how data.com drives data quality
TRANSCRIPT
How Data.com Drives Data Quality in the Cloud A Complete Breakdown of Salesforce.com’s Source of High-Quality Account and Contact Data
Safe harbor Safe harbor statement under the Private Securities Litigation Reform Act of 1995: This presentation may contain forward-looking statements that involve risks, uncertainties, and assumptions. If any such uncertainties materialize or if any of the assumptions proves incorrect, the results of salesforce.com, inc. could differ materially from the results expressed or implied by the forward-looking statements we make. All statements other than statements of historical fact could be deemed forward-looking, including any projections of product or service availability, subscriber growth, earnings, revenues, or other financial items and any statements regarding strategies or plans of management for future operations, statements of belief, any statements concerning new, planned, or upgraded services or technology developments and customer contracts or use of our services. The risks and uncertainties referred to above include – but are not limited to – risks associated with developing and delivering new functionality for our service, new products and services, our new business model, our past operating losses, possible fluctuations in our operating results and rate of growth, interruptions or delays in our Web hosting, breach of our security measures, the outcome of any litigation, risks associated with completed and any possible mergers and acquisitions, the immature market in which we operate, our relatively limited operating history, our ability to expand, retain, and motivate our employees and manage our growth, new releases of our service and successful customer deployment, our limited history reselling non-salesforce.com products, and utilization and selling to larger enterprise customers. Further information on potential factors that could affect the financial results of salesforce.com, inc. is included in our annual report on Form 10-K for the most recent fiscal year and in our quarterly report on Form 10-Q for the most recent fiscal quarter. These documents and others containing important disclosures are available on the SEC Filings section of the Investor Information section of our Web site. Any unreleased services or features referenced in this or other presentations, press releases or public statements are not currently available and may not be delivered on time or at all. Customers who purchase our services should make the purchase decisions based upon features that are currently available. Salesforce.com, inc. assumes no obligation and does not intend to update these forward-looking statements.
What is Data Quality and Why is it Important?
Quality Customer Data is Hard to Find & Maintain
Uploading and Integrating It Into Your Sales App
Keeping Data Up-to-Date and Current
Prospecting Across Many Applications
B2B Sales Rep time consumed doing research
21% Tech issue with CRM is
consolidating customer data
#1 Phone numbers change
every 30 minutes
75
(Sources: CSO Insights; Forrester Research/CustomerThink; Dun & Bradstreet)
The 3 Dimensions of Data Quality
Account Data Contact Data
Duplication
Completeness Accuracy
Account Data
Complete Contacts
Real-Time Updates
Automated Clean
Duplicate Management
Data Quality Reporting
Mobile
§ Complete Contact Information from Data.com Connect
§ Global Accounts Provided by Dun & Bradstreet
§ Real-Time Updates of Fresh Data
§ Automated, Bulk Clean of Incomplete and/or Dirty Data
§ No More Duplicates with Duplicate Management
Patented Processes for Quality For Contact & Company Data
Tens of millions of records uploaded each month
Best in class technology, algorithms & partnerships process, clean & validate records
Over 2 million members from the Data.com Connect community update 1M+ records per month
Patented process for maintaining & updating global account database
① Global Data Collection ② Entity Matching ③ D-U-N-S® Number ④ Corporate Linkage ⑤ Predictive Indexing
Over 2,000 automated quality checks & balances to ensure D&B’s high quality
standards
Accurate data on companies & people
200K sources worldwide 1.5M daily database updates
Team of Data Stewards, Analysts & Scientists maintain data quality
Data Stewards
Contact Data Quality Through Data.com Connect
Managed crowdsourcing by trusted community members 2M members exchange B2B contact information § Growing community of trusted members with shared interest in B2B contact data § 40+ Million business professionals from around the world
Community Technology Data Stewards
Incentives with patented gamification
techniques
Dedicated team monitoring data
submissions and usage
Patented approaches to screen and maintain
contacts
Data.com Connect data management techniques Industry leading combination of crowdsourcing, technology, and stewardship
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Community member adds new B2B contact
Technology validates and normalizes at time of contribution
Additional verification by data stewards
Ongoing validation through technology (i.e. 90 day email deliverability checks)
Ongoing updates by the community
for title and address changes,
etc.
1 2
3
4
5
Data.com Connect Model is Unique Deeper set of contacts, updated more frequently, with high data quality
Data.com Connect data management techniques Industry leading combination of crowdsourcing, technology, and stewardship
Social Rankings Social Incentives Self Policing
Points for Participation: § Give a contact § Update a record § Mark a record as
inactive
Ranking Authority: § 4 levels of hierarchy § Tiered contributions § Value for Participation § Rewards
§ Dedicated Data Defenders
§ Report abuse and lock accounts
§ > 12k updates/day
Community
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Data Stewards Continuously Monitoring Data Quality Data Quality Not Dictated by Community but Managed by Dedicated Team of Data Stewards
Data Stewards § Reviews some
records § Hypothesize patterns § Take action on sets
Data Analysts § Analyze large data sets § Prove / disprove
hypothesis § Look for automated
approaches
Data Scientists § Create foundation for
machine learning, automated detection, and alert work flows
Software Engineers § Build automation
solutions that increase efficiency and effectiveness
Data Stewards
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
State-of-the-Art Data Quality Technology Every Contact Undergoes Multiple Checks Per Data Field
Company • Validate 3rd party • Crowdsourced feedback
Name • Normalize • Standardizes • Gibberish Detection
Email • Validate 3rd party for
deliverability • Bounce report checks • Crowdsourced feedback
Phone • Country Normalization • External verification and
updates • Crowdsourced feedback
Title • Normalize • Standardize • Gibberish detection • Profanity detection
Address • Standardization • Deliverability check
Salesforce.com has multiple patents granted or pending on use of crowdsourcing and fact-based assessments for data management.
Technology
100% Phone Number and Email Address Coverage Customers typically experience
over 80% email deliverability 1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Guaranteed Complete Contact Information 40+ Million Contacts – each one complete with a phone number and email address
94% email deliverability
87% email deliverability
93% email deliverability
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Customer Success Across All Segments and Industries Data.com Customers Driving Business With High-Quality Data
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
COMMUNITY § New Rookie tier to raise trust barrier to entry
§ Tune member ratings to incent members to maintain contributed contacts
§ Focused Community Contests – ie. direct dials for IT decision maker contacts
The Data.com Connect Community is Constantly Evolving New Community Gamifications and Processes to Encourage More Data Entries and Better Data Quality
Community
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
DATA STEWARDS § Additional automation to enable review of more community
contributions
§ Pilot and prove technology innovations in parallel with product roadmap
New Methods to Data Stewards Monitoring Data Quality The Data.com Connect Team is Constantly Evolving
Data Stewards
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
TECHNOLOGY § Additional data quality checks ie. phone number, ownership
§ Incorporate data quality signals from web data mining
§ Continual retaining of algorithms ie. title normalization and classification
Technology Continuously Validating and Improving The Data.com Connect Platform is Constantly Evolving
Technology
Account Data Quality Through Dun & Bradstreet
Global Company Information From D&B
The Industry’s Gold Standard in Account Data
§ 233M+ companies across 200+ countries
§ 30,000 sources worldwide
§ 80+ fields of data, including D-U-N-S Number & Corporate Linkages
§ 1,000 Industries
§ 3 Million Corporate Family Trees with 17+ Million Companies Part of a Family Tree
The D&B Data Strategy Leveraging proprietary data, customer generated information, and global sources.
Global Data Sources Customer Generated Data
• Public Records
• Web Mining
• News
• Registries
• Trade Payments
• Third Party Sources
• Financial Reports
• Customer File Sharing
• IQ Feedback
• Interactive Products
Proprietary D&B Data Collection
Confidential 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 1993
Global data file created 2000
More than doubled file size in 7 yrs 2005
Worldwide Network launched Today
More than 233M
28M DUNS records
58M
100M
233M
~ 250M
~2015 Estimated 250M
Focused investments by D&B continue to rapidly increase coverage of the world’s businesses, with the majority of growth coming from the SMB sector and emerging markets.
D&B’s Global Account Coverage D&B’s mission is to fuel global commerce – and they are committed to delivering real, meaningful visibility on enterprises worldwide. To do this, D&B established the D&B Worldwide Network (WWN) - an unparalleled alliance of leading commercial information providers covering more than 190 countries.
D&B Owned Entities
D&B Strategic Partnerships
D&B’s Global Account Coverage D&B continuously works to reinforce their presence throughout the world by leveraging directly-owned D&B subsidiaries alongside market-leading partner companies.
D&B Owned Entities
D&B Strategic Partnerships
2011 Greece - In May, D&B forged a strategic relationship in Southern Europe with ICAP, the premier provider of business information in Greece, Bulgaria and Romania.
2009 Italy - D&B entered a strategic partnership with CRIF, based in Bologna, Italy, the leading banking credit information provider in continental Europe.
2008 Russia - D&B established a new joint venture with Interfax – the leading provider of business information in Russia and the former Soviet Union.
• 29M Active US Businesses • 106M Active OUS Businesses • 220+ countries • 5M daily refreshes • 1B annual trade updates
Trade Data (12K sources)
Court & Legal Filings
Business Internet Data
Business Registries
Newspapers & Publications
Telephone Interviews
Company Financials
D&B Customer Experiences
Banking Information
Mor
e th
an 3
0,00
0 G
loba
l Dat
a S
ourc
es
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Global Data Collection brings together data from over 30,000+ sources worldwide Over 30,000 Global Data Sources
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Over 30,000 Global Data Sources North America
Courts & Legal Filings Offices
Business Internet
Sites
Public Company Financial Filings
Telephone Interviews Dun & Bradstreet
Customer Experiences
Banking Information
Payment Experiences Postal Address Changes
Government Registries Yellow Pages
NORTH AMERICA DATABASE
Newspapers & Publications
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Over 30,000 Global Data Sources Europe
Business Internet Sites Insolvency Records
Company Financials (public and private)
Financial Statements/ Annual Returns
Courts & Tribunals
Payment Experiences
Business Registries
Telephone Interviews
Business Publications & News Media
Dun & Bradstreet Customer
Experiences
EUROPE DATABASE
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Over 30,000 Global Data Sources Asia Pacific
Local Directories News Papers & Publications
People's Court Newspaper
Business Internet
Sites
Administration of Industry and Commerce
National Part-Time Network for Court
Dun & Bradstreet Customer
Experiences
Shanghai and Shenzen Stock Exchange
Chinainfobank.com DUNtrade Participants
China Conformity Assessment
Direct Interview with
Businesses
Law Firms
ASIA DATABASE
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Over 30,000 Global Data Sources South America
Business Internet
Sites
Local Directories, Newspapers, Publications
Official Business Gazettes
Information Partners on Location
Banking Information
Courts & Tribunals
Payment Experiences
(Electronically)
Local Mercantile Registries
Direct Interview with
Businesses
Chamber of Commerce & Other ASC.
LATIN AMERICA DATABASE
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Over 30,000 Global Data Sources Middle East
Correspondents on Location
Registry of Companies or Ministry of Commerce
Company Name
Start Date Location
(Physical or Mailing Address)
Registration Details
Telephone Interview With Businesses
Company Name
Full Address Telephone
Fax Line of Business
Telephone Sales
Employees CEO
MIDDLE EAST DATABASE
Dun & Bradstreet Customer
Experiences
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Africa Over 30,000 Global Data Sources
News Articles
Local Telephone Directories: Telkom, Brabys
and Cybertel
Company Name Address
Telephone Number Fax Number
The Registrar's Office Registration Data
Direct Interview with Businesses
Business Activity Market Territory
Premises Branches Divisions
Employees Affiliations
Principals' Antecedents Expansion Plans Balance Sheets
Profit & Loss Projected Sales
Confirmation on New Articles
AFRICA DATABASE
Correspondents on Location
Dun & Bradstreet Customer
Experiences
The five essential elements of insight:
DUNSRightTM
Global Data Collection
Entity Matching
D-U-N-S Number
Corporate Linkage
Predictive Indicators
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
D&B’s patented method of collecting, integrating and organizing commercial data The DUNSRightTM Process
A.B.C. 123 Elm St. Bethlehem, PA 18025 (SIC) 5411-02 Chuck Smith, President
Multiple unmatched records
Alpha Plastics P.O. Box 111 Bethlehem, PA 18055 (SIC) 5411-02 Chuck Smith, President 610 882-7600
Charles Smith 123 Elm St. Bethlehem, PA 18025 (SIC) 5541-99 215 882-7600
D&B D-U-N-S Number: 12-345-6780 ABC, Inc + Alpha Plastics P.O. Box 111 Bethlehem, PA 18055 123 Elm St. Bethlehem, PA 18025 SICs: 5411 0202, 5541 9901 Key Executives Charles Smith, President Mary Worth, VP Marketing Chris White, Controller Phone Number: 610 882-7600
Single view of your customer
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Combining data into a single business, for a complete picture of customers and prospects D&B’s Patented Data Quality Process
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Data processes keep pace with global changes like mergers & acquisitions as they occur Processes that React to Business Changes in Real Time
Overview of D&B Corporate Linkage Maintenance Process
• Monitoring of news sources around the globe
• Third party sources help ensure merger and acquisitions, divestitures, bankruptcies and other significant events are correctly reflected in the database as they occur
• Automated process proactively sweeps global database to find and resolve “look-alikes”
• Daily maintenance process to detect and resolve errors globally
• Manual “Top-Down” review of “Top 12,000” trees annually
• Continuous accuracy measurement; focused on driving overall accuracy for the Top 12,000 trees to 90%+
90%
91%
92%
97%
97%
98%
98%
80% 100%
Address Deliverability
Operational Status
Employees (modeled and actual)
Year Started
SIC
Business Name
One level up linkage
U.S. Data Element Accuracy
D&B Data Accuracy Test, December 2013
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
D&B constantly tests the quality of their foundational, firmographic and associated data Account Data You Can Trust
Goals for the Data Standard Program – Top 12 Markets Global Standards Targets
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
90% of the active records in each market will contain full completeness in Identity Data Elements (Business Name, Address, Telephone, Primary Contact Name, SIC)
60-80% of the active records in each market will contain full completeness in Firmographic Data Elements (Sales or Employees and Year Started)
COVERAGE
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
>90% Accuracy of Identity Elements (Business Name, Address, Telephone, Primary Contact Name, SIC)
100% of Failure and Delinquency Scores meet predictive performance standards and are monitored annually
ACCURACY
Goals for the Data Standard Program – Top 12 Markets Global Standards Targets
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
INSIGHTS 45% of the active records will contain 1+ Trade Experience
30-40% of the active records will contain 3+ Trade or Trade Proxies
20-30% of the active records will contain Financial Attributes or Proxies 20% of the active records will contain Family Tree Linkage
Global Standards Targets Goals for the Data Standard Program – Top 12 Markets
1. Data Quality Sources
2. Data Quality Model
3. Data Quality Accuracy
4. Data Quality Success
5. Continuous Improvements
Validations
Standards
Testing
Continuous Improvement
Manual Review
Continuous Monitoring
Thousands of automated validations used to prevent incorrect data from entering database
Information policies in place for globally-consistent data processing and standardized formatting
Independent third parties routinely test and measure accuracy of data elements
Manual reviews of data ensure accuracy and identify process improvement opportunity areas
Data partners are monitored and held to quality standards; Proactive database sweeps identify trends
Goals are developed for improvement so that success can be measured
Q
ualit
y A
ssur
ance
Data Governance
Rigorous Quality Assurance Programs Are Constantly At Work and Improving The Data D&B Compiles Should Not Be Accepted at Face-Value
Privacy Processes and Standards with Our Data
“All contact data in Data.com Connect comes from community members or public sources and never from customer data managed in Salesforce.
Parker Harris EVP, Technology salesforce.com
Nothing is more important to our company than the privacy
of our customer’s data.”
Customers are responsible for their usage of data purchased through Data.com:
§ Customers are responsible for complying with all applicable regulations, e.g. Canadian Anti-Spam legislation (CASL) and CAN-SPAM
§ According to CASL: § Customers must have an opt-in relationship directly with
Canadian customers/contacts/leads to send them Commercial Electronic Message (CEM) such as an email
§ Customers need to manage CEM consent and need to show proof of oral consent, written consent, business relationship, or no business relationship with Canadian customers/contacts/leads.
Community based on trust and compliance
How Data.com Connect complies with industry standards and best practices:
§ Connect sends an opt-out notification email to new, validated contacts after they are added
§ A new contact has a 14-day period to opt out before their contact information goes live
§ Contacts can easily update their own info, manage contact preferences, or opt-out at any time
§ Connect follows best practices for email management and is CAN-SPAM compliant
