 Hello and welcome, my name is Shannon Kemp and I'm the Chief Digital Manager for Data Diversity. We want to thank you for joining the latest in the Monthly Webinar Series, Lessons in Data Modeling with Donna Burbank and sponsored today by IDIRA. Today, Donna will be discussing conceptual data models, how to get the attention of business users. Just a couple of points to get us started. Due to the large number of people that attend these sessions, you will be muted during the webinar. We very much encourage you to chat with us and with each other throughout the webinar to do so. Just click the chat icon in the top right hand corner for that feature. For questions, we will be collecting them by the Q&A section in the bottom right hand corner of your screen or if you'd like to tweet, we encourage you to share highlights of questions via Twitter using hashtag lessons dm. As always, we will send a follow-up email within two business days containing links to those slides and links to the recording of this session and any additional information requested throughout the webinar. Now let me introduce to you our speaker for today, Donna Burbank. She is a recognized industry expert in information management with over 20 years of experience helping organizations enrich their business opportunities through data and information. She is currently the Managing Director of Global Data Strategy Limited where she assists organizations around the globe in driving value from their data. She has worked with dozens of Fortune 500 companies worldwide in the Americas, Europe, Asia and Africa and speaks regularly in industry conferences. In fact, she will be speaking at one of the diversity upcoming conferences, Enterprise Data World 2017 in Atlanta, the first week of April here and just a little over a week. She'll be starting with a tutorial on best practical steps to implementing metadata strategy at that event. So Donna, hello and welcome. Hello, thank you. Always a pleasure to do these webinars and it's good to see some familiar, I would say familiar faces but familiar names and some of the names are faces are familiar as well so thanks guys who've joined us on a regular basis. And as Shannon mentioned today we are talking about conceptual data models and hopefully we'll make them less conceptual and focus more on the second part of the conversation which is really how to get the attention of business users and probably more even valuable how to get business value out of these models. So, Shannon has always done an excellent overview of me so I don't want to keep going on that. Just a few things. I am on Twitter at Donna Burbank and then today there's a hashtag, LessonsDM so if you are a Twitter fan continue the conversation online and I know there's always a lot of good questions at these events so hopefully we'll try to leave some time for that at the end. I think my bio is up there so kind of have a long background in data modeling, metadata management both in the field and with several product vendors written a few books on data modeling and probably most notable for this conversation was product manager for a while at Idea. Back then it was in Barcadero ER studio so any of the really great features of that product that you use are mine and it doesn't work if you don't like was built after I left so that's a little bit about me. A bit about the series and as Shannon mentioned these are all recorded and the slides are available too so if you missed any in the past they are still out there and I know a lot of folks who are busy during the day often have to catch them later so we appreciate that as well. So we've done some things that enterprise architecture business intelligence in the past. You'll see the theme is to really put data modeling in context which hopefully we'll do a little bit slightly differently this time but hopefully in that same theme his data modeling in and of itself is we might find it interesting but it's not really that interesting as just a standalone artifact and that's what we'll try to head on today but it's really only useful and useful to a business when you're doing something else with it like BI or understanding enterprise and that kind of thing. So that's kind of the theme we tried to cover as well as some of the new technical technology things like graph databases that are hot now so hopefully you can join us in some of the future events as well. They are all available for registration if you want to get them on your calendar early I know some people do it all in January and then just have them out there so appreciate those folks that seem to keep coming back so I appreciate that. So today the topic is going to be really about how a data model facilitates communication and that to me is really the main topic of conceptual data modeling is that business focus. So if we remember and we'll go more about this the focus is really on a business audience and so the display and the rules and the things we're capturing really should be all about the business at this level. In the second note I love distress the simplicity does not mean lack of importance and I think a lot of us need to do the more simple you make it you should have seen get less credit right and we can make it hard. Can you take a very complex topic like an organization or all the data that's flowing sort of organization and really sum it up in a one pager and that's really the beauty of a well formed conceptual data model and that goal is to make something very complicated and make it really intuitive so you can really suck out those core concepts that really make sense and as always a model or tool or any technology is only part of the solution and I think this is even more important when we are talking about conceptual data model because the whole point of its communication and for some of you I might go a little outside the line of what some people think of when they think of a conceptual data model and I'll stick to it because I think the goal is communication right so someone wants to fly a plane with you know smoke signals in the back to get the attention of the business go for it you know I think so I think we have to be a little more creative sometimes to get some of the stakeholders that were not particularly used to dealing with the cell and the process and best practices around that to get that consensus and buy in and yes that plays into some of the other things we'll be talking about throughout the year like data governance and some of those things that really are another piece of the puddle puzzle when we talk about data modeling. So again you've probably seen this framework again hopefully it's helpful to kind of bring it back to some of the core concepts so you know in our practice we always start with the top down why are we doing this what's the business strategy and how does that align to a data strategy and then we start from the bottom up as well of what all the technology and that we're talking about unstructured data here is it all relational databases etc etc and all the pieces in between and I won't go in detail on this at the time if you've joined other webinars you've seen this but what I wanted to cover here is really how data modeling and data architecture and I'm going to include metadata in this because a lot of the value in a data model is the metadata around it a really a core part of any strategy you can have a great strategy but if you don't know what your data means or where it is or how it's stored and particularly in this conversation what it means and the rules around it you're not going to get much value so we often start at the very top of a business strategy starting with the conceptual data model what are the core business concepts we're worried about we have a campaign that we're talking about customers prospects product those are the core pieces of a business and I think when you communicate that right to a business audience folks get that and you're part of the conversation and how you really can link that strategy so again you've probably seen this triangle too if you've joined other webcast but again hopefully it's helpful to put it in context so when I'm talking about business data I tend to stay either at this conceptual model or it ties into the logical and we could have a whole topic just on this is the logical model of physically focused or business focused and I'll say both and you can see that line is strategically placed it catches business rules and at this point you are starting to think about how that may apply to a data structure this the focus is still business but there is some structure around that that we're trying to put in but I would say but the conceptual level we really are really trying to model the business unless the technology or it shouldn't be at all about the technology behind the issue completely a business focus you also see that the pyramid gets smaller at the top and that sort of hits on the point I mentioned before is that less is more and simplicity is the beauty of a conceptual model so if you could think of the bottom of that pyramid where there may be thousands of physical tables can you logically and conceptually summarize those that we're really just getting that view of customers product sales inventory that kind of thing that's a beauty of a conceptual model so simple is better so you've heard me say conceptual modeling we always get the question so hopefully proactively we'll say this and I will say there's our friend Shakespeare up in the corner if you didn't recognize him or I'll say the cobblest children have no shoes right as an industry that whole point is creating common definitions we're pretty bad at that ourselves and there's a lot of different names for this model I'm talking about because I was all a subject area model some kind of business data model enterprise model seems to be the major theme is folks call that a conceptual data model so in the book that I referenced in my bio data modeling for the business my co-author Steve Hoberman actually did not a very scientific survey we had it out in that day the DM review with a big publication of what people call this thing we described this thing without using a word we've called it in our book a high level model just to avoid that conversation because in full disclosure Steve Hoberman called it a subject area model so I think those 12% go to his classes I'm just kidding a lot of respect for Steve he's a good guy so I can see him and the rest of us call it a conceptual data model so I can say ha ha Steve I'm right because we actually did a survey but I particularly don't care what you call it I think the point and hopefully that in this presentation is it doesn't matter what you call it you probably shouldn't call it a conceptual data model to a business user I would actually recommend not doing that that sounds very vague right so maybe it's your business model your business data model your business information about whatever you call it the point is communication so I won't argue semantics although I think I just spent a minute doing that I'm guilty myself but we tend to do that because we're modelers we think of semantics and definitions and sometimes we can go too far so I will talk about that more in the presentation but if you call the thing I'm discussing any of these things that's fine we'll have you in our club and as long as you know what I'm talking about I think we're fine so high level model I think is what we use in the book just to keep it neutral so and just to set the context I mentioned things about metadata so technical metadata is hugely important often folks start there because it's easy and in many cases to kind of reverse engineer I won't say managing technical metadata is necessarily easy but you know it's got to get your structure of integers in your car or our car whatever we're talking today of more of the business metadata so what do we mean by M-Squally what I mean by customer and then of course the data is the actual customers themselves but more of the reason I put a picture of Mr. John there is we'll come back to this I think visual presentation especially at the business level is so important and I like to highlight that when I build my conceptual models because really if you're thinking of a business these are concrete things one of the reasons I don't like the name conceptual data model it's actually to me the opposite we're talking with the actual concrete things that run a business right locations and customers and products and employees those are the things they're things they exist they live and breathe in some cases so kind of remind that and the picture I have there metadata again is this is a survey report I did earlier last late last year with diversity on what are some of the emerging trends in metadata management again this is broadly metadata and not specifically data modeling but what's interesting is that 80% of users of metadata are the business and when we did a survey of a lot of the customers or users of metadata a lot of the conversation was around just this that it helps business users and IT really understand the context around information and really the usage and the business rules and all of that so for those folks to say you know that the business doesn't care about data I would argue with you that I think 80% are saying they are I think there's some frustration of how we communicate to those folks in a way they understand and one tool by all means not the only tool but one tool at least at the starting point can be this conceptual data model so that they will be interested in listening I do believe in fact my next quote I just remember this from one of my consulting engagements we were the group tries to convince the you know we were in the IT department talking to business about how we have to get common definitions and almost apologizing for talking to them of we're trying to get the relationships between data and they said you mean you're not doing this already you know they sort of expected that we were doing a lot more of this business level design in our systems and we're sort of shocked that we hadn't so we were a little embarrassed but that was a positive feedback but of course you guys would do this we would want you to have these definitions and you know I think sometimes so the how how do we have this linkage between folks that you know their day job is either selling product or managing finance or you know they may use data but that's not their main number one job in the day that's ours right so how do we communicate with those folks so in those conceptual data modeling I think in data modeling and hopefully we can get over this soon I'm guilty about myself we almost feel like we have to defend ourselves for why we're doing this but it is at the core of so many things so a big driver in the growth of data and we are seeing growth there's things like business intelligence and analytics and here's an example of I think why a conceptual model is important so say you think of the traditional business user in the corner saying can you show me all the customers by region I think we all many of us in the calls have understand it a typical way we get this at least historically for data ware housing you kind of have your source systems and a data model helps there with getting kind of the data structures and how it's stored and structured you might put that in the dimensional data warehouse and kind of get your stars came in in fact we talked about that last month and that's it can be businessy but it's sort of on the on the technical side so we saw it we decided in this hypothetical situation we're going to start with a conceptual data model because we saw it on a state diversity webinar and we are all gone hell about conceptual data modeling so we started off with just a very very simple model with the basics what do we mean by customer right and we spent a lot of time on this definition because if anyone who builds definitions knows that there's an art to this and a bit of a science as well it's not just a customer's a customer or customer you know the very simple definitions a product ID is an ID of a product we've all seen those that aren't very helpful so we guys are one thought it's a person or an organization you know we can do B2B or B2C they've purchased one of at least one of our products could be more they have an active account that could be important maybe they're a product but they've dropped maintenance or you know something simple like that so we spent a lot of time on this and we were wearing it well you know are people going to think this is a little too simple is this too academic and they did unfortunately this team didn't really understand you know they thought we were okay you academics up there in your ivory tower you can go debate definition of customer duh that's so obvious we're going to go do the hard stuff we're going to you know build all that warehouse stuff we talked about in the previous slide the day reverse engineered and created physical data model for each system so that was great and they created the ETL scripts and migrated to the warehouse and then you know one of the key things I was stressed and all my whenever I can is you know the business value of this we don't do data just for philosophical reasons it's generally for some business value so often it's hard to kind of get ROI from this so we were actually saying you know that bottom bullet we can actually send out a welcome email and give people a coupon and say we can actually show ROI that we built this reporting warehouse and these are the actual results of all the people to purchase something as a result so I was pretty clever for doing that we have a location where perfectly the script for font you know we've done all the physical stuff great because we did have the models behind that well until we showed it to the sponsor and of course you know the business folks they know their customers more than anything else you know got feel of nothing else and she said you know we can't have 2,000 customers in this I know just got feel we have about 400 and Jones Tire they were actually evaluating our product for a 10% global discount and some sales are going to find you your house and break your kneecaps because you just ruined their commission and the main thing is you spent all this money in it to rebuild the port and the data was bad and that's always me in the front the one getting the heat and the guy in the back is like nope not me not me anyway what did go wrong right we did all the right stuff technically by the textbook well typical what do you mean by customer and before anyone laughs or think this is too simple I have worked for companies that have made this mistake and I don't want to call out a name because I'm sure it happens often unfortunately these kind of basic definitions what was happening is we had our quote customer database and our quote customer database used by sales those were actually prospects right that's an easy thing I mean most sales people say I'm going to go visit the customer today well often that's actually technically a prospect who corrects them right but you know if we're going to put in database those are actually different things so what happened this is actually a bit this result we sent a discount coupon to 1600 of the wrong people we gave upper management report with wrong figure and now we actually have to go back and fix it right so what's that saying if you don't have time to do it right you have time to do it again right so a lot of problems just kind caused by something as simple as a bad definition so this time we started again with a conceptual data model that our beautiful definition that's there so a prospect is a personal organization who doesn't own a product that's looking at it and a customer is someone who does and has an active account so again so simple business definition but very very important business results so again hopefully this very simple business definition kind of showed that this does have an actual business impact and the other thing is I talked about this a lot in the enterprise architecture webinar I gave a couple months ago this doesn't have to take forever I mean it can I mean some of these things maybe we don't agree with the definition of customer and it takes some iterations but sometimes it's an afternoon whiteboarding and just taking that first step and starting go a long way and yes I met a data repository behind it and published models is all great but I wouldn't be afraid of even starting because sometimes you can just flesh out some really simple stuff by a simple whiteboard and an afternoon or an hour with some folks so the importance of business definitions you've probably seen these cartoons and other forms before because I'm like hey if you have data modeling cartoon use them where else can you shade data modeling cartoons and this probably isn't funny unless maybe it's not funny at all but unless you're in the business right we're all done with acceptance testing and everything looks great and this new marketing application this little question what do you mean by customer well as we just showed that it's really hard to fix after fact you built the whole system and your basic definitions were wrong right so you get the basic requirements before you start and that's where a conceptual data model you know another data modeling cartoon so the other thing and I might differ with some others in the industry use the language of your audience in the model so a couple things one is display away this intuitive to the system folks they are a business person can't understand the model well you should build it in the way they do we use power point all the time right we think in pictures we tend to understand things so a lot of I've seen presentations from business users they didn't call it a data model but it sort of looked like what you know they might have even boxes with things like product and customer in lines between them they're just trying to suss out their things themselves and it becomes very much like a conceptual data model you know put it in the power point do the model and put it in power point if that's how folks want to see it use business terms and avoid excess detail which I think is a beauty if you think of that pyramid again keeping it simple and to totally show my ultimate nerdery this is one of my favorite clothes a shoe obfuscation so if you're not familiar with what that means it basically means avoid using big words to overcomplicate a simple term which is sort of I live in Boulder Colorado where there are many other nerds I guess because this is actually a bumper sticker that people have and in writing circles this is a joke that people use you know you can use 10 words but you could use two use two you don't use a big word just to make yourself sound smart and I think we do that sometimes in technology if we just use a lot of big words and show people how smart we are in technology won't they think we're great no actually I think we scare a lot of business people they don't want to talk to the techy ips either we're talking down to them or we're talking that geek stuff so you know a shoe obfuscation keep it simple stupid I guess the quick word and of that and talking your business's language so this raises the eternal question that I know keeps you up at night can and should a business person learn a data modeling notation so can I think I already talked to that yeah of course they can this should not be hard this is not brain surgery I think the beauty of a model is that it's simple the concepts that we're describing to be hard should they I would say we should build the models in a way that they don't have to but they can and I'll go through a quick you know 101 of modeling for those folks who might be business people on the call and might be familiar with modeling and or a nice way for you to describe modeling to your business stakeholders so here it is so again if you've never seen modeling this is a little primer for you but I think more importantly for the technical people on the call you should be able to explain your model to a business person in about five to 10 minutes quickly with slides like this and I think they should be able to get it people are smart and the model should be simple so in editing those are your nouns of the business right the who what where why when the who might be a self person what might be a product you know how we're invoicing people through an invoice and there's the demo dictionary definition of it but yeah again I wouldn't start there I just say this the nouns right you might even start looking through requirements documents to I mean business documents when you build your model and I often do this just draw boxes around the nouns right though they're often your business entities I keep hearing about products and this one must probably an entity right so nouns are the entities attributes are always to further describe that entity so again employee last name first name higher date yeah it's fairly basic I think most business people would understand this is the description the adjectives about the thing and even the model there at the bottom that's pretty straightforward we don't have to get into the fact that that thing on top of the primary key but if you did just call this unique identifier for this thing you know people can understand that then the attributes are around it so that should be fairly straightforward the people the next one are the verbs of the organization and if any of you have heard my full story which I'll shorten here I was probably one of the few I don't even know if they teach it in school anymore but diagramming a sentence I think I was six years old and they did the diagramming sentence which kind of you underline the verb and you circle the nouns and you put a downward line to a prepositional phrase very much like data modeling actually so I knew I was a data modeler back then I think it was the only kid in class that really got into it I still diagram sentences in a way it's like I often start with drawing a box around the nouns the one at the bottom of the department employee draw a line under the verbs that's really your relationship line right a department can contain more than one employee again you can take a lot of these business rules that are part of your organization that might exist in some documents and really kind of easily that way start turning it into a data model department can contain an employee a customer can have more than one account those are you know those are the relationships with sentences of your business and some tools or you can write scripts around the tools that can take your data model and generate these type of sentences so we just did a workshop in our practice actually I wasn't there my team in the UK did this but where we took the model and created these business rules and we had the meeting with the business stakeholders we read them the sentences we didn't necessarily well we also showed them the model but to kind of check the model it is you know does this sentence make sense to you no and the customer can only have one account a customer and account those are almost the same thing in fact we could call an account customer so that's wrong you might have more than one you know wallet we call them wallets what you call an account you all those kind of thing they get fleshed out by sometimes just these simple sentences and then cardinality that's a great big word that we can make of course that issue obfuscation we can make a really simple things and really techy and scary all that is is the how many right and I like this little picture if we all if you're using the IE notation it sort of looks if you look at that will go on the employee side the one to many one is really if you think of a kid how many they hold up their hand one is one finger the other ones you could turn it almost looks like a hand several fingers I actually and I am fun at parties let me tell you this is the type of stuff I talk about but when we did the data modeling for the business book again we were curious there's a lot of different notations some people like IDF some people like IE there's a lot of different ones I actually created a simple business rule and put it in a data model can run up by a bunch of my friends again I am fun and folks that one was a painter one was a musician one was an architect that built houses there was a finance person one was an engineer none of them were in data per se and I kind of said what do you think this means most of them looked at the one that was the notation here and can kind of get it well there's one thing and there's kind of many things and it looks like a department can have more than one employee I mean a lot of people got it without my even explaining anything they just buy the notation so I'm a big fan of this one but again let's not be too academic I just find that when I'm talking to business users this one seems to kind of make sense for folks so I tend to like information engineering which is crow's feet a lot of people call it or children's fingers what I call that it kind of looks like a crow's foot as well but that doesn't show the menu so anyway that is it I mean how long did that take right here's another one super types and sub types can sound so complicated but it really isn't it's either or or and right so here you maybe just draw an example a vehicle there was an exclusive or which is exclusive again in this notation it can be a car or a truck it cannot be both okay argue there's some cars that are some but in this particular business there are cars there are trucks and they're not the same thing you can't be a car and a truck at the same time you're one of the other or you could have an inclusive sub type where I'm a person and I can be a customer and I can be employee of that company and that might be kind of a thing to talk about oh wow if our employees are buying the product they get a discount do we not let them buy the product we got you could be a conversation right there and this is a joke I was spending time in marketing at a data modeling company and my boss the time was not a data model she had grown up in marketing pure business person wanted to know a little bit of modeling and I would I put this in a presentation and at first she's like what's that X and I just explained to them oh it made a lot of sense it just was a way to explain we're talking about campaign and we were segmenting customers and this is my way of saying we don't want prospects we do want customers that kind of thing and she liked it and actually kind of joke oh that's an exclusive subtype relationship and that was sort of a joke the rest of the time I worked for her she was a marketing person she loved to talk about exclusive subtype relationships that's probably rare you should not use that type of terminology when you're talking to a business but the point is this is a particularly easy one to understand and you know I've heard folk again I might go outside the lines when we say what's purely a conceptual model there's been an argument of you know R should super times sometimes be in the conceptual model I think so because I think it makes a lot of sense you're starting to plus I can talk business rules like kind of car and truck do the same thing we're a car manufacturer what about these hybrid vehicles could there be a third thing these are very important concepts and it's pretty easy to understand so yes I would say I'll put myself out in a limb that yes you could put a super type subtype in a conceptual model the other thing that might cause some disagreement with me but I'll stick to it is do we use the business terminology or a common thing in the industry is this idea of say a party right the beauty of the party model and that sounds like it's a good time is you know you could have I mean there's the certain things about a customer and employee or a client and the customer that are similar can we roll them off with this concept of party that's reusable that's a great idea to use but I would say when you're talking to business person unless you know I might be legal and there's a party in a dispute and they use that term party for their customers it means something then I would use it but say we had a party associated with an entity what the heck does that mean it could be a legal party is associated with a illegal entity which is a company and unless it means that it's very very vague there's a thing relates to thing you could you could summarize things so much that you just said you know things relate to things and everything's a thing and that gets crazy right so I particularly prefer to say things like a customer which is a product or employee works for a department because that's the terms business people are using and maybe there's some redundancy when you use the actual physical model maybe you want to do that differently but at this point we're trying to get the difference between a you know business terminology so I will ask the question they asked a lot of questions when I do this is a customer the same in the party is the customer the same as the client the customer the same as the product you know that's the type of stuff you're trying to flush out you could have two versions of customer on a conceptual data model with two different definitions and trying to say are they different you know this is made up of the end game but you're trying to understand okay maybe that's Europe's definition of customer in North America uses something different let's try to understand that more again you're just trying to communicate when you're talking about this conceptual data model and keep the focus in the business as I mentioned we can often get academic I mean one of the reasons we like right like this kind of thing is that kind of logic we go through but don't make it a logical exercise it should be a business exercise so if you're arguing let's think why we're arguing is it a different entity than yes we should really flush this out is it different names for the same entity maybe yes that's something we should be discussing could it be a super type subtype relationship and how you resolve that could be different ways maybe we keep the different names maybe we try to make it one name but again you should be arguing where it's a business definition not semantic sort of just theoretical we can I have to catch myself sometimes it's just sort of fun to start to think these through but at some point you're like does this matter does this matter to the business or are we just being academic so as a client a customer those are the same things what's the difference between an ingredient and a raw material we had a customer we give an example of in the state of Mawlington business book that there was an argument about that right some folks it is raw sugar ingredient into a piece of candy or is it a raw material ship from Brazil that's actually cane right so could be either one there's no answer to that the answer is how the business uses it again I mentioned we just did a modeling workshop from the our team in the UK and there was actually an argument whether water was a liquid so I mean this was an environment group that was doing the environmental testing so it made sense but I think I joke them I said I hope your business sponsor didn't walk by the room and you're arguing water is a lick you know some of these things can seem really academic so I think we should just check ourselves and say are we arguing this because it means something to the business that we need to resolve or does it really not matter if you call the customer or client the same thing just pick one and move on you know I think the move on is sometimes what we have to remember when we're doing these it should not draw out unless it needs to it should be as quick as possible to do these kind of models definitions are important so don't slack on them I think this is a lot of the part of the model that can be difficult and to think through and as we showed in the example you can have a lot of big business issues caused by just something like the ill-defined term you know what do you mean my customer would beat that one to death you know how are we defining household if you've been doing that in your business there's a lot of different ways is it family members is it people that live with the same address you see I like this stuff because it's just fun to think through all the different combinations or something as simple as how we're calculating total sales and again it doesn't mean you're you're the folks at the top defining this for everyone lots of times it's just showing it I have some companies working so particularly tool I like in the market also all of the different definitions are just the why I'm using this definition of total sales for this report and this one for a different report is why because you the different definitions and as long as you're clear about that that's fine it's not you're dictating necessarily how people should do it it's just being clear about it sometimes you need a dictate but not all the time and then any Italian speakers in the phone you'll get my little joke about API API means be telling those that was kind of cute so anyway but you know if we're in the financial industry what's an equity derivative what's one of my first big metadata projects was just listing all these financial terms what they need for the brokerage and then we built the technical stuff behind it but the hardest thing was getting the definitions right for some of these terms and making sure it was calculated in the same way so don't slack on your definitions that's a huge part of conceptual data modeling some tools that actually you can show it on the model I'm a big fan of that because then I think you know I think sometimes showing something like customer account that this okay yeah but we start to say you know customer is a person organization with an active account how's that different from a client well client has an active brokerage account no we can roll those into the same thing you know again that's what the business person would have to make that decision but often until you can't until you see them you know is a broker different from a sales person oh wait yeah you're right I think clients are different because those are the high net worth individuals and a sales person is any account under 100,000 us dollar you know something like that but unless you see the definitions it's often hard to see why you have client customers in model so anyway just a tool I like to use it could be anything you could export these out into a spreadsheet I would say start with a model but again it's however your audience wants to consume them because the big reason we're doing this is for that communication so again if a sign out in the street is going to get people to read this I would do that but I don't think you have to try that hard often I think that as the quote said earlier a lot of business people are actually relieved that someone's thinking about this as long as you gave it simple I don't have a slide for this one but it was a thing you know respectful times make the I think a lot of business people if they are none willing to look at your model might have had bad experience in the past I have been in meetings where there's a logical enterprise model with 200 entities printed on the wall would take up two walls and we say to the business person we're just going to spend an entire day going through all of this and you don't mind we're going through the cardinality and the relationship no surprise they don't want to do that again I had one customer that he would put five entities on a blotter you know when his sponsor came in he would have an hour of death he would just say five entities five minutes and you could just look at these really quickly when you just sort of do a little subject area every morning just I just expected 15 minutes your time or whatever often some of these questions don't take a long time if you can do it in small chunks and more importantly do it comes in a sense to them that's a business problem hey I know you're doing this marketing campaign can I just double check with you when you say customers you mean existing customers and prospects or just existing customers that wasn't hard there's just asking the question or if you need to model a small model like this I just wanted to know what do you mean the difference between customer and client I drew this out doesn't make sense to you something like that sometimes that's all you need it doesn't have to be a whole day workshop with you know thousands of entities on a wall because I would hate that as well and actually I would like that but I'm I'm weird and again human metadata I always say avoid that dreaded I just know I couldn't want to hear that we don't need to find customers I get that well you get that but your definition may not be somebody else's you know so a lot of this metadata in the company is an employee's pet you know the guy in the picture well part number oh that used to be called a component number those are really the same thing well he knew that than anyone else putting that in the model and making you know this tribal knowledge enterprise knowledge is a huge part of the model so it's talking to these people it's it's getting in the model having the different pieces of it it might be showing this person a small model kind of making sure excuse me that this is kind of published out the organization and my little my counseling for the day I think of it wouldn't the world be a better place if we all did conceptual data modeling because it really helps communication and wouldn't this help our daily life too so one of the weird things we do in the U.S. or some of us I never have had to do it but children have a summer vacation and parents say what this great idea we've got this lovely large country we're going to drive across in the summer we're going to see all the different sites and I think many a divorce and many a brotherly sisterly feud have been caused by being stuck in a car for weeks on end driving across the United States so wouldn't it have been better if we all said let's go on a family vacation what do you mean by vacation right so maybe the father's like you know I think this is a great opportunity to learn new things and I want to step in every state park in every state and learn a new fact in each one and mom goes you know I really have been working really hard I just want to read a book so you can go to your stupid state park I'm going to sit in the car and read a book and Jane has been studying at university for a while it's because dad you can stay in your stupid park and mom read your book I'm going to go out and take a run because I have been studying for weeks and I just want to get outside in the state park I'm not going to go look at the interactive exhibit dad that's so stupid and Bobby didn't want to go at all he's like for me vacation staying home being with my friend so could I just skip this whole thing altogether and Donna she's like well as long as I have my laptop and I can be building data models in the car then I'm good all right and then Ian the Brett he's like you Americans I call it a holiday you don't even know the right word right so all of this conflict just from something as simple as what should have been fun the vacation so if you think of it if we often define our terms right let's let's go to a party and if I wouldn't use my party maybe you could be annoying but I think you can see that sometimes just getting basic definitions of things before you start a project you can avoid issues so feel free to data more with your family if that really helped me hopefully an illustrative example so the other part of we've been talking about a bit is that we are visual creatures and that's other beauty I meant to you can put some of this in a spreadsheet that's fine but the beauty of a data model is that it is graphical and we tend to as humans think in pictures in fact this is an actual cave drawing found in southern France no just kidding it was actually an IDF notation that one bad joke but we do we tend to draw we try to think in pictures we draw in pictures so that's one of the beauties of a conceptual data model is that it is very graphical and pictorial so one of the things I do in most of my consulting projects and it's something that most people sending my clients around the phone they may giggle because most people do the first time until they look at it and then they're like oh this makes a lot of sense so I often just put literal pictures of things in a model right and this has been very helpful with several clients so you know I'm talking about a sales person that sells a product product in a box and then our products are all online games you know we wouldn't actually have a box for that anymore we used to sell them in stores and now it's all online so there's no ball okay well there's a you know thing right there and sales person support okay there's four reps on the phone but you know sales people are actually on the phone too we don't have any you know so I've had several customers where you know things came out of the model just from drawing it out this way the other thing it sort of makes it in fact it sort of makes a drawing of what the business is so several of my clients I have one I did some work for an outdoor industry and it was Stefan Krauss and he worked in St. Moritz and he was a skein instructor and he purchased products so he had the thing all around Stefan there was also Stefan Krauss who was an accountant in Zirc he had certain characteristics from a certain salary and because he was in the earth he was subject to GDPR you know all the sort of stories around that customer especially when you're trying to explain the connections between data or why it's important I think it really starts to click with the business person the customer down the right was he was Stefan I know he was Martin Steich he was a high net worth individual who had different accounts in different countries and he had yacht insurance and all these sort of things and it's funny during the product people start to say well Stefan would he's actually started to relate to it but that's the purpose right the client I'm at today is a healthcare provider and they're complaining with the picture we had the guy was too healthy Michael how many hospitals actually the people actually is very attractive to the band-aid on right you don't actually have sick people but the story was around you with the hospital visits and the different claims and all that so it really is kind of a nice one pager at a pictorial level I've had different aha moments with the different business people one with a lack of describe our customer you know when they're younger or they're older or so you can laugh but that's the feedback they had or you can say wow we haven't got them engaged they started to see their business in this right or another business person said oh so I get it the hard part of the warehouse is the relationships between those things right that we might have been missing a relationship yes there's no connection between a sales person and product we need to add that or you know it sort of makes it very real because you're literally seeing your business and the reason I started doing these is am a goofball and it helped me when I go into the customer a new customer a new business may have not worked with it helps me tell the story of what is this data doing why are we doing it how does it flow and I said well I will show it to other people and there's my tip for you I think it really does work because it kind of makes it very concrete for the business because we do tell stories not only pictorial but everything is a story right we can't even sleep without dreaming which is really stories right and so I think that the impact of this is what I'm trying to say is that no one cares what your data model you might have heard me say this before and it's sort of sad but they do care about the result they don't not they don't care about data modeling they want to see the real world impact what's the story why are we doing this what are the results of that model what we're trying to say is that we can't link sales to sales person because there's no relationship there or if we could think of all the great things we could do and I think sometimes we miss the so what when we do modeling you can stay too much in the database level so you know not saying you necessarily read data models to your children before they go to the point is it's not the model it's the message and I had to remind myself in fact many of my training classes I kind of say that I'm sorry but nobody cares about your model and I have to work with the customer and we do we get into what we're doing it could be anything it could be a thing you're writing an article you're writing it could be your own taxes you think you're in the middle of something you're in it and the customer you know wasn't connecting that day I think they probably had a big meeting and I had to joke to myself Donna nobody cares what your data model you know the story you were trying to tell so you know I tease myself too it's like these are things you have to remember it's why are you showing this to somebody what's the scenario another you know you might have seen my guy in the lower left but somebody encouraged me that they said they actually liked it so you're going to see him again but it reminds me this is me you saw my picture in the beginning but inside on that little guy in the lower left and so my point of the slide is that there are different personalities and different goals in your organization everybody's that guy in the middle what's in it for me right but I think there's also different personalities so we as data architects we probably went into data architecture because we're focused on things like architecture and technology and often we're sort of hired to find problems right we prime find problems and we fix them and that can often make us seem very negative right so business people they're very results or oriented a self person almost by definition is just genetically positive you know the customer said don't ask again you know and it's all about opportunity and business growth so sometimes where we clash and we can be seen as really nerdy that might be surprising so you might be wondering this little with this little guy I'll explain it to her often seen the kind of weirdo in the side of the street holding up the sign going the world is going to end if the model isn't third normal form you know we might be right maybe the world will end if our model isn't third normal form but the business person doesn't care you know why should I care that it's in third normal form tell me the what so what so that's my recommendation on the right can we be the same person on the left in time but put on the hat we all wear different hats of more of a data advisor less architect and more advisor so think of the opportunities less hey guys we can't do this with a model or this is broken or this is really hardest hey if we could link customer dating with product think of the stuff we could do so what are we doing different on the right when we're talking to the business person's language why what do they want to do that they want to get more money and want to increase sales so we could do this we're being opportunities not problems it could also be a product problem we can't link customer with product we're trying to get funding to do that but we're saying it in a positive way think all the great stuff we could do if because that's often how really successful business people live and work that's why they're successful they're the next thing and we can get annoyed by them because sometimes they forget the details and that's what we clean up after talking to them that's really what they're thinking of and that's really what I think is fun about data management why I'm still in the business because especially with data now there are so many cool things we can do so putting on that hat especially when you're speaking to the business executive that's where you become data advisor unless it architect with the weirdo with the sign in Birkenstock which we all are part of probably all of those right so just think of your audience I guess it's a message there and we do the same thing right as I mentioned before do we really we're all into our model do we care about the details of other people's jobs so think of an accountant right so we talk to the accountant we knock on his door and he's like you know we recently switched to a pool based accounting based on cast based accounting because and you're like I really just want my paycheck do I need to know all the details of the accounting system I just want my paycheck so again business people are the same thing with the data model well why would I care I just want the data for this report or I just want to make more money for sales how can I do that how can data help me so you know it's not that folks are bad people that people don't like data I think actually a lot of people are interested he just really don't care about the details of what we do because we do the same thing I give any talk you know I have a and what Christian come to fix my house I really just want the light switch to go on I don't know the deep you know sometimes you are you kind of want to know what they're doing but generally you just want the results you don't care about the details so again if you've been in any of my workshops we often kind of have this as one of the exercises and I think you know driving home think of this or if you're at work or wherever over dinner think of this how would you describe your project to the CEO in two minutes and this actually happened to be once in the U.S. we always had with the elevator pitch right you're riding up the elevator with the CEO and they ask you what you're working on or you have a sales prospect in the elevator what would you say to them in two minutes or how fast your elevator is to sell them so you probably don't want to do in the left is you know I'm working on a project to rationalize that data across data so you've lost them right that's really not very interesting to the folks but if you can put it in their context one working on that big online marketing campaign you're doing we're going to get a better customer list so that you can have better content targeting you know whatever it is you're working on think of it the context targeting you know my story this actually happened to be I was working on the customer one of my first program jobs way back in the day when I was a software engineer and I was riding up the elevator with the CEO and he did say almost literally just this so what are you working on and I told him about the program he said what does that program do and why and I couldn't answer it so it was partly a manager's fault for never filling us in with the big picture but it was probably my fault for not thinking of the big picture so focused on getting that application to run I knew on my little piece did but I really couldn't tell the big story so that was embarrassing enough that that never happened again and hopefully you think of it before that may ever happen to you or you know in anything it's not just riding at the elevator could be anything when you before you get your presentation before you do anything what's the so what and why do you care and why would some people care so that's basically yeah when we're thinking of this you know a lot more we're going to talk about but again in the spirit of keeping it simple with conceptual models we're focusing on the business business stakeholders business rules so focus on what is interesting to them have your presentation suit the audience keep it simple and think of the story right well it's your elevator pitch and don't make it overly complicated so before we over the questions to a little bit if you did have a question and it doesn't get answered today or you just want to say hi I'm here's my email I'm also on Twitter as well as our company and my personal website a little bit about my company global data strategy you do this for a living so if you want help let me know there is a metadata management course online in addition if you're at edw in Atlanta well I'll do the whole half day metadata session but if you just want the nuggets online in your pajamas you can watch this and just if you can join as a friend of the others there's a whole lineup for the rest of the year if you are interested so without further ado Shannon we can open up the questions Donna thank you so much for another fantastic presentation we had a lot of great questions coming in to answer the most commonly asked question just a reminder I will be sending out a follow up email by end of day Monday with links to the slides the recording and anything else requested throughout the webinar we did have a request for the report the metadata reports that you were showing earlier so I'll get that out to everybody as well so just diving in here Donna to the to the questions you know this can't this question came on early you may have addressed it a little bit but have you seen conceptual data models done for the bi later yes and in fact I had a slide I took out and that is something I've often heard why isn't bi your star schema isn't that just a physical thing and I had a slide that actually looked like a star and I think when you think of a star scheme I'm reporting on sales by region by customer all those things you're reporting by I mean I often just do a very high level model that is the star scheme and make sure do I understand just what those dimensions mean do I understand how we're summarizing so I think yes and I'm not sure why that is so often fought with me that well no no no of course it's a physical you're building a warehouse but if you don't get I would think even more importantly with bi if you don't get those core definitions then your reports aren't right so I and try to just understand with the business what are we reporting by and I have just kind of a very simple you know might be a you know the entity in the middle and some stars around it and then that kind of makes sense this is the main thing we're reporting on this is the slices of the dimensions we're reporting by so yeah I think it makes a lot of sense for you so as I know Dr. Peter Chen the inventor of ER model diagram considers ER model as conceptual and wished it to be used by the business people from the very beginning but with very what has been changed in since Peter's time and approach well I mean I think I would agree with them that you know there are business people that can understand a pure data model I think I'm just seeing a trend in the business in general I mean I had to kind of switch my thinking you know often it's been very prescriptive you know you should have designed and then you implement and you build and I almost think that everybody is like Wikipedia and Wikipedia you know this whole idea of crowd sourcing so I guess I've just been more creative in my approach and so you know some of these things like pictures I think you know I've sort of seen even this metadata repositories where there's comments and threads and almost like a Slack channel for metadata so I think I think some of the core concepts are the same he's done a lot for the industry but I think and I have to get my brain around it often and say you know this is a good thing it's not as structured might be more agile but I think so some of these creative maybe it is a Slack channel for metadata maybe it is a picture that just gets the idea across and maybe you put it in the model later or maybe it's iterations of the model I think the smaller bite size chunks and the ADD world with small amounts of time happens so I think the core concepts are the same I think we've gotten a little more creative on the front end for some of this when we talk about the conceptual model my two cents on that one sure so one of the challenges I find in building data models is showing the relationships in a way that is simple and palatable for business users do you have any opinions or recommendations on software options for developing data models the relationships I have two ways I show those let's see if I can go back to my screen one answer I think the relationship well one answer as I'm looking for my slide is I would I show only the relationship that makes sense I mean if you're doing a bigger model sometimes it is those relationship lines that get really gnarly the lack of a better word very quickly so that's one thing and there's many of the data modeling tools in the market I never say the name of any tool so stop asking me if I had people that have nothing you did help me get that question I will not answer that one you can hide and show on the same model different relationship lines but if we think of the one that I had that was just the gosh I'm losing my mind to sleep in the day just this kind of one that the story kind of one I will do this in a PowerPoint and just if I'm really just breaking down the core concepts I've been doing PowerPoint just so PowerPoint line and this almost looks more like a data flow and it's not typically you're one of any relationships but it depends what you're trying to show if you're just showing basic these things relate to these things and the connections are hard I show that if I'm showing relationships I do it more in the way that we showed with the IE and I will show crow's feet because I think that doesn't make a lot of sense I think very quickly even on this one of what I like about that is that the customer can have one or more employees and he starts showing the cat and cat cardinality and I often hide the detail you'll see here in this modeling tool is just just the name and you know it kind of simplifies it so I understand a crow's feet notation and most of the data modeling tools have that so it's not necessarily always the tool it's either PowerPoint if you're doing really simply or in the modeling tool use the notation but just simplify it don't show all the attributes and all of the relationships just a few of them that makes sense some longer sure yeah so when building an enterprise conceptual data model should you capture all attributes key attributes or no attributes oh good question who answered that one up so that was another I would say here's a great consultant answer it depends no but I'll say what depends up I think if you're just trying to get this main I'm trying to get to the good department of one one employee then yeah I think sometimes hiding that detail is good for an example sometimes if I like to show attributes if they're not overwhelming probably not every single attribute for example I'm talking about customers and I show gender code no we only sell the corporations that wouldn't have a gender so I think if the attributes show help clarify the meaning I definitely show them well what was the department well we have the department had you know these example might not be good because it might be odd but we're not obvious oh my hero the example I had with a car or a truck well what are the attributes of trucks that would be different from the attributes of car and that might help answer if they're different so if it's helpful I add them I probably would say you don't unless there's only a small amount of attributes show all of them I would say probably never show because that's when you're starting getting the logical and it gets overwhelming so it's either none or that small subset that kind of adds clarity to your discussion of these are the things that seem to be different or this is what makes up a customer that's different than a client or something like that I hope indeed so you know how do we incorporate this data modeling into the agile methodology I think it is perfect for agile methodology partly when you're doing a model right it is a small subset and often when you think of agile or a sprint or a requirement space it's getting that requirements from the business and turning them into stories so this would be at the very beginning just making sure we're getting the right requirements is a huge part of it and the beauty of these conceptual they should be iterative you should change them as I mentioned before but arguing are you using that thing that's exactly the point the point is communication so it should change around then at the beginning of a sprint or during a sprint you can use this to help kind of vet out the data aspect of it and it should change and it should generate discussion and kind of turn into the story it's the context band the story you're doing in your sprint so I think it's perfect for because it's so quick it isn't like you have to develop a whole physical or logical model you can just start to do some of the conceptual stuff is that out just the pieces that make sense or get clarity I am all we draw a model and start to have questions do you mean X or Y or Z and often that model can help be clarity generate clarity is a lot faster all right I think we have time to sneak in one more question here do you have recommendations considerations for the taxonomy standards selected for building a conceptual model no I mean I think I mean taxonomies can mean different things for people some people like it kind of super type sub types I would say whatever makes sense often taxonomies are good outside of a model sometimes kind of just showing a hard article approach and kind of a list a structured list helps so I think whatever kind of makes sense for the business I kind of sometimes like this approach kind of a hierarchy kind of taxonomy I think I think also sometimes everyone would be really different about semi-sectonomies which I'm being vague in my answer the other thing that often comes up is if I have different meanings of a thing customer could be a client you know I often will just show kind of the out of the line relationships to those rather than try to think oftentimes people try to force everything to be the same or it might just be a taxonomy or hierarchy so I think sometimes the conceptual model can help flush that out well Donna thank you so much for another fantastic presentation and thank you thanks to our attendees for being so engaged in everything we do we just love the questions that are coming in but unfortunately that's all we have time for today just a reminder I will send a follow-up email by end of day Monday with links to the slides the recording of the session and also include the research report that we worked with on with Donna that she talked about on metadata and thanks to idea for sponsoring today's webinar and helping us make it all happen we appreciate it and I hope everyone has a fantastic day thanks Donna thank you