Webinar: Master Data Management in the Cloud

Your data and applications are on premise and in the Cloud so why limit your MDM options to the ground?

Until now, MDM solutions were complex and expensive on premise software, not designed to address the new challenges and opportunities of SaaS and other cloud-based apps and data. Cloud MDM changes all that!

Watch us demonstrate how an innovative Cloud MDM solution quickly solves these real-world master data management problems:

• Synchronize master data across Salesforce, cloud and on-premise apps
• Ensure data quality across ERP, CRM and eCommerce systems
• Enable data governance by empowering your business users to take control

See a demonstration on how Cloud MDM supports data management, real-time synchronization and simple-to-configure data governance.

Cloud MDM Webinar Video Transcription

Elias: Welcome to today's webcast on Master Data Management in the Cloud. My name is Elias Terman, and I'm Director of Product Marketing at SnapLogic, and I'll actually be moderating today's event. Today I'm joined by 3 experts in the areas of MDM and Integration. Christophe Barriolade is an MDM luminary and the CEO of Orchestra Networks, the creators of SmartDataGovernance.com. Zeb Mahmood is Principal Product Marketing Manager MDM at SnapLogic, and brings with him eight years of experience at IBM, where he helped deliver MDM solutions to global 2000 companies. Last and certainly not least, we have Craig Stewart, Manager of Solutions Engineering at SnapLogic, a 20-year veteran of the Integration Industry. So, quick agenda: we're going to talk about SnapLogic and Orchestra Networks, and talk about why is MDM so difficult, give you an overview at a conceptual level of the Cloud MDM solution, and then talk about some of the challenges and opportunities in getting data in and out of a Cloud MDM solution. And then we're really going to spend the majority of our time in the software [life]. First Christophe will give a demo of SmartData Governance.com, and then Craig will take us under the covers and see how data moves in and out of this MDM system. And then we'll end with a robust Q& A session, time permitting. So with that, I'd like to pass it over to Christophe, and why don't you tell us a little bit about Orchestra Networks?

Christophe: Thank you Elias, good morning everyone. A few words about Orchestra Networks. We are a multi-domain Master Data Management vendor. The company has been founded 11 years ago, and we today have a global presence across Europe and North America. For more than 10 years we are selling an Enterprise MDM software called [EVX], with more than 60 large accounts live, worldwide. Last year, we launched the industry's first Cloud MDM solution. It is called SmartGovernance.com, fully based on our Cloud MDM technology, but available as a service in the Cloud.

Elias: So, a little bit, thanks, Christophe, a little bit about SnapLogic. I mean, basically what we've done at SnapLogic, we've taken a modern approach to Enterprise Application Integration. It's a 100 percent internet technology for integration in the age of the internet. We were founded by Gaurav Dhillon, our CEO, who also founded Informatica, where he led that firm from a garage operation to a publicly-traded, multibillion dollar software company. We've been fortunate in that we've had great validation by the analyst community, including winning the Gartner Cool Vendor Award, as well as the Ventana Research Cloud Computing Excellence Award. And, we've had great success solving a wide variety of complex integration problems within leading enterprises across the globe. Our Headquarters are in San Mateo California, and we also have global offices in Chicago, New York City, London, and Munich. So, what is SmartDataGovernance.com? SmartData Governance is a Cloud Master Data Management solution that can really help improve the health of your business. So, your data is no longer tethered to a specific business application, but instead lives in the Cloud. It's completely web-based, and can be easily linked to all your on-premise and cloud-based applications, via SnapLogic's integration platform. There's no hardware or software to install, and you'll see how it all works in just a few minutes. But before we show you the solution live, we wanted to give everyone an overview of some of the challenges people encounter when implementing MDM. So, Master Data, when woven together, creates the DNA of your organization. Your operations need to have the same view of your Master Data, whether you're selling a product, or hiring an employee. [Analysis and] reporting relies on the same Master Data hierarchy, whether you want to know sales per product, or revenue per region. So, why is something so logical so difficult to implement? Well, you've got different teams in different locations that need to share data at different points in its life cycle, often modifying it along the way. A business tends to grow organically, with solutions sprouting out wherever needed. Put simply, when data is duplicated, for example to serve a particular business application or operating unit, it's inevitable that the data will become inconsistent. And this is further complicated by the burgeoning number of Cloud Apps being adopted across your enterprise. So, what do people really need? Well, they need a single Master Data Store that will enable them to do what they do. Like providing a unique definition of Data Quality rules that everyone can adhere to. Of course, there needs to be proper roles and access rights, and there needs to be a way to quickly integrate with applications. SmartDataGovernance.com works for all your reference master data. And that's important, because often-times your different data domains are linked. For example, your HR Codes and employee data is linked to your stores and offices, which is linked to your reference data about locations and countries, just to give one simple example. A successful MDM Program starts with the business users and keeps them in the loop. From the definition of a data model, managing hierarchies, to the governance of master data over time. Of course, doing everything through a web browser is great, but how do you get your application data in and out of SmartDataGovernance.com? SmartDataGovernance.com allows you to easily snap into your Cloud and on-premise data sources through the SnapLogic integration platform. SnapLogic can be deployed on-premise, in the Cloud, or in a hybrid deployment model. The choice is yours. And as you'll see shortly, the SnapLogic Designer provides full access to your master data. I've asked Zeb to take us through a quick illustration of MDM within a large retail organization. Let me pass the controls over to Zeb.

Zeb: This is great, this is awesome. So, what you just talked about, what I wanted to do was just take a minute or two, and step through a very realistic scenario, right? So, like you mentioned I've been working in this space for, since 2002. I've done implementations myself, and one of the first implementations that I worked on, it was pretty much what you're looking at on the screen right now. So we had a bunch of store systems. Product data was coming in. [inaudible 06:57] step back, imagine yourself to a large retailer, right. So you have several vendors, which are sending you data, and the data is coming into a standard system, maybe a Global Data Synchronization network, which is very well structured, very well defined by now, it's been there for a while. But then you also have your internal systems, which are sending you product hierarchy information for example, for your financial systems. And then you will have these small number of vendors who are not on GDSN. Or, even some who are, they have some attributes which are not covered by GDSN and so you will have attributes which are coming in through different sources. Now, you bring all of those together, they're coming in, you have a nice master data management repository that you've set up. You've enforced a nice data model, required hierarchies, your workflows for data governance are in there, and it's nice, clean harmonized data. Very good. Now you're embarking on a multi-channel commerce strategy. So you have your products being sold through print catalogs, traditional print catalogs. You have the same products being sold through your .com website, or maybe even a b2b website, so imagine yourself to be like Staples. And then you have these in-store kiosks, very common these days, right. So you have [PriceCheck], people can go and self-serve customers, and check the price on a product. So all of this is going well, and you've done an amazing job, and you think "Yeah, hey, you know, awesome. I didn't really need to use any data integration in there, and I have point-to-point integrations, everything's working just fine." But let's go to the next slide. And, as we go to the next slide, you see here there are, this is how I would have done things now, you know. Through my experience, what I've learned is that it's very important to realize up-front where you're headed, right? So let's plan it like that, let's have a nice data integration there in-between. Let that be the gate-keeper for your data how it's coming in, and then you have the data for the [inaudible 08:56], all the vertical processes that come with data governance, master data management, in-between. And then you have that data being sent out. So I know you're still looking at me like "Hey, Zeb. Big deal, all right? So I had already done the point-to-point, now you put another integration layer between. What's the big deal, here?" Let's go to the next slide. So as we go to the next slide, this is what's happening. This is very realistic. Because you've done a good job, because you have a great master data management repository, which has brought in good product hierarchies, clean data, consistent data, complete data, all the attributes are filled out. So, you have a lot more systems that are asking for that data. So on the target side, on your right side, now you're embarking upon selling your products on global commerce. They want the data as well. You know what, the Customer Support said we want to look at the same data that you're selling so that when people call us we know exactly what to talk about. And even on the target, on the source side what you're looking at is different systems sending you data which were not sending it to you earlier. Now, the consumers are more environment-conscious, so they want to know the carbon footprint. They want to know how much recycled material was used. The packaging information is coming in to help with supply chain systems. So, point being that you are growing the number of systems that you integrate with both from your source side, your target side, and this is what you will see if you have really done a good job at master data management, which I'm sure you have. So, plan ahead, bring in your data integration architecture before this thing happens where you have to go back to the whiteboard and say "Hey, you know, we're going to take six months just to re-architect." You architect in a way which can scale for the future, you have been foresight-ful, you're not caught blind-sided. So, with that I'll leave you.

Elias: Thanks, Zeb, that was a great overview, appreciate that. I think the point here is that today's hybrid environments are changing quickly, right. And, and that's why it really makes sense that you'd want 100 percent internet technology for integration in the age of the internet. And that's exactly what SnapLogic does. Like the web itself, SnapLogic is built on lightweight yet powerfully strong restful architecture, where all integration resources have a URI, so users and infrastructure can be anywhere. So these Snaps, or intelligent connectors, you download them from the Snap Store in seconds, making it easy to connect to a wide variety of both on-premise and cloud-based data sources. And you can even make your own Snaps. This enables lines of business to access the data sources they need, but through a controlled and standardized environment. And our Visual Designer, which Craig will show you in a few minutes, it really encourages adoption across the enterprise. So, there's Snaps for database connectivity, there's enterprise Snaps, SAP, Social Media. There's over 100 Snaps in the Snap Store, and when you download these Snaps into SnapLogic, basically what you have is this rich palette of modular components that are re-usable, that you can drag and drop onto a visual canvas to create that first integration in minutes. And you can even monitor your entire integration ecosystem on your iPad, 24/7. But enough talk. let's see both solutions in action. What we're going to do is Christophe, let's start by taking a look at SmartDataGovernance.com, and then we'll go over to Craig, who'll show us what's happening under the covers as we move application data in and out of the MDM solution. So I'm going to pass the controls over to Christophe. Take it away, Christophe.

Christophe: Thank you very much, Elias. So, now I propose to start our demonstration by showing SmartDataGovernance.com. So what you see here on my screen is the Home Page of the website, where you can find a lot of details about the features of the solution. As I am a user, I can now log in into my own MDM account. So, again, it's 100 percent browser-based, so all the features of the MDM are available in a web browser, and it is fully cloud-based. So when I log in to my account, you can see that I have access to many MDM projects. I'm going to use this one called SnapLogic MDM, where you can see I have access to my directory, and to some statistics. So when I connect into this MDM, I have now access to all the features of the solution. So first, what you can see on this screen, we have a few tabs, that group the main features of the MDM. Data allows me to group and manage the content of my MDM. Data Spaces allows me to manage different versions of my master data, we'll see it later. Then we have Modeling for defining my own data models and my own workflow models. Then we have the Main Workflow UI for end-users. And finally, Data Services for [inaudible 14:17] data integration and [capability]. So first let me start by showing you some examples of master data I am managing with the solution. First, I'm selecting the version. It means that I can work in multiple parallel versions on my MDM. So, I will start using the reference, which is my main production version of my data. Then I can select basically the master data domain. So you can see here six different domains, like products, stores or organization. It's unlimited so you can define any type, and any number of data domains, and you can even link two different domains together. So let's have a look to product, here. You can see we have various data objects, such as product families. So I can directly view the content of my master data. So here it's just a simple reference table. If I select now Products, you can see I have more than 800 products in my MDM, and I can view it as a table. I can search, so you can have full-text search, multi-criteria searches, and I can select Record to view the details on all the fields of my product. I can also organize my master data as hierarchies. Actually here, a product is linked to a family which is linked to a category. So if I go here to the view menu and display my product hierarchy, I will have access now to a tree view of all my master data, and I will be able to manage this hierarchy. I mean, Move Nodes, Attach Nodes, etc. So basically, this user interface allows any end-user to access the MDM and to make some updates. Before showing you a change management scenario, let me show you how easy it is to change the data model, the underlying data model of my MDM. So, going now to the data modeling part of the solution, I can select my product data mode, and let's say we need to introduce a new attribute for our product. So, again, it's web-based, very easy to do and without any coding. I can go here and drill down to my product object, and decide to add a new field. So let's say we want to introduce say, date for our product, and also we need to create a small business rule that says each product can have up to three dates. So simply, we just have to state that the maximum number of elements for product date is three. And that's it. Now I just have to republish my data model, and on the fly it will update my MDM and all the user interface of the solution. So let's do that. I confirm this publication, and go back to my data set. So if I select one of my products, you can see that here I have a new attribute called sales date. If I look at the documentation I can see the type, the rule, and I can now fill up to three dates for my product. So, it's very easy to prototype and to define your own data model using this solution without any complex development or complex development to do. Now, let's talk about version, because in MDM, versions are very often because you probably want to have your current version of your data, keep track of the past, and also work in the future. Let's say we want to introduce a new product for next quarter and also to reorganize our product hierarchy for the next quarter. What I can do is go to the Data Spaces tab, and create a new Data Space. So it's like a new version or a new staging area for my MDM. I can call it Next Quarter Product, and now I can work inside this data space. So you can see on the left side of my screen I am working inside this new data space, so which means I am totally isolated from the main data space. So let me do some updates. So I will display my hierarchy, and I will decide that some of my LCD monitors here will be moved under a new family. Okay. So I will move it here under integrated monitor for instance. I decide also to rename a product category for the next quarter, so I will select this category and say now it's called Display. I can also make dates in other parts of my MDM. For instance, if I now select Stores, I can manage my different stores per regions. And actually, here, we can see the geographical hierarchy where I have all my stores. And again, what I can do is update and say, what, Central Region will have a different manager, for instance. So, I am now still working in this data space, but now let's say we are in the next quarter, and we want to apply those updates to the main MDM, because we want to publish those new master data to all the target systems both on-premise and in the cloud. So what I need to do is merge my data space with the main one. And this solution will compare those two different data spaces and show me the differences. So if I go to this category object I can see we have an update on one category, which has been renamed. If I look at Product we can see that three products have been updated. Actually, they have been moved to another family. And finally, in one region, Central, the manager has been changed. So you can review each individual update and then decide to merge those two data spaces. Another important feature of SmartDataGovernance.com is the workflow. In the real life, you probably want your business users to update the master data via step-by-step processes. So with very clear security, and change management and approval processes. So SmartData Governance comes with its own workflow. It's fully configurable, and end-users have access to a very simple UI for reviewing their workflows and doing their tasks. So here you can see that I have one pending task under the new store [office]. Actually we are creating a new store in San Mateo, so if I select this workflow, you can see a graphical display of the workflow status. Actually, the first task, called create store, has been completed a few minutes ago by your user. Actually it was myself. And now we need to approve this the region of the store. So I can take this task, actually, I also received an email for notification, and I can review the status. So I can say okay, San Mateo, it's in California, so I will validate. I can also check the location so here we have an integration to Google Maps. And then I can add a comment and say "Okay for me," and accept the new store. Super-simple. There's no training required for end-users and it means that a lot of people across the company can be involved in data governance day-to-day, using the solution. So, before Craig shows you the second part of the demo which is basically the integration, he will show you how easy it is to connect to SmartDataGovernance.com, pull some data updates and populate any target system. Actually, one of the objects we'll deal with is in my product catalog, the Manufacturer table. So here I will go back to my main reference master data set and create a new manufacturer so I'm sure the solution will take this new data and populate my target system. Elias, before Craig continues the demo, any questions?

Elias: Sorry, I was on mute. Someone asked a question about the language button in the upper right hand corner. I think what they were getting at is this, if you were more comfortable for example in a language other than English, can you use the interface in a different language, I think is what they were getting at. Christopher: Yes, sure, so the solution is fully multi-language, and here in my demo I have just configured English and French. But actually you can design your MDM and your master data model in any language. And when I switch here to French, you can see that both the user interface, but also my data model, are translated into French. So it means the same MDM can be viewed and manipulated by users in multiple languages.

Elias: Fantastic. Thank you, Christophe. Now what I'd like to do is switch it over to Craig, and let him show you what's happening under the covers here, as we move data in and out of the MDM. Take it away, Craig.

Craig: Okay. Thank you, Elias. Hopefully you can hear me?

Elias: Yep.

Craig: Thank you. Okay. So, SnapLogic is the technology that we're using to move data in and out of SmartData Governance. And that is enabled by adding a Snap to the SnapLogic environment, and SnapLogic has a comprehensive set of different applications which we support through the provision of a Snap. And they're all available for download from the Snap Store. And if I just browse across here in my browser, you can see in the Snap Store we have various different applications, we can search for them in various different ways from the Snap Store. Just a few of those that you can see on this page. We've got obviously things like Sales Force and SAP. Some of those with prices associated with them. And some of them like [Box], and SmartData Governance which are built and provided in the Snap Store for free in cooperation with the application vendors, to actually enhance the ability to get data in and out of those applications very easily. So, what do I mean by very easily? Well, the first thing that we do when we want to use an application with SnapLogic is we add that application to the Foundry. And to do that we just simply install the new Snap, having purchased it on the Snap Store it gives us a link. We just need to add the link into the Snap Foundry here, and it will install it automatically and provide for us the access, the components to be able to work with that application. So SnapLogic itself includes all the functionality to be able to move and manipulate data. Accessing databases, files, converting data, filtering it, sorting it, all of those kind of things that you're going to need to do whatever the applications you may be working with. In addition to those, in my environment here, I also have a set of other Snaps all in place, including SmartData Governance and Sales Force. If I look at the SmartData Governance Snap here, you see that in common with many of the other Snaps, we get a set of components which we can then use to actually access the applications that we're wanting to work with. So, in that SmartData Governance that Christophe was working with a short time ago, we had that Manufacturers table. And if I wanted to actually read that data out of that manufacturers table, and perhaps upload that into my Sales Force environment. I want to read that table, so I can create a new pipeline. So I'll just make a new feed here, we'll call it My Pipeline. I just take that Read Table component and drag it onto my pipeline, which is going to be my data flow, and all I need to do now is configure that. So when I want to configure that, I have earlier created my connection, so I just drag my connection down to my object here, and having done that, I can then use the Suggest functionality that the Snap provides. So, I then can say "Okay. Show me the things that I can then select from that connection." If I just move this up here, we can see that inside there now I have a drop-down on my data space, and it was the reference data space that Christophe was using there. So I'll just save that. Within that reference data space, again, press Suggest. Then it goes and [introspects] onto it to see. . . Now I can choose the data set, and it was the product data set that I was looking at within there. And again, suggest one more time, and then we can then actually see that we can then select the manufacturers table. And this time, it will actually, when I press the Suggest, go and get me all of the data, the metadata of that particular table. And in this case it's just those couple of fields, there. And that then gives us access, adds those to the output feed so that we can then go and get that data and do whatever it is that we want to do with it. This time I'm not going to use any of the inputs, so I can just get rid of that. And if we want to preview the data, the object here just gives us access to be able to go and query this in real-time. This is going out into the Cloud [instance], of SmartData Governance and pulling that data back. And there you see that new manufacturer that Christophe added. Okay. So now I know that I'm getting the right data out of that component, I might want to then upload that into my Sales Force. And earlier I used the Sales Force Wizard to generate me the necessary components to be able to work with Sales Force itself, and inside there I've got these different components to be able to read and update my account table. So I'm just going to take those manufacturers and actually do an upsert of that data into Sales Force. And in order to do that, if I take the data out of the Read component and drag it to my Upsert component, it will automatically then apply SmartLogic's SmartLink technology which will map automatically for you the identically named fields, the fields that you've mapped previously; so it learns as you use the tool, it actually learns the things that you link together. And then the third phase is some fuzzy logic, which is how it works out that country links to billing country. The other things in here I've linked: Manufacturer ID to name, automatically, it's remembered that I've done that previously. And I may also want to add things like, I'm going to put an external key in there. I need to identify what that's going to be if I'm going to use the upsert function of Sales Force of course. So I'll just save those mappings, and within the Sales Force object here, you'll see that we do things like batching the data into Sales Force by actually taking the data sets and making up to 200 rows at a time in each batch that we push into Sales Force. It automatically here picked up that SAP idea of the external field that we're going to use to do the upsert, so all of that information is ready to roll now. But you may also see that on the bottom half of this object here, it's actually got an output. It's all very well knowing that we've thrown this data at Sales Force, but it's also useful to be able to figure out whether it actually worked. So generally what we do is we will take that output from Sales Force, and interpret that. Maybe in this case, I'm just going to put that into a Writer component. And because I just dragged that from the Foundry, it's un-configured, so I'm jut going to call this Audit.csv. And what that will allow me to do is actually see what the output is from that upsert so we can see whether it did inserts, updates, etc. So we can interpret that in various different ways. So this pipeline is actually going off talking to this cloud application, the SmartDataGovernance. It's then taking that data, upserting it into Sales Force, and then giving us the results of that to be able to see what's actually there. Now, if we just save that, we can then execute this from within my environment here. Just press that execute button. Off it goes, and actually doing this in real-time in the Cloud, pulling that data and moving it across. We can then look at the execution log, and we can see that in a text form, or if we go across to the stats here, we can then see what it's actually done, how many rows it read and wrote each time. So here you see it read from the SmartData Governance, 58 rows, and when it did the upsert, it then wrote that result out to the file. So, let's go and actually see, take a look inside that audit file, and let's see if it actually wrote that data as we wanted it to. Well there we go, we can see the IDs of the objects. It was successful. We can see any errors that may have been generated, and whether the component was actually created or not. So this gives us a comprehensive ability to understand what's actually going on in our environment. In order to work with any of the many applications that we can, we just use these drag-and-droppable components that allow us to then just move the data, and concentrate on the complexity of our data and our particular business processes, rather than actually having to work with the details of how to make connections to these different application whether they are cloud-based or on premise, whether it's a thing like a file or a database or a much more complex application like SAP. And any of the components that we have in our environment, we can then integrate the data from any of these different applications. So we can integrate not only with SmartData Governance and Sales Force, but if you had SAP in your environment, or Oracle eBusiness Suite, you have the ability to integrate all of these different applications together, and insure that the data is flowing and that you are taking advantage of something like SmartData Governance to insure the quality and integrity of the data that you're actually using within your business processes.

Elias: Hey Craig, I've got a question. Someone was asking a question. Oh, I'm sorry

Craig: . . . [brief] demo, yep.

Elias: Someone was asking a question, sorry to interpret you, but. . .

Craig: Go ahead, Elias.

Elias: Yeah, thank you, Craig. Someone was asking a question about scheduling, like how do you schedule the movement of data in and out of the MDM, and other systems.

Craig: Okay. So, this task that I've got here to read that data and upsert it into Sales Force, if we want to actually make that happen on a regular basis, each of these pipelines can be scheduled. So we have built into SnapLogic a scheduler. And if we just take a brief look inside there, we can specify when we want this to actually happen, whether it's once a day, once an hour, or right down to every minute, to actually make this actually move the data as and when it's required. We can specify solutions that we want it to happen Monday to Friday but not on Saturdays and Sundays, etc. And we can also specify the things like what should we do when we've completed this, do we send a notification out automatically to say this happened successfully, or whatever. So, we have scheduling built into SnapLogic, but of course many customers may actually have some enterprise scheduler functionality in-house already. And you'll see throughout SnapLogic, these URIs available. Now, if we want to actually execute this from somewhere outside, we can just use the URI here, and ask SnapLogic to execute it. Obviously we'd need to parse in the authentication parameters, because you need to make sure that you have the right authority to execute a particular task. You can parse in parameters, you can have this executing, being called for instance from Sales Force. You have customers using [AppEx] triggers within Sales Force whenever a new account is created, and notifying SnapLogic to then move that over to other applications. And equally we could have SnapLogic just holding the application or database and moving that data which may have changed, moving the [Deltas] across, essentially. So we have the ability with SnapLogic to be SnapLogic oriented and using the SnapLogic Scheduler to make things happen, or we can actually integrate with third-party schedulers, the likes of Tivoli and other enterprise functionality that you may want to use to make stuff happen, through the standards-based approach. That's one of the beauties that having a product like SnapLogic being entirely built on the technology of the cloud, is actually able to represent. Any other questions there, Elias? Okay. In that case. . .

Elias: Sorry, I was on mute. I apologize. I was talking away and I was on mute. There was a question about multi-tenancy. SmartDataGovernance.com is a multi-tenant solution, but it also, the architecture allows for the physical isolation of data sets. In the case of SnapLogic, basically the way we offer it, it can be deployed on-premise, using standard commodity hardware and software, on-site in your data center. We can offer it as a service, as a virtual cloud server, or it can be done in a hybrid environment. So really, the choice is yours. There's a lot of great questions that are still on the call, and what we're going to do is make sure that we follow up with everybody afterwards, but...