Mastering the MDM Journey: Start with Trusted Data

On today's webinar on Mastering the MDM Journey, and Starting with Trusted Data, I am Melanie, a solutions engineer at Tamr, and I'm joined by my wonderful colleague and director of product marketing, Tatiana. So welcome, Tatiana.

Hi, Melanie. Hi, everyone. Thank you for joining us today. I'm excited to be here and share some insights on what is an MDM journey, how other companies navigate their own unique MDM journeys, and, yeah, share a couple of real world examples of customers who do that successfully.

I'm excited. I some of you may have been on the other webinars before.

And as I always do, I'll just give you a general walk through. And if this is your first time at the webinar, I'm super excited. Hopefully, you'll find something useful, today both for yourself and the organization that you associate yourself with, I should say. Alright. So in the next thirty minutes, we'll walk through the four critical stages of MDM. Those will be assessing, improving, reviewing, and operationalizing, and we'll walk through what those exactly are later.

We'll explore the biggest challenges organizations face, how they approach the MDM journey, and share real world examples of companies transforming their data strategies.

At Tamar, we call MDM a journey because tackling messy silo data at scale is never a one time fix. It requires, continuous learning, automation, and adaptation to drive real business value. So it is challenging, and we hear you. We hear this all the time.

So we have this webinar to show you how you're not alone and what it is that you can really do, to work on this. But before that, we wanted to run a quick poll. So please take a moment to rate how far along your organization is with its MDM program. Some of those arch options are not started, planning, implementing, or in production.

So on the right side of your screen, that's where you will be able to see that call, and we'll give you a couple of seconds to answer that.

Alright. Let's see.

Oh, these are interesting results. I like this.

Okay. So different stages. That's gonna be very, very helpful. I like how it's kind of spread out.

I see there's five votes for not started, six votes for not started. Or did somebody think they started and they just went back on it? Okay. That's possible.

We have twelve for planning, ten for implementing, and ten in production. Alright. So I think all of you will be able to gain something today.

This will be, hopefully, an extremely useful session. But before I get started, with the webinar with Tatiana, I'll just give you a little more information, and I'll stop with the planning. But I am a planner, and I want you all to feel comfortable during the webinar.

On the right side of your screen, you'll also see a q and a box, so feel free to submit any questions at any time. These are private, so don't hesitate to ask whatever is on your mind, and we'll address them throughout the session and in the q and a at the end.

But, also, we have, a team that will be able to answer those questions. But that's it. No more giving you advice on how to go to q and a and answer polls. Let's kick things off.

So Absolutely like that.

You've worked with many organizations, on their MDM programs. What's the biggest challenge that they face?

Yeah. That's a great question. So, as you, of course, know, we work with customers of different sizes that operate in different industries.

They are at different, you know, technological maturity levels. And, there is one thing that, you know, that is in common, among them. And what it is, it is that they know that they have a problem with their data, and they know that MDM is the right solution for them, but they don't really know where to start. Right?

They don't really know how to tackle this problem. And so, you know, we spent some time learning from those who do so successfully, who do great progress, you know, who show great progress and, success on their unique MGM journey. And so we decided, you know, to chart this uncharted territory, by, kind of specifying those four key steps of an MDM journey to help others navigate their MDM journey. And so what are those four steps?

So the first step or stage is assess. This is a stage where you, spend time understanding where it is, you know, where where you are, what's the state of your data, and where do you wanna go. Basically, understanding what's the destination. Right?

What's the business goal? Why do you want, you know, to embark on this journey? So second stage, improve, is when you, improve the quality of your data by cleaning, enriching, curating it, and delivering golden records, trustworthy golden records that contain the best data, you know, the best enterprise, data that you have available.

So then moving on to stage three, review. This is a stage where you take this improved great quality best version of your data and share it with business users, not only business users, but those end, end users, data consumers who use the data on a daily basis to do their job. Right? It could be salespeople, marketing organization.

It could be procurement. You name it. But, basically, the people who actually use the data on a daily basis. And then last but not least, operationalization.

This is the stage where you connect the best data that you have available about your customers, suppliers, etcetera, to your operating systems.

And here in this stage, I think it is very important to stop for a second and mention that we've seen many customers, many organizations actually start at the very, like, last stage. They kinda either skip or rush through the first three stages, and they jump right into operationalizing their data. But the trick is that when you do so, when you operationalize the data that isn't ready, it causes a lot of, problems down the road. And we will talk about it in more detail later, in our in in today's webinar. But it is, I just wanted to kind of highlight it here right at the start that is very important, right, not to jump straight into operationalization.

And we are, here to help you navigate through the first three fundamental steps to get you ready for operationalizing your data.

Mhmm.

Well, I have a lot of questions for all of those stages, and I think that the people that are on the webinar, might too. But I think in any attempt that I personally have on targeting a new goal and organizations that I work with in tackling MDM, they start with the assessment, hopefully, at some point.

So how should organizations better assess where they are in their MDM journey? Because I have to admit, I don't think that everyone really knows where they are in their MDM journey. There's the people that really hope they're somewhere but may not be there.

So how do how do they better assess it?

Yeah. Of course. And, yes. I totally agree with you and understand that there is no, you know, one fits all approach.

But there are some, you know, rule of thumbs, you know, some tricks and, that, I'm here to share today. So I guess the first and one of the most important things you do on your RMDM journey is try to understand what it is that you try to achieve, not on the data standpoint, but from the business standpoint. Right? Connect your data your data with the business strategy.

Right? And once you do that, it will help you prioritize or identify the data that, you wanna focus on. Right? It is a very bad idea to kind of boil the ocean, right, and tackle your data in its entirety.

And so focusing on only on the data at first, that would be essential to making sure you are, you know, achieving the, you know, the destination of your choosing, would be, you know, the the first thing you would want to do. So to do so, you wanna make sure that you engage diverse stakeholders as I mentioned earlier. Right? You wanna make sure that this, you know, exercise is not done just for the sake of having great, you know, good data.

It is for the sake of making, you know, people's life better and helping them do their job better. Right? And so connect to diverse stakeholders across organization to make sure, you know, your data strategy aligns with the, business strategy. And to do so, ask those uncomfortable, you know, critical questions, to get to, you know, to to the point where you need to be.

Mhmm. It's interesting you bring this up now because this morning, I talked with Burberry, who is one of, Tamr's customers. And they talked with, somebody who leads a lot of different data teams and different data functions. And they always say, involve different stakeholders and ensure that you focus on what the most important asset you have to your business is because there's so much data, that you could tackle, but it just doesn't always make sense. Some things are more distracting than they are, useful.

And with with that in mind And, actually, yeah, I just I just realized that I, wanted to add one more thing as you were talking about, you know, this conversation that you had earlier.

What we've seen is that, many organizations kind of tried to, you know, rush through this process or skip the stage altogether and just make assumption. Let's say, oh, my data is, whatever, sixty percent, you know, ready. And, as a result, you know, surprise, surprise, it is not. So it is very important not to rush through this process and actually, you know, take a short step back and, you know, make sure your starting point is right.

Because, you know, without understanding the current state, you cannot really plan and prepare for successful future state. So, basically, future starts tomorrow or today. Right? So, yeah, just wanted to add it here.

Yeah. And for the people that are on the call, are there any questions, that they could also internally ask that you can leave them with, when they go and assist with their teams?

Yeah. That's a tricky one. And, I guess, again, you know, starting from very simple questions that turn out to be very hard to answer. You know, how many customers do I have?

If there is a issue, you know, problem in my data, you know, typo something, how quickly can I identify it, and, you know, how easily can I resolve it? You know? I bet there are many people on this call or in this webinar, right, who's been in a conference room, you know, on a meeting, and then a question is asked, how many customers do we have? Or name top three customers, and different departments have different answer to that.

Right? And then you spend hours arguing over whose data is right. So, to avoid those situations, you need to kinda find a middle ground and, start from there.

Mhmm.

Alright. We've assessed, and now we want to improve. And many companies struggle with data cleaning at scale. Is there a way for AI and automation to help in making this process more efficient?

Oh, that's, an easy yes.

You know, I think, from, you know, where I sit, I believe that AI is, you know, should be a cornerstone of this improve stage. And just, you know, to remind everyone, improve, the stage is about, improving the quality of your data, by cleaning, curating, and reaching it. And AI can do those boring and monotonous data cleaning tasks for you. Right? It is it allows for scalability, flexibility, efficiency, speed, you know, you name it. So why not, ask it to do it for you, right, to do the heavy lifting for you while he were focusing on some more strategic, you know, things?

So, yeah, hope this answers your question.

I you, yeah, it definitely does answer the question in using AI. I mean, we talk a lot about AI native MDM and with Tamr being the only actually AI native, MDM tool out there.

A lot of the times, people are curious how could I utilize it to help me out. Another thing that on this improvement, space that I hear often from prospects and customers is, well, what about, like, third party enrichment? Do you have a take on that? How much does it help with the improvement, and when should companies really consider it?

Yeah. Of course. And so, overall, I do believe that when used correctly, and thoughtfully, third party data is, you know, helps a lot to, a, improve matching accuracy, and, b, fill in the gaps in your data to make it more usable, complete, etcetera. But I wanna, you know, kind of bring this point home that when used correctly. If you just go out there and buy, you know, all all all sorts of, referential data and kinda throw it all together, it won't be that helpful.

But there are ways how enrichment can support and improve data quality. So, for example, as, of course, you know, at Tamr, we built a vast corpus of referential data, which we use to match her customers' internal records against. And, by doing so, we help them, drive their matching accuracy up and, again, fill in the gaps in the data that they need, you know, that needs to be, filled in. So, yes, external data, I'm a big proponent and support supporter of of it when it is used, you know, thoughtfully.

I agree with you. I I always joke that if you had a Lenny protected corporation, maybe you would get a customer or a company three sixty for a Lenny protected corporation.

But everybody, I am not a real corporation. Not yet. Not yet. But if you're using referential match, at least you'll check the legitimacy.

So when you you say you have golden records, you actually have them, and you remove Lenny Partaki Corporation from your data.

So that's out.

Yes.

Alright.

And you also might figure out, yes, that, you know, Tatiana Incorporated is actually a subsidiary of Eleni Incorporated.

Right? And so, there's so much value in external data, but you need to be thoughtful about how you build it in the, you know, the, the process.

Yeah. And, also, not all enrichment is the right enrichment. Right? I I think, one capability that I find very useful is there is a page where you can see what like, which of these referential datasets would actually help me. Some will give me eighty percent coverage, some a lot less. So it's always something to consider, to figure out what the right fit is.

Speaking of improvement, maybe we get to a state where we do have our customer three sixty. There's people in the crowd that are in this review, phase. So you emphasize engaging business users in this process. What's a practical way to ensure that they participate appropriately?

Yeah. There I think there are a couple of things that are worth mentioning. So first and foremost, as I mentioned earlier, it is very important to, engage business users, you know, end user data consumers in this, exercise. Right? You don't want to stop at the data team because data team is are, you know, are not the ones who use this data on a daily basis and know it better.

So but when you are sharing this data, right, when you engage those diverse stakeholders, you wanna make sure that the data is presented in such a way that it's, you know, easy to consume, easy to understand. So you don't want, let's say, salespeople to be staring at a large, you know, Excel spreadsheet with hundreds of attributes, right, not really knowing what to do with it. Right? So you wanna make sure that they can interact with with this information easily.

And so, for example, at Tamr, we have those so called three sixty pages where, it's almost like a, you know, Wiki page, right, with all the best data about, let's say, a customer, kind of all structured in a very user friendly manner. And most importantly, it allows you to submit feedback really quickly and easily. Right? So you just open a form and you say, hey.

You know, there is a typo in Tatiana's last name. I think it's wrong or something like that. And then you kind of click okay. Right?

And it goes right, into the data curator's hand hands, and, you know, this person can action on this feedback almost instantaneously. Right? And I think at this stage, it is very important to remember that it's not only about, you know, improving data quality and iterating iterating continuously and, making sure there is alignment between, you know, the data and the business needs. Right?

What it is that, business org actually needs in terms, in terms of data, but also keeping people in the loop, asking for their feedback helps improve, you know, trust within the organization just kind of through this collaboration, and I think it is very, very important.

Absolutely.

Genuinely. I mean, precisely that. Because you could have a process where we utilize AI and machine learning in order to achieve a specific result. And it does provide a lot more time and back into users' hands so that they can focus on other critical things that they, care about day to day. However, I think there is always the importance of keeping that human in the loop and what happens with edge cases, and using AI in order to surface those edge cases for people to review.

I think it's critical for how organizations function because you don't want things to be a black box.

Definitely stand by that, fully.

After reviewing, a lot of people, start thinking about, okay, I am operational now. Right? And sometimes there's this very thin line between review and operationalize and what does operationalizing really mean. And I think the biggest risk in operationalizing MDM is really pushing incorrect data into business systems. So how can teams avoid this?

Do not skip the first three stages?

So as I mentioned earlier from our experience, we do offer So you said do not skip the first three stages.

That that's what you said. Yes. Exactly.

Yeah. So, as I mentioned earlier, I think at the very beginning, what we see often is that organizations are so excited about operational ease operational operational use cases that they jump straight straight into it without, fixing the data first. But, unfortunately, this does result in, you know, all sorts of downstream complications, errors, and, you know, risks risk exposure.

And so, yes, our advice would be here to ensure that data is ready. And how to do it goes through the first three, stages that we've just, discussed.

In in defining, though, operational MDM, if you can if you can do so for us, is it for everyone? Is operational MDM really for everyone?

That's a wonderful question. I get it all the time, so that's why I come prepared with the slide. So no. The quick answer is no.

Although we do advise companies to strive, you know, to, to aim to have their best data available at all consumption endpoints, including those operational use cases. Right? So, basically, connecting the best data they have, to their mission critical systems, right, operating systems.

We totally understand and acknowledge that it might not be, you know, the right fit for every organization. And the reasons you know, there are several reasons for that. It could be timing. It could be, you know, technological maturity.

It could be just the specifics of the use case. It could also be, you know, just it's not a priority for now. But you need to remember that, you know, never say never, and tomorrow can you know, things can change. And so our advice our advice here is typically go through the first three stages, make sure your data is ready, start getting value from those three stages, and let's say, you know, continue using this, you know, best version of your data for analytical use cases.

And then when the need arises, when you and your organization are ready for operational use cases, you will have, a, the data ready. Right? And everyone will already be, you know, on on board with it. So, again, just start going through the first three stages and kind of building a solid foundation for operational use cases down the road if it is not a priority for you, you know, as of now.

Are there organizations that have nailed this MDM journey, and so immediate impact or what what does this really look like in real life beyond just the stages?

Yeah. Of course. And again, I've done my homework, so I have a couple of examples of very, you know, impressive work that other organizations done.

And I want to You're the best you're the best, person to bring on the webinar, Tatiana.

You've done all the prep. I you don't need me here. I can just leave. Everybody, you can maybe we should put another poll. Do we need a let in again for these webinars? Yes or no?

I do. I do. I totally do. So a couple of examples. And before I jump into, you know, sharing their stories, I want to kind of, you know, start with it, with a, you know, note or disclaimer. Right?

As you listen to those customer stories, you will notice that both, those customers, they operate in different industries. Right? Financial services versus, health care. They had completely different starting points.

They had completely different, you know, destination in mind. And, we were so excited to meet them where they needed us the most and, you know, partner with them and help them walk through their their unique MDM journeys. And so with that, let me jump into the first story. Right?

Mizuho Bank. So they had very, kind of legacy, approach to data management, and their goal, their business strategy was to grow their business in the US. So a couple of success factors that, you know, played critical, role for them were, first, choosing the modern, flexible flexible MDM platform that allowed them to implement faster and, you know, and scale and also, you know, go through changes and updates, whenever they needed.

So, secondly, machine learning driven entity resolution proved to deliver better matching accuracy, better matching results.

And, one interesting piece of feedback that they also shared was that, they try to focus on the quality of the output data versus the input because this way, it was much easier to connect it to, you know, kind of real world problems rather than just, you know, some hypothetical ones. And so a couple of pitfalls that they, you know, recommend avoiding are, first, data quality issues delay progress. And second, do not basically go with operational use cases before you fix the data. Otherwise, it causes, you know, problems down the road. And And so another example would be from AM and Healthcare. And, again, completely different, starting point, different industry. Everything is different.

So they had a solution in place which they, wanted us to replace. And this solution, it was not scalable, not flexible enough, and lacked real time, capabilities, which was critical to, delivering instant, excellent, exceptional clinician experience. Right? And this was their main goal. So a couple of success factors that they, mentioned were before doing anything, clearly define the problem that you're trying to solve for and align it to business strategy.

And then evolve business users early and often. And lastly, pick the right technology partner. And so a couple of pitfalls that they can suggest avoiding, at all costs, do not boil the ocean. Right? Again, focus on the, key entities that will have the biggest impact on the business.

Secondly, do not rush with implementation.

Right? Make sure you implement things, the way that works best for your organization depending on, right, how things function and, you know, all the processes internal processes that, you have in place. Last but not least, do not overlook the importance of external data.

Okay. There is, I think a lot to think about there, and I'm sure, if anyone is interested more in these particular use cases, there is more information, that you can find.

But we gave a lot of information to all the people at the webinar today. So if they leave with one thing, other than the fact that Eleni Partagie Corp doesn't exist, I don't have a company, what's one piece of advice to a company just starting their MDM journey?

Read our ebook.

But, just joke aside, I guess I would say a few things. I know you're asking for one, but I have three in my, sleeve. So the first one is remember that MDM is not a project. It's a journey.

And and when you're embarking on it, just remember to start small, show progress, and iterate as you as you go. Right? You might learn that something isn't exactly the way you wanted it to be. And so keep this kind of, you know, mindset, that you need to iterate as you go.

And last but not least, don't do everything yourself. Use, you know, modern technology. Use AI. Use machine learning to help you along the way and do the heavy lifting, for you so you can focus on more strategic, you know, strategic components of this, MDM journey.

Thank you, Tatiana.

We're almost up on time and with a minute left, for everyone. You've seen mastering the MDM journey is essential for organizations looking to turn fragmented, inconsistent data into a trusted business ready asset. And following a structured approach, other than it making you organized, such as assess, improve, review, operationalize that Tatiana gave us today, organizations can overcome common data challenges and ensure their data is not just clean, but also actionable.

So thank you so much for being part of the discussion today. We hope you're leaving with fresh insights, into how Tamr can empower organizations like yours, and we'll be sending out the webinar recording soon. But if you'd like a deeper, little dive, feel free to request a demo to see Tamr in action. Maybe we will meet each other again.

You can watch a video on how we build golden records, and you can also check out our latest ebook, on the MDM journey, which Tatiana just mentioned a couple moments ago.

Anything else, Tatiana, that you'd like to leave the audience with?

So if you are interested in learning more, about MDM journey and, hearing, firsthand from Mizuho Bank and AMN Healthcare. And if you are attending Gartner's data analytics summit, in Orlando next week, make sure to stop by our booth and, stop by our speaking session. But thank you for having me today. It was, great, you know, joining you and the rest of the audience.

I hope this was helpful. And, yeah, good luck on your MDM journey. And if you have any if you need any help or have questions, feel free to reach out.

Thank you so much, Tatiana, for your time. And thank you to everyone for joining the webinar today. And we'll see you very soon for the next webinar or at the Gartner, Orlando Summit next week. All right. Bye.

Thank you. Bye.

‍