How to increase Data Quality: What to Standardize and Centralize, and Where Does it Matter Most?

How to increase Data Quality: What to Standardize and Centralize, and Where Does it Matter Most? How do you get the data you need? And what do you do ...
Author: Ashlee Ellis
0 downloads 0 Views 257KB Size
How to increase Data Quality: What to Standardize and Centralize, and Where Does it Matter Most? How do you get the data you need? And what do you do with it, once you’ve got it? Dave Rocheleau, Sr. Manager, Human Resources Information Jane Dobie, Manager, Human Resources Analytics

About RBC Financial Group Canada’s largest financial institution: Canadian operations – 1,300 branches, 58,000 employees in Canada, 16,000 employees in US, 4,000 internationally, in Europe, Caribbean, Asia, and Australia, 800 branches internationally. US banking operations in North Carolina, Atlanta, and Florida, and country-wide mortgage sales/service Wealth management with RBC Dain Rauscher Capital/Equity markets operations in New York Life insurance operations with Liberty Insurance, South Carolina Market Capitalization - $39 Billion Revenues - $17 Billion

•1

HRIS History Implemented SAP HRIS in 1997, replacing legacy inhouse system. Implemented SAP Payroll in 2000. Major investments in Employee Self-Service (ESS), and Manager Self Service (MSS) My Information Updater Web-based Base Salary Decision, Short Term Incentive Decision tools, Applicant tracking, and E-learning systems Major drivers of data quality (when it counts)

Recent implementation of SAP’s Business Warehouse (BW) – discovered greater demand for quality of data.

Employee Self Serve

•2

Centralizing & Standardizing the Data Elimination of separate systems Continued investment in SAP development SAP is AUTHORITATIVE source for HR Data BW is the standard reporting system Same answer every time, standard definitions (example: Turnover – Voluntary/Involuntary)

Data quality education Working with Business HR, Service Partners, Platform leaders to understand importance of good data (P&L, Balance Sheet comparisons)

Process re-engineering, MSS investment

Appetite for Data Traditional approach – Saratoga metrics, benchmarking, “what are the other banks doing?” Increased complexity – regulatory, compliance based, overall need for: Better data (need accurate data to conduct analysis) Richer data (what don’t we have that we need) Wider data (from external sources, socio-economic and academic)

Some business/HR leaders starting to ask the tougher questions: What if? What about? What would happen if we… ?

•3

Creation of HR Analytics group New function for HR Demand driven – need for higher level of analysis in HR Group is in its infancy New hires – mathematics and statistics (50%) HR/Business knowledge (50%)

HR is evolving from 100% reporting, and little analysis Traditional – give us the data and we’ll troll through it until we find something interesting… Current – Starting some detailed analyses Future – HR will use the data to address the organization’s Human Capital requirements

Getting the Business to buy in (and HR) Choose real data issues and connect them to the bottom line – Benefits, Pension, Pay, Bonus Share the data with the Business Leader/HR Advisor Basic Demographics – age, gender etc. (HR ABC’s) Debunk some of the myths/conventional wisdoms – turnover, churn, movement Build upon small successes

Conduct the best analysis you can Get them hooked, and then have them help you drive the Data Quality even more.

•4

A Success Story (even with limited data) In 2001, RBC created a new destination / career sales role – Financial Planner (current incumbents came from 2 similar roles). Would be the pattern for future career roles Sales & Relationship account management Manage 200 – 300 clients Focus on building client base/portfolio/book size : Share of wallet Responsible for assets/credit books of $35 to $50 Million Compensation plan with dual tiers Asset/Credit Mix Book Size/Growth, including retention

Original Question – Can we determine “Time to Contribution? What we had: Book Size Position Tenure (Current) Position Tenure (previous position) Previous Position Geographic Region 5 quarters of performance Sales/Growth Target

What we did not have: Education/Accreditation (ESS field – not every employee has entered it) Credit/Investment Mix No information on target setting No information on seasonality No information about product demand – credit vs. investment

Combination of HR/Business data

We worked with what we could get.

•5

Initial Results Discovered that the research suggested the upper limit for book size should be lower than the current point where the book is split. Regardless of actual sales/growth target, performance averaged 2.3% with little variation Targets were not predictive of real growth (in some cases – negative correlation to growth for large books) Higher performing employees came from one role New hires significantly underperformed relative to peers

What did this mean? We were unable to answer the original question asked Identified additional opportunities to collect additional data Opened the door to having more discussions with the business Started to have conversations to: Take the initial findings and build on them with further analysis Consider changes to the role design to better meet business objectives Help the business understand potential key factors for success in the role Make potential design changes to the compensation plan

•6

What’s next for the Financial Planner Analysis? More analysis Raised more questions than we answered Identified additional data requirements – direct link to Data Quality

More conversations Business/Sales executive Sales Managers High performing employees, average performers HR partners – compensation, learning, talent management

What’s next for Human Capital Analytics More marketing to the Business Leaders and HR Additional analyses underway Looking at enterprise populations for Base Salary Increases, Short Term Incentive Payouts Studying large populations – destination roles versus stepping stone roles

Identifying importance of good data, and driving additional data quality requirements.

•7