Editor’s note: In a November 2016 SandHill article about Master Data Management (MDM) technology, Stibo Systems predicted enterprise, multi-tenant MDM solutions in the cloud will improve in 2017. In this article, we follow up with an interview of Christophe Marcant, Stibo Systems’ senior vice president of strategy & communication, for deeper insights into what we can expect as this technology evolves.
In what ways is the MDM market changing now to better address companies’ needs and expectations?
Christophe Marcant: MDM technology helps companies provide accurate, complete and consistent information to their customers and allows them to understand their expectations. What has changed is the approach toward implementing such a solution.
Historically, emphasis was on establishing a “system of record” that could include a single view of the product, customer, etc. The idea was that by centralizing definitions, models, business rules and storage, the enterprise would be able to dispatch information to any audience across the organization or outside to business partners or consumers. However, the rate of change is accelerating and it is no longer sufficient to be in control of one’s data via a “system of control.”
Today it is becoming essential that an organization can adapt to ever-changing demands and be enabled by data rather than hindered by it, becoming “data agile.” We expect we will see more demand toward the creation of “system of engagement” where the emphasis is placed on creating next-generation communication and collaboration capabilities.
There has been a lot of buzz around machine learning and the value it can bring to organizations. How do you think it will affect the MDM space?
Christophe Marcant: While the focus used to be on centralizing – and controlling – data to better distribute information, we anticipate that the market will now demand that MDM solutions adapt/react to data demands quicker, whether internally or externally. There are some exciting developments in artificial intelligence and machine learning. There is great opportunity to leverage machine learning (ML) to adapt data from a source to a consumer faster. Rather than focusing on enforcing format and meaning to facilitate exchanges, ML will allow us to discover patterns in data, propose associations, correlations and adaptation. The more data a system processes, the more it will learn. It will then allow for any new data to be taken in consideration as needed, by relying upon prior interactions. This will render traditional extract-transform-load approaches a thing of the past.
MDM appears to be moving towards a more decentralized approach. How will this affect multi-party data pools and data exchanges?
Christophe Marcant: While the trend toward leveraging greater amounts of data faster and in unexpected ways could indicate that data pools might not be as important in the future as they have been in the past, the jury is still out. In my opinion, data pools have offered three important components to the community: (1) a shared and agreed-upon data model to describe products, (2) an exchange capability to move data from point A to point B, and (3) a contractual aspect by knowing that your information is going to be leveraged by legitimate consumers and likewise that any information you receive was from an authoritative source.
With the advent of on-demand data exchanges, powered by ML, the participants will no longer have to worry about the formats and semantics but still will be on the hook to enforce the contractual aspects of those exchanges.
Blockchain, which is the technology that has been highly publicized through its association with Bitcoin, could also play a role here. Blockchain is important because it allows for the creation of a micro-contract infrastructure between the participants. This infrastructure can enforce when and where information is used and leveraged and ensure its authenticity.
You have been involved in the MDM and product information management (PIM) space for some time now. Is the ongoing cloud demand affecting your outlook on the solutions Stibo is building?
Christophe Marcant: I mentioned earlier that motivations for MDM systems remain the same but that expectations are evolving. There is no doubt that cloud, as in multi-tenant SaaS, offers a quicker time to value by allowing participants to share and access data faster. However, this is achieved at the expense of flexibility.
Purpose-built MDM apps, offered as SaaS, are both market and business-case specific, which may be the reason that we sometimes see an increase in demand for these solutions. Having said that, on-premises or hosted/IaaS solutions remain essential for companies that care about flexibility and their ability to leverage their investment across the enterprise.
Please explain the trend toward multidomain master data management and what is driving it.
Christophe Marcant: Again, this goes back to the evolution from system of control to system of engagement. If the emphasis is put on connection and exchange possibly over storage and enforcement, then it is important to remember that data comes in many shapes and/or forms. Some is easily understood, while others we may not be aware of yet. There is a great benefit in connecting data of different types and origins.
Over the last two to three years, our customers have shared their interest in mastering multiple types of data, or what we call data domains, and relating those with one another. What they tell us is that mastering product data is key, mastering customer information is urgent and being able to connect that data is imperative as it allows them to address untapped opportunities.
In fact, they now realize there is possibly more value in connecting the data to derive beneficial information than in the original data itself. For them, multidomain master data management is the proverbial pot of gold at the end of the rainbow.
Christophe Marcant is senior vice president for strategy & communication for Stibo Systems, a global provider of multidomain Master Data Management (MDM) solutions. For 10 years as a retailer, Christophe focused on how product information can be leveraged in e-commerce and omni-channel initiatives. Prior to joining Stibo Systems, he led SapientNitro’s PIM practice, where he advised clients worldwide as they considered new product information strategies. Follow him on Twitter.