting and assessment Early Warning Capabilities No early warning capabilities Predictive Capabilities First discussions of data-driven predictive capabilities begin Project Management Integration of data into project management is first discussed Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  24. 24. Stage Four terra firma
  25. 25. Beginning of data directory and inventory discussions Data Availability Recognition Data is reliably recognized Accessibility Storage Publication Unstructured data begins to be processed for features • Data is reliably stored • Data normalization and warehousing become focal point Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  26. 26. Measurement Strategy • Master data management standards and roles emerge • QA and QC begin to be applied to data • Key metrics are crystallized and understood across organization • Most “low-hanging” data has been identified, even if not yet available or measured Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  27. 27. Reporting • Dedicated reporting resources are available • Reporting frameworks and BI tools receive increased investment and focus • Reports are regularly distributed to wider audiences Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  28. 28. Actionability Reporting and Assessment Department reporting and assessment incorporates data Early Warning Capabilities No early warning capabilities Predictive Capabilities First experiments with data- driven predictive capabilities begin Project Management Integration of data into project management begins Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  29. 29. Stage Five the golden years
  30. 30. Data directory and inventory are published Data Availability Recognition Data is reliably recognized Accessibility Storage Publication • Data normalization and warehousing are implemented • Data normalization and warehousing are implemented Data is reliably stored, both from structured and unstructured sources Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  31. 31. Measurement Strategy • Dedicated master data management roles and standards are present • QA and QC are reliably applied to data • Key metrics are crystallized and understood across the organization • “Next wave” metrics emerge, often from strategic thinking • e.g., strategic recognition of “customer experience” and subsequent measurement Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  32. 32. Reporting • Dedicated reporting resources are available • Reporting frameworks and BI tools are available • Reports integrate external data sources (e.g., industry or region baselines) • Reports are regularly distributed to wider audiences Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  33. 33. Actionability Reporting and Assessment Department reporting and assessment is driven by data Early Warning Capabilities Early warning capabilities begin to develop Predictive Capabilities Data-driven predictive capabilities begin to drive decision-making Project Management Project management is driven by data Process Improvement Process improvement discussions around workflows emerge Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
  34. 34. Tell you only what happened in the past Historical Metrics • Typically based on averages or aggregates, unconditioned on specific facts • Good view of the forest health, but hard to see if individual trees/patches are doing well • Typically show running tallies or only incorporate COMPLETED records • Often difficult to translate into direct, actionable decisions on the ground Not why Not how No information about the future
  35. 35. Can tell you why or how something happened in the past Historical Analytics Identify Outliers Maybe unique, maybe areas for improvement or investigation Not future-oriented Can’t necessarily provide information about the future • Ability to present more specific, focused analysis • Ability to analyze both the forest and the trees • Typically based on statistical methods from causal inference
  36. 36. Can tell you what may happen in the future Predictive Analytics Predictive Capability Both YES/NO and probability/range Predictions Ability to predict outcome for SINGLE events Application Ability to reformulate as KNOWLEDGE or guidelines/policy Machine Learning Typically based on machine learning methods
  37. 37. Historical reporting Examples from Legal Question: What did we spend on settlements and legal expenses last quarter? $1.2M Question: On average, how many effort hours does staff counsel spend on the discovery phase of a non- compete dispute? 25 hours
  38. 38. Historical Analytics Examples from Legal Question: What factors drove time to close lease negotiations last quarter? Question: What factors drove legal and compliance expenses last quarter?
  39. 39. Examples from Legal Predictive Analytics Question: Should we settle this dispute at outset? • The counterparty is expected to accept an initial offer • The dispute is predicted to settle for $100k, with legal expenses of $15k • If an initial offer is not made, this dispute is expected to cost $50k in legal expenses and has a 25% chance of going to jury trial. Question: How many staff counsel effort hours do we expect to spend on this negotiation? • An estimate of 18 hours, with 90% confidence that the dispute will fall between 13 and 30 hours
  40. 40. Historical reporting vs. historical/predictive analytics Varying Skillsets Historical Reporting • Database administrators • SQL Developer • Reporting developer • Cognos • BusinessObjects • Teradata Historical/Predictive Analytics • Data scientists and statisticians • Data analysts • R, Python, SAS developers • Chief Data Officer
  41. 41. Baseline strategic decisions and resources against models Baseline Comparisons Decision-makers • Doesn’t mean autopilot • Suggested decisions and justification or data should be presented to users • Users have an obligation to either: • Follow recommendation – agency/decision primarily owned by model • Reject recommendation – provide justification and explanation, incorporate into future model iterations Analogy: Pilots Pilots don’t get to choose any flight path; flight planning systems present optimized routes and pilots have to request an override Analogy: CPAs An individual accountant doesn’t get to choose what GAAP is. If they decide to calculate something in a way that doesn’t comply with the model, they have to provide a justification to the auditors.
  42. 42. Need to baseline or adjust for counterfactuals Counterfactuals • Example: We lost negotiations on renewal term provisions in 6/10 deals this quarter. Is this good or bad? • If base rate is 3/10,this is probably good. • If base rate is 8/10,this is probably bad. • Example: Our average outside counsel rate rose by 1% this year. Is this good or bad? • If base rate is 4% increase, then this is probably good • If base rate is -2%, then this is probably bad. 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 2012 2013 2014 2015 Arbitration Mediation Litigation
  43. 43. LexPredict: Legal Data Strategy Maturity https://www.lexpredict.com @lexpredict Thank you for reading.
  44. 44. Empowering the future of legal decision-making LexPredict Background Importing best-in-class processes and technologies from leading global industries LexPredict is an enterprise legal technology and consulting firm, specializing in the application of best-in-class processes and technologies from the technology, financial services, and logistics industries to the practice of law, compliance, insurance, and risk management. We focus on the goals of prediction, optimization, and risk management to enable holistic organizational changes that empower legal decision-making. These changes span people and processes, software and data, and execution and education.
  45. 45. Spanning the breadth of business LexPredict Offerings Consulting Services Process, technology, data, and strategy Data Products Historical and real-time data products Training Programs Standard and custom training on-site and online Communication Services Change management and marketing Software Services Custom legal and statistical software solutions
  46. 46. Empowering the future of legal decision-making © LexPredict 2012-2016 @lexpredict www.lexpredict.com contact@lexpredict.com

×