The last interview with Alessandra was a huge hit with the audience and this time, we have the views from a completely different industry. So glad to have a chance to welcome our dear friend and champion, Ramesh Kalava, Information Management Leader at Citrix. Ramesh graduated from University of Madras with a specialization in Computer Applications. He has a deep experience of 17 years on data and information management with multiple organizations.
Its a pleasure to have you today to talk about yourself and your work. Thank you for allowing us to interview you. Really appreciate your time. Can you please tell our readers a bit about yourself?
I am currently managing Information Management practice at Citrix Systems to develop & implement DI/DQ/MDM solutions and to improve the account data quality for better decision making at Citrix Systems. As a practice leader, I help develop enterprise architectural standards & processes for data in my organization. I also drive implementation of the same for all domains to reduce the complexity & ambiguity.
What is the business problem you are solving currently and what impact does it have on the organization? Can you please elaborate with a few examples on how the master data journey helped fuel these use cases?
Master data is indispensable and the key driver for any organization, because it is the foundation for multiple business user cases including improve data quality, data stewardship, reference data management, hierarchy management, business process management etc. During my 13+ years of journey on MDM, we have solved multiple business problems using various MDM process and technologies. Some key use cases are
- Account de-duplication :
- Account de-duplication is key for marketing operations ( campaigns, market penetrations, lead conversions etc). Creating single source of truth by enriching the single source of truth using third party data vendors enables effective decision making
- Customer 360 / Customer Master:
- Enabling Customer360 for the Supply chain and Sales is key for business operations. Integrating third party data vendors to standardize and enrich is imperative for analytics and dashboards to convert the leads and improve renewal rates. Hierarchical information facilitates effective territory management & executive dashboards.
- Partner Management /Hierarchy : This is another key business use case to solve Territory management for sales reps.
Editor’s note: We agree wholeheartedly. In fact, as a startup, we hit so many new use cases of MDM everytime we talk to a new industry or customer.
What according to you are the key pillars behind a successful master data management project?
Based on my experience, the following pillars are key for any data quality/MDM project
- Define Your Business Problem
- Stakeholders Sponsorship
- Enable Data Governance
- Continuous Monitoring and Prioritize
- ROI Measurement
What are your views on agile data mastering?
Master data projects required high budget to implement and most of the time stake holders don’t see the value immediately. Moving to agile processes allows stake holders to see the immediate value and ROI. This helps to improve the trust and sponsorship commitment. We have implemented multiple data quality / MDM projects using agile processes. Our recent project Customer360 is implemented using Agile fundamentals. This helps to be transparent with stakeholders by providing demos on every sprint.
Editor’s note – Great, keep it simple, use agile tools, deliver smaller sprints.
According to you, what are some key aspects about Customer Data Management that you would advise our readers to keep in mind?
- Customer Master is key for any organization. Defining business problems would help us to focus on data problems like data de-duplication, multiple hierarchies ( Sales, Legal, DUNS, Product, Region and etc.,)
- MDM projects are not IT projects. Establish great collaboration with business stake holders
- Define data champions and owners
- Enable KPIs to measure and monitor
How excited are you about the potential of AI in MDM? Where do you see the maximum impact of applying AI on the MDM discipline?
I am excited as I see that Master data with predictive analytics and enabling AI helps data stewards to take better decisions. It also improves accountability and productivity.
Please name few data practitioners whom you admire and would love to hear from in this interview series.
So nice to get in touch and talk with you Ramesh, many more master data projects for you!
Comments, questions, suggestions, feedback – contact us here. Thanks!
Posted on June 19th, 2020
Our guest today on the Master Data Maestro Series is Alessandra Almeida from SANOFI. Alessandra has over 15 years of experience in Data Management and Data Analytics working in global settings and has always focused on transformation through innovation. She is passionate about promoting a Data Culture in corporations to bring value to businesses. Alessandra is always motivated to connect data to corporate strategic objectives and enjoys most in working with team members and stakeholders at all levels on utilizing data as a true fuel to digital transformation.
Let’s hear what Alessandra has to say about data and MDM.
Thanks for spending your time with us Alessandra. Can you please tell us more about yourself ?
I graduated with a master’s degree in Data Management and Analytics and continued with formal training in leadership and coaching. I have worked in senior positions for various global consulting companies. I previously worked as SANOFI LATAM head for Information Management and Analytics. I am currently a Senior Director at SANOFI within the Enterprise Information Management organisation, focusing on a streamlined implementation of customer MDM for the whole company.
When did you start working on master data management?
As you can see by my mini curriculum, “Data” was always part of my career and recently, in 2018, I was assigned as the Global Head for Customer MDM at Sanofi, starting from ground zero, building the vision and implementing it in a multi-year program roadmap.
You have done a lot of work on digital transformation. It would be lovely to hear about what it involves and how it integrates with your MDM work.
One of the drivers for MDM strategy is to support Customer 360 View. Understanding Customer engagement on multi-channels is key to address customer needs properly. MDM at Sanofi started to provide Customer Golden Profile to MCE Program (Multi Channel Engagement). We had many learning on the road, especially talking about consent and data privacy. We have a lot still to accomplish, but MDM is gradually being recognized in the company as a “One-Stop-Shop” for customer profiles. Customer 360’s aim is to provide a complete view of how a company is working with their customers and MDM has a key role here, providing the right profile to different downstream departments and unlocking the customer data for analytics.
Editor’s note: Thats correct, we are also seeing a changed perception for MDM as the key enabler for Customer 360.
Having said that, consent must be very well managed across the systems, and having a centralized way to collect the consent and distribute it to the different systems will give a clear picture on how this customer wants to be contacted, or not contacted at all. In my opinion, Customer MDM is not the right place to store the consent, but it can provide the right profile to be connected to the consent. Having a consistent Consent strategy, along with a preference management platform, very well aligned with Customer MDM is key to establishing a Customer 360 strategy.
Another perspective of MDM is data privacy. GDPR (or any other privacy regulation) establishes that we should be able to track how data will be stored, anonymized or deleted as per the Customer’s request. I am summarizing and simplifying the concept of regulations, it is much broader than that. To deal with GDPR requests, we must have a consistent data flow, checking customer’s requests since the ingestion (if the data is coming from a 3rd party vendor, IQVIA for instance) and make it flow across different downstream systems through an orchestrated and coordinated process.
MDM will own the role to implement the request on MDM side and trigger the downstream as the following steps, but each and every downstream must be accountable for their internal process of completing customer’s privacy requests.
What are the biggest challenges you foresee in the healthcare industry around data quality and data integration, especially with patient, HCO and HCP data?
I will focus HCO and HCP, as they are my core domains at this moment. We need to establish the right way to collect this data from different sources and have a strong capability to master the data and create a golden profile that will be pushed out to downstream systems. And when we talk about HCP, it is not only the physicians, but all Healthcare professionals, like nurses, pharmacists, dentists…any professional that the company wants to connect with. Same with HCO. It is not only about hospitals or clinics, we are targeting here pharmacy chains, wholesalers, distributors, etc… This rich and robust environment will allow Healthcare organizations to unlock analytics and really achieve Customer 360.
What would you say are the key benefits of an MDM implementation?
A well implemented master data system provides the following
- Improving the consistency and quality of key data domains
- Increasing the accuracy of the answers to important data-driven questions
- Enabling seamless flow of data across the value-chain to address critical needs
You have done a lot of prior work on Sales & Marketing analytics data, quality and information governance. Please tell us more about your challenges and outcomes.
The commercial area is demanding a lot of quality data to produce consistent reports and analytics. We see today that the CRM platform should not be accountable for customer profiles. They will consume the Golden records, coming from MDM and have a bi directional connection with MDM to process any data change request coming from field sales. And other platforms should also come and feed themselves directly from MDM. In that way, analytics on top of DataLakes or DWHs will have a consistent MDM ID coming from different transactional systems and being able to produce consistent analytics results. Having said that, Data Governance is mandatory, to define clear quality rules, measurement controls and people skilled to answer data quality inquiries. For big corporations, the biggest challenge is to balance Global vs Local. It will depend on the ambition that the company has in order to manage the data ecosystem.
Editor’s note: It is an interesting point you brought out. Unless we integrate the MDM with the datalake and datawarehouse, or have datalake native MDM systems, the MDM itself becomes a silo and not that useful.
For Sales & Marketing, on the customer domain side, a full centralized mode will never work (in my opinion 🙂 ). That’s why I am suggesting those 2 options.
1. Decentralized – a data governance at local level can solve the discussion.
2. Hub & Spoke – Some components of the data strategy can be managed centrally, so a Global Data Governance must be in place as well as a good Metadata management strategy and Reference Data Management strategy.
MDM Implementations have a reputation for being extremely expensive and behind schedule. What are your thoughts on this?
MDM implementation can be painful for sure, if we want “boil the ocean”. My recommendation is to start with data that matters and define MVPs (Minimum Value Product). We worked like that and we went live in 5 countries in the first year, 4 more in the second year and evolution on the first 5 ones. By June this year we will add 3 more countries live, meaning delivering globally around 8 millions golden records. This year we moved our program for SAFe Agile Framework methodology, and we run 12 weeks iterations, with 6 sprints of 2 weeks, composed by clear deliverables. In this way, even if having a full picture can take a while, we are adding blocks of MDM to deliver business value. For example, we can start with 2 data sets that represent 80% of the profile and then in the next iteration we can add more data sets in order to enrich this profile.
The agile machine is beautiful as we have different teams working in parallel with different components to be delivered. It is a well-orchestrated machine that works much better than long projects with longer time to deliver results.
Editor’s note: Agile data mastering is being advocated by us and other successful practioners in this series as well. Great to see this as a recurring theme on this interview series.
Regarding costs, it varies. Everything is comparable to business value that a specific feature will bring to justify the investment. Budget is centralized and coordinated by my group and priorities are defined based on a limited capacity, clear prioritization criteria and business value score. A prioritization SterrCo is in place with business to have a common sense on what is better now for the company. For sure we manage a backlog that will be considered and discussed on the next PI (Program Increment).
What advice would you like to give to a customer starting their MDM journey?
Create a Vision (long term) and define your short term objectives.
Find your ambassadors on business and IT side
Select a tool that is easy to adapt and customize
Identify your key partner to deliver the project
Communicate progress and marketing/celebrate all milestones.
What in your view is going to be the impact of COVID on MDM and Healthcare companies?
COVID-19 is bringing a deep rethink about on ways of working in many areas, and it wouldn’t be different on Healthcare. I will emphasize my comments on the Commercial area for Customer Domains, but for sure, other domains are heavily impacted as well. If we look for example at Patient, Vendor, Employee.
For customers, especially sales field force, less and less face to face visits are allowed. In this situation, having a solid master management process that gives the ability to connect the dots between CRM, Medical Systems, Transparency, Events and Finance will help to empower the reps for better digital communication with customers as well as improve assertiveness for digital channels for campaigns or institutional content. Not mentioning the high importance on Product Launch, that will require a big synchronized effort between different teams to achieve the right customer target and get the best result on the launch as possible.
MDM is considered in healthcare companies as a back end solution, but gradually it is taking a protagonist role, especially if we don’t have more the luxury of a full human connection and we more and more need to rely on data to make decisions. The foundation for a good decision making strategy is a strong data management in place, encompassing MDM, Data Governance and Data Quality.
You have a great mix of working on analytics, digital transformation, MDM, Data Quality and Data governance. In your opinion, how does it all tie together?
In my point of view, the glue for all those buzz words (:-)) is “Data itself” . MDM/DG/DQ are foundations to provide a good analytics strategy. We should take a look at one domain holistically and put a strategy in place that comes in 3 layers
1st Layer: Business Information Architecture (BIA) – Here we should work on pure business outcomes and business value. What to expect from the data? What do we want to achieve? What are business questions we need to answer?
2nd Layer: Data Architecture – In this layer, we will define data flows, understand data lifecycle, data maturity model and data readiness (sometimes, very late on the road we find out we don’t have the data. We have something on excel files in someone’s machine and this is very troublesome). Data Architecture must give us the clear view of our roadmap, starting by more mature datasets and defining what to do for the others to achieve the minimum readiness point to be part of the data architecture. In that layer we identify also data governance aspects (not into many details, but main highlights, as well as data privacy and security flags for PII and/or some reserved markets that we need to align data security aspects.
3rd Layer: Technology Architecture – What is the tech stack we will use to deliver. Discuss how technology can streamline ingestion to consumption through modern platforms (cloud services) and APIs. Here is the moment to define the MDM Engine (tool).
Editor’s note: This is a wonderful way to break the problem down!
Please name a few data practitioners whom you admire and would love to hear from in this interview series.
I’ve been following Scott Taylor’s articles. I also had a chance to work in an eyeforpharma event as a speaker, with Jessica Federer in a panel she was leading. I don’t know if she is dedicated to data only, but we had good talks about governance and compliance on data ecosystem for pharma.
Pleasure learning from your experience Alessandra, you have really described the MDM need, planning and execution strategy in detail. Grateful that you shared your thoughts with us during this very busy time.
We wish you the best.
Comments, questions, suggestions, feedback – contact us here. Thanks!Posted on May 20th, 2020
Buvana is a Data Management Practitioner with 10+ years’ global experience in the insurance and financial industries. Possessing extensive data management experience with analytical, technical and interpersonal skill sets, Buvana has demonstrated a strong ability to manage Big Data initiatives. She has a proven track record of achieving strategic business objectives using data driven solutions while demonstrating an expertise in producing enterprise data strategies. Known by colleagues for being a driven team-player and action oriented visionary with fast capability of learning new techniques, Buvana has successfully implemented enterprise-wide data initiatives. We had the pleasure to hear from her about her journey and her outlook on data for the Master Data Maestros series.
Pleasure to have you as a guest today Buvana.
I want to thank you for the opportunity.
You successfully delivered digital marketing data solution for MetLife’s property and casualty business. What were the project challenges? How did you overcome them? How did it impact the bottom line?
Some key Project challenges were lack of customer 360 to consistently identify our customer and as a result, we were not able to trace their journey through the life cycle of digital marketing. We were never able to tie the campaign outcomes to the actual campaign itself to measure the success and ROI of the campaign.
MDM was the data foundation that helped us establish trusted views of the customer and link their campaign activity to the unified customer views. Our MDM program enabled our business to trace the journey of the customer and measure KPIs across the marketing funnel. These insights helped digital marketing stakeholders to tie the outcomes of the campaigns to the ROI. For example, cost of launching a campaign is say $2M but as a result of our digital campaign , our conversion rate (ie. converting clicks to quote to a sale) has yielded an increase in sales by $16M. Additionally , we have a huge opportunity to preform cross sale and upsale outreach in subsequent campaigns.
Editor’s note: That is a phenomenal achievement, an increase in sales to the tune of $16m through trusted view of customer built by the MDM. Congratulations!
You have worked on Enterprise Data Lake initiatives. It would be lovely to hear about what it involves and how it aligns with your MDM work?
Effective data lake architecture is a pre-requisite to attain maximum value of a MDM program. Pairing of MDM and data lake is instrumental in driving value for data initiatives. Historically , MDM synergizes the capabilities within transactional systems and datawarehouses to extract critical data points to enable the master data. With advent of data lake architecture , MDM has the ability to extract critical data real time, extract critical data from unstructured data sources seamlessly through a scalable data ingestion process. Data lake design offers various storage zones that helps manage application/ transactional data and master data in an optimal fashion.
MDM solution paired with a scalable ingestion process in Data lake and flexible storage methods where transactional, application, unstructured datasets , alternate datasets can coexist is a critical data foundation enabler to support all our business use cases and offers extensive analytical capabilities.
Editor’s note: Agree. MDM can not be a standalone system. It has to be natively integrated with the datalake.
What are the biggest challenges you foresee in the BFSI industry around data quality and master data management?
For MDM and data quality management to be successful, it is critical for an organisation to have a data champion at the Senior leadership level and achieve organisational alignment.
While an MDM program can be quite intimidating due to the complexity, best way to drive success is think big but start small – focus on critical business case that business is focused on, align the data strategy to business use case, establish MDM as a program to support the data strategy and then keep this program as evergreen to support to additional use cases . So best way to achieve success is baby steps and focus on tangible deliverables that has an immediate impact on the business outcomes to drive value.
Editor’s Note: Very well said, start small, think big! Agile data mastering is here to stay.
What would you say are the key benefits on an MDM implementation in the Insurance sector?
Insurance industries business strategies support offense (revenue focused) and defense (operational efficiency, compliance /regulatory) strategies.
MDM data strategy supports both offense strategies – driving sales /profit and defense strategies like GDPR , HIPAA regulations , CCPA etc.
Editor’s note: Happy you brought this up, it resonates with our experience too.
You have done a lot of prior work on sales & marketing analytics data, quality and information governance. Please tell us more about what were the challenges and outcomes.
Our data foundation is critical enabler for enterprise analytics capability . An effective data foundation must be supported with comprehensive information governance to drive certified trusted data to support analytics effort.
The key challenges were lack of data foundation and not having data driven mindset across the enterprise.
A well governed data foundation is critical to drive business insights for an organisation. Often times , we think about technology capability but we should start with business drivers – what outcomes are we targeting for , what success looks to us and how we leverage data to help accomplish our business outcomes.
Also, data literacy is another challenge – stakeholders need to adopt common vocabulary and common semantic understand of the critical data.
Data quality has always been reactive than proactive – we fix the data problems after the fact – this has resulted in data debts because the organisation spends more time fixing bad debt in a reactive manner.
Hence it is imperative that information governance framework is critical for data foundation that aligned with business drivers.
Information governance ensure data quality and promotes data literacy – that is critical to drive business insights using enterprise data assets.
In the recent Magic Quadrant for MDM solutions, Gartner has stated that “MDM continues to shift from reluctant to indispensable spend across a broader range of industries”. What are your views on this?
While I agree to this statement from a research /theoretical standpoint , there is more to this statement in the practical world. The shift reluctant to indispensable spend requires few intermediate stages. I would include reluctant to acceptance spend that will eventually drive indispensable spend . The biggest challenge is shift from reluctant to acceptance spend – accepting that the MDM spend is critical first is key and then the spend to transition to indispensable over time. The spend transition is slow in the practical world.
Analytics are a focus area for most enterprises, big and small. Can you please elaborate some examples you have seen where an MDM implementation fuelled the analytics journey?
I can highlight two critical analytics use case where MDM has been instrumental to the success of these use cases .
- To improve Net promoter score of our customers . MDM enables 360 of our customers and enables operational efficiency of our customer service . This results in better customer experience and opens up huge opportunity to promote our organisation through customer centricity.
- Customer Life Time Value – Again MDM allows marketers understand current product relationships and determine value of the customers . Through MDM , we have opportunity to analyze what is the propensity of our customers to buy other products , thereby increasing their future life time value.
You have a great portfolio of work on analytics, data lineage, data governance, MDM and data quality. In your opinion, how does it all tie together?
For an organisation to be data driven, it is imperative to have MDM program and promote data literacy – bringing together stakeholders and driving towards business outcomes and establishing organisation alignment becomes critical for a successful MDM program.
For credible trusted data domains established through MDM , it is absolutely essential to have data governance framework in place to align people , process and technology . This will ensure organisational alignment, promotes data literacy and helps in establishing enterprise data stewardship model in place for trusted information quality.
While MDM becomes data foundation , data governance drives data literacy and improves data quality by having controls in place through well governed process that manages the data , align people who drive the data quality and technology that enables the data and provides insight on data lineage so that we have full transparency on primary sources of the data.
For a successful analytics initiatives, it is imperative to have consistent , cohesive and trusted data and ensuring we have a good understanding of the data.
A well governed MDM program is a key enabler for analytics initiatives.
Pleasure learning from your experience Buvana, thanks for taking time out and we wish you the best.
Comments, questions, suggestions, feedback – contact us here. Thanks!Posted on May 7th, 2020