+1.855.360.3282 Contact Us

Efficient & Compliant Business with Trusted Reference Data Management (RDM)

Reference data, a subset of master data (lower in volume, variety and volatility), is generally uniform, enterprise-wide and often created by external standardization bodies. Let’s say your customer’s address is a part of a master data record, then the Zip Code and State fields are reference data. It is the kind of data you will find in dropdowns and lookups–restricted values that you can choose from within a field on a form.

The value of reliable reference data cannot be undermined. Due to the nature of IT application development and the reliance upon off-the-shelf application systems, reference data is all too often isolated in silos within many different systems. Inconsistent reference data across multiple systems can cause invalid transactions (state and zip code mismatches), revenue leakages (bad discount codes) and compliance risks (improper tax codes).

As a part of transactional records, reference data is grouped with associated master data and transactional data, and is needed for both operational and analytical master data management enterprise use cases to provide attributes, hierarchies and key performance indicators. Traditional mapping requires human judgment as well as manual synchronization and remediation of reference data. This is neither efficient nor reliable.

To ensure accurate reporting and analytics, proper governance and operational efficiencies, enterprises require a standardization system that makes it easier to define, map, manage and remediate reference data across the organization.

Reference data management is an integral part of Modern Data Management and needs to be a part of your data management strategy. Thinking about reference data in isolation or as an afterthought leads to expensive rework and compliance risks. Since Modern Data Management includes graph technology to establish relations across people, products and places, interesting capabilities result when combined with reference data. With the graph, reference data become pivoting attributes. For example, let’s say physician’s specialty is the reference data, as it needs to map to multiple systems and different physicians. Now speciality_code can be a pivoting attribute, which enables you to drill into a specialty to see the physicians across the organization, and other information relevant to the specialty. The graph makes such relationship management simple.


Today’s data management solutions need a user-friendly solution to define and manage reference data across multiple functional areas, industries and data domains. Whether customer, product or supplier data, Reltio Cloud RDM is a simple, business user-driven application that is adaptable to business needs across any use case required to preserve values and mappings between reference data sets–both in a domain and across domains.

Unlike other legacy MDM tools, that charge separately for basic RDM capabilities, Reltio Cloud Modern Data Management Platform as a Service includes core RDM functionality built-in. Being built-in makes it much simpler to ensure that there is consistent reference data for all downstream operational applications. By managing complex mappings among customer, partner, product and supplier data domains, and managing their interrelationships, enterprises will improve data quality and reduce compliance risk.



Governance of reference data is vital–manual or custom RDM often lacks change management, audit controls and granular security and permissions. Due to the complexity in managing and governing reference data, an RDM solution should include a seamless, intuitive user interface to manage lookups, and ensure data consistency across systems with version control, security and access controls. Reltio Cloud’s RDM facilitates remediation and improvement of reference data quality along with mapping to localized data, which helps with global harmonization. Built-in workflow capabilities, such reviews, approvals, history and audit trails help make structural changes to reference data with complete governance.


RDM data are often managed by business users who want to maintain, manage, standardize and remediate reference data at their fingertips. They need complete visibility into the “Crosswalks” for understanding data change impact. Collaborative curation of information through fine-grained workflow and governance allows cross-functional teams get the most accurate information real-time. Teams should be able to flexibly deliver information to downstream applications or provide access through embedded widgets within operational applications.


In many cases, organizations purchase or subscribe to third-party data sources for verified reference data. Lines of business want an easy way to connect to third-party reference data sources to enrich the existing data. Data as a Service within a modern data management platform lets you connect to such data sources, and merge the data with other master data for your data-driven applications.


A Modern Data Management platform lets you connect to existing MDM, operational applications and third-party data sources for real-time integration. User-friendly interfaces with import and export capability help map reference data sources quickly, and eliminate the burden of managing reference data sets. Reference data from multiple source systems require no transcoding, translation, custom code or IT involvement. Configuration, Lookup and Transcode REST APIs are available to manage reference data through integrations. A multi-tenant cloud platform ensures ease of provisioning and zero downtime upgrades. Deployed in the cloud, you will be delivering value faster than ever possible without the overhead of managing the infrastructure for this highly critical and available data.

Mona Rakibe is a Director of Platform Product Management at Reltio. She’s an expert in data management technologies with a specialty in content management and BPM, having worked for companies such as EMC, Oracle and BEA Systems.

Challenges in Leveraging Big Data in Retail

Ankur Gupta, Sr. Product Marketing Manager, Reltio

1. Consolidate and cleanse data from various sources:

Retailers want to bring data together from multiple internal, third party subscriptions, public, and social sources to create a complete and accurate view of their customers. They want to uncover relationships, not just between consumers and products, but locations and family members as well to solve the householding issues. They want a single source of truth of customer data across functional areas and a reliable data foundation for accurate customer segmentation and identification of the influencers.

2. Gain relevant insights from omnichannel data:

There were several discussions around retailers wanting to blend interaction data from various channels with consumer profile information, giving marketing, e-commerce, and customer support teams visibility into customer preferences, product interests, and channel choice. Retailers want to deliver insights like churn propensity, lifetime-value, and abandonment rates to relevant teams in the context of their role and objectives. Furthermore, many leading retailers are leveraging machine learning and predictive analytics to suggest next-best-actions to send relevant and consistent information, across all channels, to the customer and find opportunities for up-sell and cross-sell. However, there is still a concern about the reliability and completeness of the data utilized to run such analytics.

3. Create a global product master:

Several retailers want to create a complete product or SKU profile to understand the supply chain relations, contract adherence, consumption per location, overall global business value and even social sentiments about their brands. They want a worldwide real-time view of the product, especially during a launch, to gain critical insights into accurate targeting and managing key influencers in the marketplace, designing relevant promotions and devising social media strategy.

4. Break data silos across departments:

Retailers are looking for ways to encourage collaboration across teams, in real time. With global multi-functional teams, multi-product portfolio, and big data scale consumer information, it is critical to allow as well as secure access to a convergence of information, with the proper level of role-based access and visibility. Data management has to be a shared responsibility across all functional groups and tools for social curation of internal data in the form of annotating, workflows, tagging, and voting allow every member to contribute and continuously improve data quality and the enterprise knowledge.

5. Exchange data with external parties:

There were some interesting discussions about the possibility to share the data externally with the suppliers to establish a single holistic view of the supply chain. Historically, most retailers do not have the infrastructure to process and make transaction-level data accessible easily. Fortunately, this technology is now available as Data as a Service (DaaS). Retailers can efficiently carve out a data view in the cloud and share it with partners or even monetize their data to create new revenue streams. The advantages of retail data sharing include improving on-shelf availability, better demand forecast accuracy, and improving the customer experience, among many others.

6. Be compliant:

With so many teams working with consumer data, retailers need comprehensive auditing and tracking features to guarantee compliance. They want a historical trail for any data merged or updated and want to get alerted to abnormal data viewing patterns by application users for possible information breach or theft. Compliance and transparency need to be inbuilt into the data management rather than treated as reports developed as an afterthought.

According to a McKinsey study, the continued adoption and development of big data levers have the potential to increase US retail productivity by more than 0.5 percent a year through 2020. Such a boost in profitability is especially significant in a sector where margins are notoriously tight.

Are you ready to address the above pain points and turn your big data into a valuable asset? Answer these seven questions to learn how prepared you are to manage your retail data effectively.

Final Results are in! The Winners of Data-driven Madness 2016 are …

========== April 18, 2016 ==========

And the final results are in! The winning business initiative or technology of the 2016 Data-driven Madness tournament is …

Reliable High Quality Data

winning easily over Master Data Management in the technology region. Final results and full bracket below. Thank you everyone for participating. In summary, this tournament indicated that Reliable High Quality Data is a business imperative, that is crucial as the foundation for any predictive analytics or machine learning for relevant insights. It also showed that Master Data Management, a stalwart technology and discipline for over 10 years cannot stand alone as the only offering to meet business demands. In fact, for many Reliable High Quality Data now goes beyond just MDM. 

We will be publishing a full interactive infographic of the tournament shortly. Contact us to indicate your interest in receiving a copy.

The winner of the apple watch is Ryan McCormick with a narrow 3 point advantage over ericbless who receives a $50 amazon gift card for finishing second. As does Charles Joseph who won the random drawing among all participants. Thank you all for taking part in Reltio Data-driven Madness 2016!


========== April 11, 2016 ==========

The championship game is here! 

Anyone can still enter because Championship correct selections score 10 points each! Also as a bonus, a random drawing will occur on everyone who votes for the championship winner for another mystery prize.

Here is the near final look at the bracket the voting percentages and participant leader boards. Scroll down for the highlights of the round and instructions on how to play. Click here to enter your selections for the Championship game.

And the updated leaderboard. It’s a 3 horse race, with Ryan McCormick taking the lead.  ericbless in second and is only 3 points away. Big Dan scored 0 in the final four and drops to 3rd.


Highlights from the Final Four round (Feel free to provide your thoughts in comments below)

Business Region 1 vs Region 2

  • #1 Reliable High Quality Data won easily over #1 Relevant Insights Delivered. Indicating that accurate data was a prerequisite for insights.

Technical Region 1 vs Region 2

  • #1 Master Data Management narrowly squeezed by #1 Predictive Insights showing that a changing of the guard is nearly here. Many expect MDM to be standard moving forward, so the votes reflected that assumption. However MDM was still too critical and not widely adopted to be beaten this year

Preview of Championship game

It’s all about data quality and reliability in the final match up. The question remaining to be answered is that if business expects #1 Reliable High Quality Data, is #1 Master Data Management the only and the best way to deliver in this modern data management era.

Results will be published and the champion crowned at the end of the week. Voting closes Friday 12am PT.

========== April 6, 2016 ==========

Even though March Madness is in the books,  Reltio Data-driven Madness 2016 continues and for the first time in history, all #1 seeds have made it through to the final four!

Anyone can still enter because Final Four correct selections score 5 points each.

Here is the latest look at the bracket the voting percentages and participant leader boards. Scroll down for the highlights of the round and instructions on how to play. Click here to enter your selections for the Final Four.

And the updated leaderboard with “Big Dan” still holding serve with a slim 2 point advantage over Ryan McCormick.  Fernando drops to 4th with a DNP in the last round, ericbless gains ground and is only 5 points away. Charles Joseph is still mathematically in the running for the Apple Watch!

Some highlights from the Elite Eight round (Feel free to provide your thoughts in comments below)

Business Region 1

  • #1 Reliable High Quality Data blanked #2 Always Available in a Villanova vs. Oklahoma style beat down

Business Region 2

  • #1 Relevant Insights pulled away from #6 Recommended Actions highlighting that people want to know what’s going on with the data

Preview of Final Four Business Region 1 vs Region 2

Battle of #1 seeds in the business region begs the question. Is it enough to get #1 Relevant Insight? Or does the data have to be #1 Reliable and High Quality for the insights to be correct and matter?

Tech Region 3

  • #1 Master Data Management  trumped #6. Graph Database but not by an overwhelming margin. Many of the crowd realized that some of the best MDM platforms today already have Graph embedded  

Tech Region 4

  • #1 Predictive Analytics squeezes past #2 Data-driven Apps showing some indecision. Do you want just analytics and the #1 seed? Or #2 Data-driven Apps that are both analytical and operational in nature? This was one of the best games of the tournament with analytics winning out.

Preview of Final Four Tech Region 1 vs Region 2

In almost a carbon copy of the Business Region semi-finals, the Battle of Tech Region #1 seeds begs the question. Is it enough to use #1 Predictive Analytics? Or does the data have to be based on a #1 MDM foundation for reliable data quality for the insights to be correct and matter?

========== April 1, 2016 ==========

It’s down to the Elite Eight in Reltio Data-driven Madness 2016. Surprisingly all #1 seeds have made it through so far. Now it’s crunch time.

Anyone can still enter because Round 4 correct selections score 4 points each.

Here is the latest look at the bracket the voting percentages and participant leader boards. Scroll down for the highlights of the round and instructions on how to play. Click here to enter your selections for the Elite Eight.

And the updated leaderboard with “Big Dan” still holding serve with a slim 1 point advantage over Fernando. Ryan McCormick has dropped down to 5th (UPDATED due to a scoring/identification error Ryan’s Sweet Sixteen score has been updated and he is still in the running), ericbless and Charles Joseph are all within striking distance of the Apple Watch!

Some highlights from the Sweet Sixteen round (Feel free to provide your thoughts in comments below)

Business Region 1

  • #1 Reliable High Quality Data cruised into the Elite Eight with a dominant performance over cinderella #13 Team Collaboration
  • #2 Always Available survived a scare from #3 Closed-loop ROI

The chance for a final four berth pits #1 vs #2 in a fascinating conundrum. Is access to data anytime, anywhere more important? Or does that data have to be reliable in order for it to matter? 

Business Region 2

  • #1 Relevant Insights crushed #5 Easy-to-use with their starters resting in the 2nd half for the potential face off with #6 Recommended Actions
  • #6 Recommended Actions trumped #2 Real-time delivery in an upset, proving that getting support and guided help wins over speed on insight in this case

#1 vs #6  match-up here is almost too close to call. Do you need relevant insight first, before you get recommended actions? Or will you trust recommended actions to be based on relevant data and insight?

Tech Region 3

  • #1 Master Data Management a crowd voting favorite shimmied into the Elite Eight, over underdog #13 Semantic Analysis.  Chants of “MDM, MDM” could be heard in the crowd. Coach of #13 Semantic Analysis was quoted as saying “We’re a young technology, we’ll be back next year”
  • #6. Graph Database outlasted #7. Machine Learning in OT, with 7 lead changes in the final minutes.

In another #1 vs #6 game, the analysts feel that #1 MDM has the edge because all modern data management platforms have built-in Graph capabilities. Meaning that #6 Graph Database can’t possibly win. However those who are on legacy MDM systems, might feel they have to go for #6, but they may encounter scalability issues with typical Graph DB technology when they hit higher volumes.

Tech Region 4

  • #1 Predictive Analytics shut out #4 Elastic Scalability in an absolute drubbing
  • #2 Data-driven Apps won easily over #3 Hybrid Cloud, showing that IT values delivering business facing capabilities over infrastructure preferences

This final tech region 4 game is a doozie. Again who you favor depends on your perspective. Do you want just analytics and the #1 seed? Or #2 Data-driven Apps that are both analytical and operational in nature? This is one of the most anticipated games of the tournament.

========== March 24th, 2016 ==========

It’s Sweet Sixteen time in Reltio Data-driven Madness 2016. Not surprisingly all #1 seeds have made it through so far. Now it becomes a true test.

Anyone can still enter because Round 3 correct selections score 3 points each.

Here is the latest look at the bracket the voting percentages and participant leader boards. Scroll down for the highlights of the round and instructions on how to play. 

And the updated leaderboard with “Big Dan” holding a slim 1 point advantage over Fernando. Ryan McCormick, ericbless and Charles Joseph are all within striking distance of the Apple Watch!

Some highlights from the second round (Feel free to provide your thoughts in comments below)

Business Region 1

  • #2 Always Available had to survive a massive scare to beat out upstart #7 Free Text Search. It was looking like a lost cause but pulled off a Texas A&M style comeback to win in double OT
  • #13 Team Collaboration’s Cinderella run continues! Easily blowing away #12 Role-based Access, who in the last round pulled off their own upset.

Business Region 2

  • #1 Relevant Insight Delivered won easily over #8 Self-service Reporting. The crowd was heard chanting “You got Served!” at the #8 seed, proving that the fans want things automatically delivered on a platter
  • #6 Recommended Actions Provided was taken to overtime by favorite #3 Third-party Data Sources setting up a clash with #2 Real-time Delivery in the Sweet Sixteen

Tech Region 3

  • #13 Semantic Analysis a relatively new first timer to the tournament beat crowd favorite #12 Columnar Database in a sort of semi-structured vs. structured showdown
  • #6 Graph database continued to thump the competition, pushing aside #14 Schema-less
  • #7 Machine Learning ousted #2 Workflow with a barrage of well judged shots, that not surprisingly got better as the game wore on, and they received more data 

Tech Region 4

  • #1 Predictive analytics continued it’s date with destiny, even predicting in the post game conference that it would cruise into the final four
  • #3 Hybrid cloud showed that it was twice as nice by beating #6 Public cloud
  • #2 Data-driven apps once again destroyed its opponent and it wasn’t even close

========== March 20th, 2016 ==========

Round 1 of Data-driven Madness is complete! While there were some tough matchups, there were relatively few upsets. Most of the seeded teams moved on to the second round, where competition will now prove to be tougher.

You can still enter because Round 2 correct selections score 2 points each.

Here is the latest look at the bracket the voting percentages and participant leader boards. Scroll down for instructions on how to play. 

And here is the leaderboard …

Some highlights from the first round (Feel free to provide your thoughts in comments below)

Business Region 1

  • #13 Team Collaboration pulled a mild upset over #4 Structured Processes proving that democratization and input rules
  • #7 Desktop Access beat #10 Mobile access in a last minute thriller

Business Region 2

  • #3 Third party sources won easily over #Social Media/Public sources indicating that everyone still as a little distrust over uncurated data

Tech Region 3

  • #12 Columnar Database beat #5 Relational Database handily in an upset, signalling a changing of the guard
  • #6 Graph Database thumped #11 Document Database in a rout, in a battle of newcomers
  • #14 Schema-less handily beat #3 Fixed Data Model, showing that flexibility is king

Tech Region 4

  • #1 Predictive Analytics moved on by beating #16 Business Intelligence, illustrating that it’s better to look forward, than back
  • #8 Microservices crushed #9 ETL, indicating that given the horse power batch isn’t really relevant anymore
  • #6 Public Cloud beat out #11 Private Cloud. Next round’s match up between #6 Public cloud vs #3 Hybrid cloud (who vanquished #14 On-premises) should be fascinating
  • #2 Data-driven Apps destroyed #15 Process-driven Apps, no surprise as they are the future of the enterprise and strong favorites in the tournament

========== Challenge kickoff and instructions March 12th, 2016 ==========

It’s that time of the year. Just like March Madness, Data-driven Madness has brackets, and an ultimate champion. Some of the head-to-head match ups are “lay-ups” (pun intended), and others are a little “I’m on a desert island and I have to choose one” hard.

To play:

1. Fill in the final selection championship round here. Select the winner of each of the head-to-heads by forecasting and selecting the winner of the second round. Your selections will also cast a vote for each selection, and the highest number of votes for each will move on to the next round.

2. You get a point each time your selection moves on. After each round is completed, you will be given the chance to select the winners of the next head-to-head rounds. Second round winners count for 2 points and so on.

3. The winner of the Apple Watch will be the person with the most number of points at the end of the tournament. Unlike March Madness, you can join at any time until the end of the tournament when all the results will be tabulated.

4. Tie breaker on points will be determined by the person who selected the Champion, and then further tie breakers will be by the number of correct entries in the final four and previous rounds.

Show Me (and let me Act on) the Data! The Days of Master Data Only Reports are Numbered

For over 10 years master data has been locked away in MDM tools with limited visibility of the profile and state of the data. Even the data stewards, tasked with the manual resolution through workflow queues have had poor feedback from traditional tools as to the state of their most important asset. MDM vendors’ general response to this challenge has been either to:

  1. Expose the hub metadata through a published model and to allow customers to “roll their own insights” via BI tools such as Microstrategy and Business Objects

  2. Leave it to the users to export the data out to a separate data mart or warehouse, or worse excel spreadsheets for review

Typically the lag and latency associated with analyzing the data quality is of concern, since an agile organization has data continuously coming into the system. Like the painting of the Golden Gate bridge, the work is never done. When you reach one end point, you just turn around and start again. 

Furthermore, narrowing efforts to focus on the data that needs the most attention has always been a challenge. There is only so much manual effort that can be applied as resources and expertise are limited.

Being able to quickly filter and see data issues, or have the system provide the alerts and recommended actions is a capability that is available in today’s modern data management Platform as a Service (PaaS). And since all of this should be running in a multi-tenant cloud, using a browser-based and mobile interface, any filtering and creation of lists of data can easily be stored as a URL and shared with colleagues for collaboration.

While traditional MDM vendors are trying to incorporate this level of functionality and basic reporting, most companies have moved on.


Companies expect not only in line insights and reporting directly from within the tool, they now want access for frontline business users. Not just so they can see the quality of data for themselves, but to allow them to make changes, offer suggestions, and comments at the point of engagement. 

These capabilities can only be found in a new generation of data-driven applications. They provide complete transparency as to the quality of the data, feature complete social collaboration and curation functionality at scale, as well as allowing third party data to augment and replenish gaps in quality in real-time via Data as a Service (DaaS). In other words, reporting is nice, but action speaks louder than reports.

Another reason why reporting on just master data is limiting, is the increased demand to have not just customer entities, but product, organization and other entities and affiliations to provide the big picture. Additionally all related transaction and interaction data are required to be more closely tied into profiles, hence the continuing requirement to create data marts and warehouses.

So merely improving the reporting capabilities of the MDM tool is necessary but not sufficient. Frontline business users want operational execution and the ability to immediately correct or provide input about the data from the same interface they are using for their daily operations, not just the ability to retroactively get reports about the data. Better still, the system should have the smarts to provide recommended actions in the context of workflow to guide both business users and data stewards as to what to do next. Or to take action on their behalf.

With all the amazing modern data management technology available, your company should expect and get more by going beyond just MDM, by providing your business teams with fully integrated data-driven applications. The next time someone offers to “Show you the data!”, ask them whether you can do something about it together as a team.