Category Archives: Master Data Management
“Inaccurate, inconsistent and disconnected supplier information prohibits us from doing accurate supplier spend analysis, leveraging discounts, comparing and choosing the best prices, and enforcing corporate standards.”
This is quotation from a manufacturing company executive. It illustrates the negative impact that poorly managed supplier information can have on a company’s ability to cut costs and achieve revenue targets.
Many supply chain and procurement teams at large companies struggle to see the total relationship they have with suppliers across product lines, business units and regions. Why? Supplier information is scattered across dozens or hundreds of Enterprise Resource Planning (ERP) and Accounts Payable (AP) applications. Too much valuable time is spent manually reconciling inaccurate, inconsistent and disconnected supplier information in an effort to see the big picture. All this manual effort results in back office administrative costs that are higher than they should be.
Do these quotations from supply chain leaders and their teams sound familiar?
“We have 500,000 suppliers. 15-20% of our supplier records are duplicates. 5% are inaccurate.”
“I get 100 e-mails a day questioning which supplier to use.”
“To consolidate vendor reporting for a single supplier between divisions is really just a guess.”
“Every year 1099 tax mailings get returned to us because of invalid addresses, and we play a lot of Schedule B fines to the IRS.”
“Two years ago we spent a significant amount of time and money cleansing supplier data. Now we are back where we started.”
Please join me and Naveen Sharma, Director of the Master Data Management (MDM) Practice at Cognizant for a Webinar, Supercharge Your Supply Chain Applications with Better Supplier Information, on Tuesday, July 29th at 10 am PT.
During the Webinar, we’ll explain how better managing supplier information can help you achieve the following goals:
- Accelerate supplier onboarding
- Mitiate the risk of supply disruption
- Better manage supplier performance
- Streamline billing and payment processes
- Improve supplier relationship management and collaboration
- Make it easier to evaluate non-compliance with Service Level Agreements (SLAs)
- Decrease costs by negotiating favorable payment terms and SLAs
I hope you can join us for this upcoming Webinar!
“If I use master data technology to create a 360-degree view of my client and I have a data breach, then someone could steal all the information about my client.”
Um, wait, what? Insurance companies take personally identifiable information very seriously. The statement is flawed in the relationship between client master data and securing your client data. Let’s dissect the statement and see what master data and data security really mean for insurers. We’ll start by level setting a few concepts.
What is your Master Client Record?
Your master client record is your 360-degree view of your client. It represents everything about your client. It uses Master Data Management technology to virtually integrate and syndicate all of that data into a single view. It leverages identifiers to ensure integrity in the view of the client record. And finally it makes an effort through identifiers to correlate client records for a network effect.
There are benefits to understanding everything about your client. The shape and view of each client is specific to your business. As an insurer looks at their policyholders, the view of “client” is based on relationships and context that the client has to the insurer. This are policies, claims, family relationships, history of activities and relationships with agency channels.
And what about security?
Naturally there is private data in a client record. But there is nothing about the consolidated client record that contains any more or less personally identifiable information. In fact, most of the data that a malicious party would be searching for can likely be found in just a handful of database locations. Additionally breaches happen “on the wire”. Policy numbers, credit card info, social security numbers, and birth dates can be found in less than five database tables. And they can be found without a whole lot of intelligence or analysis.
That data should be secured. That means that the data should be encrypted or masked so that any breach will protect the data. Informatica’s data masking technology allows this data to be secured in whatever location. It provides access control so that only the right people and applications can see the data in an unsecured format. You could even go so far as to secure ALL of your client record data fields. That’s a business and application choice. Do not confuse field or database level security with a decision to NOT assemble your golden policyholder record.
What to worry about? And what not to worry about?
Do not succumb to fear of mastering your policyholder data. Master Data Management technology can provide a 360-degree view. But it is only meaningful within your enterprise and applications. The view of “client” is very contextual and coupled with your business practices, products and workflows. Even if someone breaches your defenses and grabs data, they’re looking for the simple PII and financial data. Then they’re grabbing it and getting out. If the attacker could see your 360-degree view of a client, they wouldn’t understand it. So don’t over complicate the security of your golden policyholder record. As long as you have secured the necessary data elements, you’re good to go. The business opportunity cost of NOT mastering your policyholder data far outweighs any imagined risk to PII breach.
So what does your Master Policyholder Data allow you to do?
Imagine knowing more about your policyholders. Let that soak in for a bit. It feels good to think that you can make it happen. And you can do it. For an insurer, Master Data Management provides powerful opportunities across everything from sales, marketing, product development, claims and agency engagement. Each channel and activity has discreet ROI. It also has direct line impact on revenue, policyholder satisfaction and market share. Let’s look at just a few very real examples that insurers are attempting to tackle today.
- For a policyholder of a certain demographic with an auto and home policy, what is the next product my agent should discuss?
- How many people live in a certain policyholder’s household? Are there any upcoming teenage drivers?
- Does this personal lines policyholder own a small business? Are they a candidate for a business packaged policy?
- What is your policyholder claims history? What about prior carriers and network of suppliers?
- How many touch points have your agents and had with your policyholders? Were they meaningful?
- How can you connect with you policyholders in social media settings and make an impact?
- What is your policyholder mobility usage and what are they doing online that might interest your Marketing team?
These are just some of the examples of very streamlined connections that you can make with your policyholders once you have your 360-degree view. Imagine the heavy lifting required to do these things without a Master Policyholder record.
Fear is the enemy of innovation. In mastering policyholder data it is important to have two distinct work streams. First, secure the necessary data elements using data masking technology. Once that is secure, gain understanding through the mastering of your policyholder record. Only then will you truly be able to take your clients’ experience to the next level. When that happens watch your revenue grow in leaps and bounds.
“Not only do we underestimate the cost for projects up to 150%, but we overestimate the revenue it will generate.” This quotation from an Energy & Petroleum (E&P) company executive illustrates the negative impact of inaccurate, inconsistent and disconnected well data and asset data on revenue potential.
“Operational Excellence” is a common goal of many E&P company executives pursuing higher growth targets. But, inaccurate, inconsistent and disconnected well data and asset data may be holding them back. It obscures the complete picture of the well information lifecycle, making it difficult to maximize production efficiency, reduce Non-Productive Time (NPT), streamline the oilfield supply chain, calculate well by-well profitability, and mitigate risk.
To explain how E&P companies can better manage well data and asset data, we hosted a webinar, “Attention E&P Executives: Streamlining the Well Information Lifecycle.” Our well data experts Stephanie Wilkin, Senior Principal Consultant at Noah Consulting, and Stephan Zoder, Director of Value Engineering at Informatica shared some advice. E&P companies should reevaluate “throwing more bodies at a data cleanup project twice a year.” This approach does not support the pursuit of operational excellence.
In this interview, Stephanie shares details about the award-winning collaboration between Noah Consulting and Devon Energy to create a single trusted source of well data, which is standardized and mastered.
Q. Congratulations on winning the 2014 Innovation Award, Stephanie!
A. Thanks Jakki. It was really exciting working with Devon Energy. Together we put the technology and processes in place to manage and master well data in a central location and share it with downstream systems on an ongoing basis. We were proud to win the 2014 Innovation Award for Best Enterprise Data Platform.
Q. What was the business need for mastering well data?
A. As E&P companies grow so do their needs for business-critical well data. All departments need clean, consistent and connected well data to fuel their applications. We implemented a master data management (MDM) solution for well data with the goals of improving information management, business productivity, organizational efficiency, and reporting.
Q. How long did it take to implement the MDM solution for well data?
A. The Devon Energy project kicked off in May of 2012. Within five months we built the complete solution from gathering business requirements to development and testing.
Q. What were the steps in implementing the MDM solution?
A: The first and most important step was securing buy-in on a common definition for master well data or Unique Well Identifier (UWI). The key was to create a definition that would meet the needs of various business functions. Then we built the well master, which would be consistent across various systems, such as G&G, Drilling, Production, Finance, etc. We used the Professional Petroleum Data Management Association (PPDM) data model and created more than 70 unique attributes for the well, including Lahee Class, Fluid Direction, Trajectory, Role and Business Interest.
As part of the original go-live, we had three source systems of well data and two target systems connected to the MDM solution. Over the course of the next year, we added three additional source systems and four additional target systems. We did a cross-system analysis to make sure every department has the right wells and the right data about those wells. Now the company uses MDM as the single trusted source of well data, which is standardized and mastered, to do analysis and build reports.
Q. What’s been the traditional approach for managing well data?
A. Typically when a new well is created, employees spend time entering well data into their own systems. For example, one person enters well data into the G&G application. Another person enters the same well data into the Drilling application. A third person enters the same well data into the Finance application. According to statistics, it takes about 30 minutes to enter wells into a particular financial application.
So imagine if you need to add 500 new wells to your systems. This is common after a merger or acquisition. That translates to roughly 250 hours or 6.25 weeks of employee time saved on the well create process! By automating across systems, you not only save time, you eliminate redundant data entry and possible errors in the process.
Q. That sounds like a painfully slow and error-prone process.
A. It is! But that’s only half the problem. Without a single trusted source of well data, how do you get a complete picture of your wells? When you compare the well data in the G&G system to the well data in the Drilling or Finance systems, it’s typically inconsistent and difficult to reconcile. This leads to the question, “Which one of these systems has the best version of the truth?” Employees spend too much time manually reconciling well data for reporting and decision-making.
Q. So there is a lot to be gained by better managing well data.
A. That’s right. The CFO typically loves the ROI on a master well data project. It’s a huge opportunity to save time and money, boost productivity and get more accurate reporting.
Q: What were some of the business requirements for the MDM solution?
A: We couldn’t build a solution that was narrowly focused on meeting the company’s needs today. We had to keep the future in mind. Our goal was to build a framework that was scalable and supportable as the company’s business environment changed. This allows the company to add additional data domains or attributes to the well data model at any time.
Q: Why did you choose Informatica MDM?
A: The decision to use Informatica MDM for the MDM Trust Framework came down to the following capabilities:
- Match and Merge: With Informatica, we get a lot of flexibility. Some systems carry the API or well government ID, but some don’t. We can match and merge records differently based on the system.
- X-References: We keep a cross-reference between all the systems. We can go back to the master well data and find out where that data came from and when. We can see where changes have occurred because Informatica MDM tracks the history and lineage.
- Scalability: This was a key requirement. While we went live after only 5 months, we’ve been continually building out the well master based on the requiremets of the target systems.
- Flexibility: Down the road, if we want to add an additional facet or classification to the well master, the framework allows for that.
- Simple Integration: Instead of building point-to-point integrations, we use the hub model.
In addition to Informatica MDM, our Noah Consulting MDM Trust Framework includes Informatica PowerCenter for data integration, Informatica Data Quality for data cleansing and Informatica Data Virtualization.
Q: Can you give some examples of the business value gained by mastering well data?
A: One person said to me, “I’m so overwhelmed! We’ve never had one place to look at this well data before.” With MDM centrally managing master well data and fueling key business applications, many upstream processes can be optimized to achieve their full potential value.
People spend less time entering well data on the front end and reconciling well data on the back end. Well data is entered once and it’s automatically shared across all systems that need it. People can trust that it’s consistent across systems. Also, because the data across systems is now tied together, it provides business value they were unable to realize before, such as predictive analytics.
Q. What’s next?
A. There’s a lot of insight that can be gained by understanding the relationships between the well, and the people, equipment and facilities associated with it. Next, we’re planning to add the operational hierarchy. For example, we’ll be able to identify which production engineer, reservoir engineer and foreman are working on a particular well.
We’ve also started gathering business requirements for equipment and facilities to be tied to each well. There’s a lot more business value on the horizon as the company streamlines their well information lifecycle and the valuable relationships around the well.
If you missed the webinar, you can watch the replay now: Attention E&P Executives: Streamlining the Well Information Lifecycle.
Step 1: Determine if you have a customer data problem
A statement I often hear from marketing and sales leaders unfamiliar with the concept of mastering customer data is, “My CRM application is our single source of trusted customer data.” They use CRM to onboard new customers, collecting addresses, phone numbers and email addresses. They append a DUNS number. So it’s no surprise they may expect they can master their customer data in CRM. (To learn more about the basics of managing trusted customer data, read this: How much does bad data cost your business?)
It may seem logical to expect your CRM investment to be your customer master – especially since so many CRM vendors promise a “360 degree view of your customer.” But you should only consider your CRM system as the source of truth for trusted customer data if:
· You have only a single instance of Salesforce.com, Siebel CRM, or other CRM
· You have only one sales organization (vs. distributed across regions and LOBs)
· Your CRM manages all customer-focused processes and interactions (marketing, service, support, order management, self-service, etc)
· The master customer data in your CRM is clean, complete, fresh, and free of duplicates
Unfortunately most mid-to-large companies cannot claim such simple operations. For most large enterprises, CRM never delivered on that promise of a trusted 360-degree customer view. That’s what prompted Gartner analysts Bill O’Kane and Kimbery Collins to write this report, MDM is Critical to CRM Optimization, in February 2014.
“The reality is that the vast majority of the Fortune 2000 companies we talk to are complex,” says Christopher Dwight, who leads a team of master data management (MDM) and product information management (PIM) sales specialists for Informatica. Christopher and team spend each day working with retailers, distributors and CPG companies to help them get more value from their customer, product and supplier data. “Business-critical customer data doesn’t live in one place. There’s no clear and simple source. Functional organizations, processes, and systems landscapes are much more complicated. Typically they have multiple selling organizations across business units or regions.”
As an example, listed below are typical functional organizations, and common customer master data-dependent applications they rely upon, to support the lead-to-cash process within a typical enterprise:
· Marketing: marketing automation, campaign management and customer analytics systems.
· Ecommerce: e-commerce storefront and commerce applications.
· Sales: sales force automation, quote management,
· Fulfillment: ERP, shipping and logistics systems.
· Finance: order management and billing systems.
· Customer Service: CRM, IVR and case management systems.
The fragmentation of critical customer data across multiple organizations and applications is further exacerbated by the explosive adoption of Cloud applications such as Salesforce.com and Marketo. Merger and acquisition (M&A) activity is common among many larger organizations where additional legacy customer applications must be onboarded and reconciled. Suddenly your customer data challenge grows exponentially.
Step 2: Measure how customer data fragmentation impacts your business
Ask yourself: if your customer data is inaccurate, inconstant and disconnected can you:
· See the full picture of a customer’s relationship with the business across business units, product lines, channels and regions?
· Better understand and segment customers for personalized offers, improving lead conversion rates and boosting cross-sell and up-sell success?
· Deliver an exceptional, differentiated customer experience?
· Leverage rich sources of 3rd party data as well as big data such as social, mobile, sensors, etc.., to enrich customer insights?
“One company I recently spoke with was having a hard time creating a single consolidated invoice for each customer that included all the services purchased across business units,” says Dwight. “When they investigated, they were shocked to find that 80% of their consolidated invoices contained errors! The root cause was innaccurate, inconsistent and inconsistent customer data. This was a serious business problem costing the company a lot of money.”
Let’s do a quick test right now. Are any of these companies your customers: GE, Coke, Exxon, AT&T or HP? Do you know the legal company names for any of these organizations? Most people don’t. I’m willing to bet there are at least a handful of variations of these company names such as Coke, Coca-Cola, The Coca Cola Company, etc in your CRM application. Chances are there are dozens of variations in the numerous applications where business-critical customer data lives and these customer profiles are tied to transactions. That’s hard to clean up. You can’t just merge records because you need to maintain the transaction history and audit history. So you can’t clean up the customer data in this system and merge the duplicates.
The same holds true for B2C customers. In fact, I’m a nightmare for a large marketing organization. I get multiple offers and statements addressed to different versions of my name: Jakki Geiger, Jacqueline Geiger, Jackie Geiger and J. Geiger. But my personal favorite is when I get an offer from a company I do business with addressed to “Resident”. Why don’t they know I live here? They certainly know where to find me when they bill me!
Step 3: Transform how you view, manage and share customer data
Why do so many businesses that try to master customer data in CRM fail? Let’s be frank. CRM systems such as Salesforce.com and Siebel CRM were purpose built to support a specific set of business processes, and for the most part they do a great job. But they were never built with a focus on mastering customer data for the business beyond the scope of their own processes.
But perhaps you disagree with everything discussed so far. Or you’re a risk-taker and want to take on the challenge of bringing all master customer data that exists across the business into your CRM app. Be warned, you’ll likely encounter four major problems:
1) Your master customer data in each system has a different data model with different standards and requirements for capture and maintenance. Good luck reconciling them!
2) To be successful, your customer data must be clean and consistent across all your systems, which is rarely the case.
3) Even if you use DUNS numbers, some systems use the global DUNS number; others use a regional DUNS number. Some manage customer data at the legal entity level, others at the site level. How do you connect those?
4) If there are duplicate customer profiles in CRM tied to transactions, you can’t just merge the profiles because you need to maintain the transactional integrity and audit history. In this case, you’re dead on arrival.
There is a better way! Customer-centric, data-driven companies recognize these obstacles and they don’t rely on CRM as the single source of trusted customer data. Instead, they are transforming how they view, manage and share master customer data across the critical applications their businesses rely upon. They embrace master data management (MDM) best practices and technologies to reconcile, merge, share and govern business-critical customer data.
More and more B2B and B2C companies are investing in MDM capabilities to manage customer households and multiple views of customer account hierarchies (e.g. a legal view can be shared with finance, a sales territory view can be shared with sales, or an industry view can be shared with a business unit).
According to Gartner analysts Bill O’Kane and Kimberly Collins, “Through 2017, CRM leaders who avoid MDM will derive erroneous results that annoy customers, resulting in a 25% reduction in potential revenue gains,” according to this Gartner report, MDM is Critical to CRM Optimization, February 2014.
Are you ready to reassess your assumptions about mastering customer data in CRM?
Get the Gartner report now: MDM is Critical to CRM Optimization.
Did you know the 2014 Brasil World Cup is actually the World Cup of Data? In addition to the visible matches played on the pitch, eShops will be in a simultaneous struggle to win real-time online merchandise customers.
Let me explain. Jogi Löw, the manager of the German team, is known for his stylish attire. At every major event, each European Cup and World Cup, he wears newly designed shirts and suits. As a result, when television audiences see each new article of clothing, there is a corresponding increase in related online retail activity. When Löw began this tradition, people didn’t know that his outfits were made by Strenesse. As a result, people searched using the keywords “Jogi Löw Shirt.” This drove traffic to the eShop with the best search engine optimization, giving them more conversions and more revenue.
If a manager’s attire drives online retail sales, imagine how much demand there is for the jerseys worn by the most visible World Cup athletes? Many of the these players have huge social media followings. Consider the size of the social media followings of Ronaldo, Kakà, Neymar, Ronaldinho and Wayne Rooney:
There is huge demand for these player’s jerseys. This demand will only increase as the games progress. Once the winner is decided, Google searches will rise for phrases like “World Cup Winner Jersey 2014 of xxx”. Some refer to this as the super long tail. And research does show that search queries with 3 or more words have better conversion rates than queries with only 1 or 2 words.
Who can predict the winners?
What happens if a fairly unknown player scores the last goal in over time? How will that event impact social media activity and search engine volumes? Who will be able to leverage this activity to sell the relevant merchandising products fast enough? The eShop with the best data will have the quickest response. And the eShop with the quickest response will get the traffic and the revenue.
The world cup is a battle. The early bird closes the sale. It’s time to play the World Cup of Data.
In my marketing classes, I like to share on the works of Michael Porter’s Competitive Strategy. This includes discussing his three generic business strategies. We discuss, for example, the difference between an “efficiency strategy” (aka Walmart) and an “effectiveness strategy” (aka Target or even better, a high end service oriented retailer). I always make sure that students include in their thinking on differentiation the impact of customer service.
One of these high end service oriented retailers is using technology to increase its customer intimacy as well as holistic customer knowledge. Driving this for them involves understanding when customers use their full price and off price customer purchase channels. I was so fascinated about their question that I decided to ask the font of all wisdom, my wife. She said that her choice of channel is based on my current salary or her projected length of use of an item. So if she is buying a jacket that she wants to use for years, she will go to the full price channel but for a dress or pair of shoes for one time use like a Wedding, she will go to the lower priced channel. Clearly, there is more than one answer to these questions. This retailer wants to understand the answers by customer segments.
To create an understanding of each customer segment, this retailer wants to create a “high fidelity” view of data coming from customers, markets, and transactional interactions. This means that that they need two new business capabilities. First is a single integrated view of their customers across channels and the ability to see the cause and effect of customer channel selection decisions. Do customers spend more time at the full price channel option when, for example, sale offerings are going on?
To solve these problems, the retailer has implemented two technology approaches, master data management to bring together its disparate views of customer and big data for quick hypothesis testing of customer data from structured and unstructured sources. With Master Data, they get a single view of customer across differing IT systems. For separate customer specific analysis they have created operational and analytic views on top of the MDM system. And while they have an enterprise data warehouse and multiple analytical data marts, they have also created a HADOOP cluster to test hypothesis about the cross channel customer segments. They are using the single view of customer regardless of channels and transaction history to understand when customers use which channel and as well what marketing or other campaigns pulled the customer in. With this, they are creating inferred attributes for customer market segments.
Clearly, the smarter the retailer gets, the greater the differentiation the retailer services can be to customers. At the same time, the data let’s the retailer optimize marketing between channels. This is using data to create service differentiation.
Before I joined Informatica I worked for a health plan in Boston. I managed several programs including CMS Five Start Quality Rating System and Risk Adjustment Redesign. We recognized the need for a robust diagnostic profile of our members in support of risk adjustment. However, because the information resides in multiple sources, gathering and connecting the data presented many challenges. I see the opportunity for health plans to transform risk adjustment.
As risk adjustment becomes an integral component in healthcare, I encourage health plans to create a core competency around the development of diagnostic profiles. This should be the case for health plans and ACO’s. This profile is the source of reimbursement for an individual. This profile is also the basis for clinical care management. Augmented with social and demographic data, the profile can create a roadmap for successfully engaging each member.
Why is risk adjustment important?
Risk Adjustment is increasingly entrenched in the healthcare ecosystem. Originating in Medicare Advantage, it is now applicable to other areas. Risk adjustment is mission critical to protect financial viability and identify a clinical baseline for members.
What are a few examples of the increasing importance of risk adjustment?
1) Centers for Medicare and Medicaid (CMS) continues to increase the focus on Risk Adjustment. They are evaluating the value provided to the Federal government and beneficiaries. CMS has questioned the efficacy of home assessments and challenged health plans to provide a value statement beyond the harvesting of diagnoses codes which result solely in revenue enhancement. Illustrating additional value has been a challenge. Integrating data across the health plan will help address this challenge and derive value.
2) Marketplace members will also require risk adjustment calculations. After the first three years, the three “R’s” will dwindle down to one ‘R”. When Reinsurance and Risk Corridors end, we will be left with Risk Adjustment. To succeed with this new population, health plans need a clear strategy to obtain, analyze and process data. CMS processing delays make risk adjustment even more difficult. A Health Plan’s ability to manage this information will be critical to success.
3) Dual Eligibles, Medicaid members and ACO’s also rely on risk management for profitability and improved quality.
With an enhanced diagnostic profile — one that is accurate, complete and shared — I believe it is possible to enhance care, deliver appropriate reimbursements and provide coordinated care.
How can payers better enable risk adjustment?
- Facilitate timely analysis of accurate data from a variety of sources, in any format.
- Integrate and reconcile data from initial receipt through adjudication and submission.
- Deliver clean and normalized data to business users.
- Provide an aggregated view of master data about members, providers and the relationships between them to reveal insights and enable a differentiated level of service.
- Apply natural language processing to capture insights otherwise trapped in text based notes.
With clean, safe and connected data, health plans can profile members and identify undocumented diagnoses. With this data, health plans will also be able to create reports identifying providers who would benefit from additional training and support (about coding accuracy and completeness).
What will clean, safe and connected data allow?
- Allow risk adjustment to become a core competency and source of differentiation. Revenue impacts are expanding to lines of business representing larger and increasingly complex populations.
- Educate, motivate and engage providers with accurate reporting. Obtaining and acting on diagnostic data is best done when the member/patient is meeting with the caregiver. Clear and trusted feedback to physicians will contribute to a strong partnership.
- Improve patient care, reduce medical cost, increase quality ratings and engage members.