Industry sign banking new york permission slip now
all right good thanks so again welcome everyone it's nice to see a packed room or something that has been a exciting journey for myself as well as those at Deutsche Bank when it comes to tableaus disruption which is not a bad thing it's a good thing it's something that actually impacted the way that we did business intelligence bank alright we're back so again today the the topic is on the tableau Center of practice it's another fancy word for a center of excellence one that we adopted because we know practice makes perfect so that's something we're going to be working with and talking with all of you and hopefully it'll be enjoyable for you I'll be joined on stage where or will be Co stayed Co staging this with arm and rahzar so Armen manages or co-head of our structured financing and I'm the tableau Center practice manager as well as I'm a bi product manager and I'll talk a little bit about that when I come up here in just a moment but at this moment I want to hand over to Armen is he's going to talk a little bit about how tableau came into Deutsche Bank how it impacted his area and his area specifically when it comes to finance and exactly what were the challenges that they faced which would then introduce why the tableau Center practice was so needed so out your no retirement thanks al ok so I think what I wanted to do here is to tell a little story about what happened to Deutsche Bank you know we were I think like a lot of people in this room okay we okay thank you so we had problems like that the beginning of implementing tableau might be lots of feedback negative feedback so just telling a story and I'll just I'm just gonna talk a little bit about kind of what we done and just a little bit of the background I come I run a big business to deutsche bank you know 350 people and you know work in the front office I have traders that work for me I have loan originators I have people that basically originate business for the bank and so I came to Joy Bank about five years ago coming from another Wall Street organization goldman sachs and when i did that i inherited like many of us when you work at a big company i inherited data that was you know old stale broken etc and i was working in a risk business inherited one of the largest risk businesses at Deutsche Bank had when I when I first joined and I was really struggling with how to look and see our data really kind of a timely basis in a way that was comprehensive to be able to think about how we manage our risks and how we manage our people around those risks and the real challenge we had was that we sat in many geographical different locations we sat in variety of booking systems etc etc and it just really didn't work together so we ended up coming up with a plan as to you know how are we gonna fix this and it was really an immediate need so at first we went to kind of our traditional IT department this is you know three or three and a half four years ago and said you know can you help us and the answer was we can build something for you but it's gonna take a long time right and anytime you want to change something it's gonna take a long time so I didn't have the time for that I needed to really deliver results starting a new job etc and seeing really a problem with the way the bank look at its in manage its risk and so we reached out to one of our you know one of our consultants who recommended tableau and tableau is really our thick to sit on a variety of different systems etc and so we ended up really and this is this is I she came to the tableau conference three and a half years ago or three years ago and really sat in one of these sessions and listened to a bunch of people from a variety of companies talk about how they got tableau to work and I got the idea from that conference which was to accelerate the anything cars which was we couldn't get the IT department to really do anything right so we ended up doing and you know it sounds like there's some familiar faces to that and you know grins but I think what we ended up doing is we tried to do it on our own so we did it from the businesses perspective we brought tableau on our own servers and you know we got we did it from a regulatory and compliance perspective correctly but we didn't really have the support of IT when we first did it and there was a lot of challenges and push backs because people were really concerned around you know there's jobs associated with these these non self-service business intelligence tools and it ended up being really really challenging for us to deliver so we took the kind of what I like to call the Trojan horse method and the other one there there you go we ended up really it was sort of you know the Trojan horse grassroots is kind of how we thought about it but we really connected to all of the source systems that sit with an organization and then worked with all of the folks that sit in the organization from legal and compliance risk technology operations our senior management team and our core vision team of really three or four developers that were developing workbooks for for our needs but I think the one the one key component around this is and I think a lot of people struggle with this a lot of folks that are in this room are new to tableau but some aren't but we have currently which we'll see in our next slide we have now I think we're over 8,000 users and it's rapidly growing so we think we're going to be a 20,000 and you know the next year and a half and it just what I wanted to highlight is that there's two ways to approach it we have a lot of clients that use tableau and we ask them the questions and we hum of these conferences we asked the same questions which is you know what is the best way to get your data fixed and tableau being our best solution for it which is do you take the corporate board push down from the top this is my data strategy or do you do this grassroots in certain areas that need it and then push up and in some cases you do both in my organization the senior management didn't necessarily get the data needs they didn't understand that you could take something like tableau sitting across many many systems and create powerful powerful sets of information for people to save money to bake bake make better decisions etc and so we had to do it from a grassroots perspective now since we've developed it and we have a center of practice which Sal's going to talk about in details is there's a a senior management view that from the top down is this is how we need to architecture our data and so what I wanted to make sure that people are aware is that both ways work we've seen it working in organizations in both ways but from our perspective we started from the bottoms up grassroots Trojan horse and it was successful we ended up delivering you know lots of growth and lots of value because of this grassroots initiative so we took it immediately and started sending out kind of workbooks and we use a very kind of analogue way to do that in the beginning I think a lot of people are challenged when they start using self-service business intelligence to say when do I send out the report we took the view that especially in our world where markets move all the time that the data is never static and is never going to be a hundred percent accurate and if you take that assumption set then you don't go and try to fix all the data a hundred percent before you deliver a report to somebody we took the view that we were like if we can get data to be 75 to 85% accurate and again it was you know you can't exactly tell then we'd start pushing at reports to senior management we push out reports with the views and see see all the people that are responsible for the data the owners of the data and literally send out like a PDF or an HTML file with a report that would deliver something that was important to senior management and you be amazed how much faster that fixes the data and having somebody sitting and like a room somewhere saying I don't really understand this data let me go fix it first and that was a very very powerful tool that got our data accuracy to close to nineteen ninety-eight ninety-nine percent for almost every report that we'd send out and you know because he would call it like that doesn't look right right I know how to fix it it's wrong we'll fix the source system so I think that's a really important you know piece of the puzzle for us was really and we take that thesis now every time we connect to new data source we try to scrub it to make sure there aren't wild anomalies but we don't think we can get perfect I think that is not necessary that every organization does that and again we make sure that we label the reports to be very clear that these aren't you know actual reports they're not something you should rely major decisions on but ultimately you do end up making decisions based on the reports that you see and so I think this this slide is showing us is that you know we obviously were able to quickly deploy it we're able to connect to multiple data sources and then we had immediate savings from a very low cost you know footprint I think the other you know key point to this is and this is just a philosophical one which is there's this whole concept of this crystal clear data Lake right everyone has this idea let's have a data link I don't in our organization and a lot of organizations I don't think crystal clear de lakes exist right we sit on top of like dirty puddles and messy streams right and swamps right of data and we connect on top of them with tableau and so we then tried to cleanse some of these muddy waters and settle some of the you know in efficiencies or the inaccuracies associated with the mud but we don't ever try to we don't try we have taken many attempts to like cleanse data make this perfect data Lake and it takes forever right it's just never we never get there and so we've taken the view goes back to kind of 85 percent sort of thesis is that we are going to have data that's not accurate but the only way to really you know cleanse the water is to use to expose it so you know this is sort of talking about our grassroots initiative and you know when you're gonna hear Sal is really kind of development that's happened since we started this is a grassroots initiative but you know we started you know over four years ago three or four years ago really with nothing right we had a few users in my and my team that were you know I asked them to step out of there these are people that have you know run regular businesses or have regular jobs to do on their day to day basis to become experts of the data and to build and work with our our technology teams and our consultants that helped us build these workbooks to really develop the first set of reports that we had as an organization again we're at seven thousand four hundred and sixty seven users today again I think that number will probably double over the next year year and a half because at this point and this is why you're gonna hear from Sal is it became an organism that we couldn't control from my business right I was elder always helping folks to try to socialize you know the success we've had and the benefits we got from tableau but it became just unwieldy for us to manage from one part of the business you know Deutsche Bank has 90,000 employees and so you know my my team of 350 you know couldn't couldn't manage you know adapting to that so you know I think you're gonna see us develop these tools across all geographies and you know the the challenges around geographical become really really important right you have latency issues you have you know is that person awake who supports the business and from you know from New York to Singapore to answer questions the other really important point beforehand another cell is that we really are now like one of our biggest challenges and we've had this rapid growth but one of the biggest challenges we've had now is really the plethora of workbooks that just exist I can't tell you how many sales reports if I search for sales reports I have on my screens I can't find what I'm looking for and so we've been asking we you know I'm part of the the Financial Service Executive Council is prototype oh we meet once every every quarter in New York with all these kind of financial firms and we all have the same challenges I'm sure it's not just the financial services firms so we've been asking tableau to help us to develop you know sort of a landing page right that has thumbnails of things that can be forced to you as reports that you need to see but also you can create your own favorites except we've kind of created our own version of that and it's kind of a hack to be honest but it's it works because then I can look at a screen pull up a you know a single web page every single day and know these are the reports that I can see and they can push to me what the little most latest the latest version is so that's been something that's been challenging for us and we think it's a challenge for others at least in our in our industry and but it's something that you know you'll see is it's hard to track the right report or what report you want to have and it becomes you know you create an end up creating duplicates which is inefficient so the way we've tried to tackle that as well is we've created sort of a social interaction we don't have a you know a tableau conference but we have town halls effectively where we have a sharing of ideas and we have a lot of workshops which sales team runs which we talk about the types of reports you go not give out anonymize data reports let's say here's the sales report that this team created and here's the HR report this team created so we can try to create the least amount of duplication and the max amount is kind of social sharing of you know really successful reports and so I think those are the one ways we've thought about it we're we're still in our nascent stages and are all excited about you know the artificial intelligence piece of this of what tableau is thinking about and you know natural learning language learning which is going to be super useful kind of on a go-forward basis but I think that we we really believe that the way that this tableau is happening so quickly is similar to what happened you know in the 50s or the 60s right I mean you used to take a piece of paper and you would send it down a chute and then people in a word processing office would type of a piece of paper and then send it in a memo interoffice mail to somebody else now we all have word on our desktops now we all have Excel on our desktops to manage numbers we think that self-service and below is going to be that way where you will eventually have everybody will know how to use these tools and so our thesis is to try to push it to users end-users as aggressively and quickly as possible the technology teams are there to support and both you know regulator from our perspective a regulatory perspective compliance perspective and you know and you know to make sure that we have a fit in the most efficient form in terms of the way that we deliver our technology and so it's really kind of a business driven process from us it's been driven by myself have been a champion of tableau at the bank and I'm happy to see pass it off to Desalle where it's become too big and then pass off to him on this on this conversation right Thank You Armen so as you heard the business at this point in time from the 2016 prior to that was taking it on themselves to to push to monopolize tableau to socialize it among their teams but again as you heard Armen say sad it grew larger than they anticipated so they needed some support and full disclosure - Armen I don't know if he knew this but when when you talked about IT pushing back I was I was the one not wanting tableau in but there's a reason for that there's a reason yeah yeah so as the product manager I'm responsible for ensuring that we do not have too many products of the same kind to redundancy cost and so forth and we already had for bi products you know we had Cognos s ap QlikView you know spot fire out there so they're all in there and then I heard this tableau and I'm like come on why do we need another one what I wasn't prepared for was the social as
ect the the visibility that tableau gave to the business this whole self-service culture that I was preaching but we couldn't really get there with the other BI tools was available with tableau when I saw it firsthand in fact we did a contest between some developers and Cognos and tableau and the tableau wasn't a developer just a business general business person with a data set just blew it out of the water and I knew right then that I could not push back any further thus we took on this task so you're gonna hear a lot about this and you've probably heard this if you were at the Charles Schwab one other cos you have to plan so for us we already had a very mature IT platform managing other BI platforms but we needed to change our thinking about how we were going to support our men's group and other groups who wanted to be able to take tableau be efficient be consistent and to do so I had to go back to the drawing board but before that we had to take a look at what the landscape looked like across Joyce shabang in GCT in other areas I needed to actually talk and find the stakeholders identify what they really needed from this product to support their businesses this was some really cold calls I would talk to them on the phone I would ask very you know deep questions about their strategic paths and so forth and then what I found out kind of like what you see from the picture I was just watering one at a time I kind of saw that there was no way that that was going to grow I needed a larger irrigation system right I needed a community to be able to support growth there was no way we're gonna be able to do it just one-off so this is my brain dump here this is me locked in a room on a white board and we have plenty of these white boards in our office where I just started to put everything I could possibly think good bad and the ugly but it came to business intelligence self-service you'll see some really keywords that are gonna pop out and they're gonna pop out because later on I show you some of the final product here but things like governance training we had you know something in that right left corner they're business orientated professionals we went through different iterations those are really your tableau champions those who are embedded in the business who become tableau ambassadors and specialists we needed to identify those as well but as you can see from this quote you try to get as many good ideas get rid of the bad and you're hopefully left with something that can be very useful so I'm gonna walk through a couple of the prep steps that each one of you can probably take for your individual businesses these are our steps that in it go in somewhat sequential order they don't necessarily have to but one thing you do need to identify is what we had to do is figure out what personality you wanted to be where you're going to be heavily on the IT where you're just managing the administration or are you gonna get your hands dirty where you're going to actually impact change in the business by giving best practices and governances be a support team that's what we had to do and then from there we needed to learn from those who were already successful teams not just generally within your own business so last year when I was trying to formulate all of this I heard Jessica Briggs Sullivan from Charles Schwab you may have heard her she's done a couple of these I listened in because I couldn't come to the conference and I was just impressed with how she put together a plan of action around creating a center of excellence so I called her up I LinkedIn contacted her she spent some time with me and my director on exactly some of the steps and lessons learned that would help me so again I pass it on to any of you and those who may watch this later on that it's good to reach out to your peers and find out what what they're doing so as to get a better frame of mind you need to build a plan a charter a mission statement something that will brand you and your team that will also help generate excitement about what your practice is you also need buying it from senior management you need yourself an arm in it okay you got to find yourself an arm in out there one who not only at the senior leadership but who's also passionate about intelligence in the business insights growing being strategic and in this particular case using tableau to be able to do all of that and that really helps generate even more effectiveness in building your your center of practice the next thing we needed to do is locate your partner locate and partner your business friendlies these are those that are inside of the business both at senior levels or in different parts of your business for us we went to look in HR we want to in the risk teams we wanted to look in the sails and identify people who would want to partner with us as we push this forward a good group that we found was our agile ambassadors many groups are developing these agile methodologies and teaching safe and agile practices within their business they are great partners to pick with to work with self service like tableau because agile and tableau really work hand-in-hand then we wanted to begin the relationship with tableau and this is where I found a huge advancement to building my vision of the center of practice other vendors I've worked with weren't as embedded with their their customers as tableau is so I was able to get a strong relationship with individuals like Jeremy Blaney Peter Varga Miranda Oscar all these are ones who are working directly with us they helped us to kind of identify customer insights that we would not have found on our own and they were instrumental in getting us where we wanted to be faster and then one thing that seems to get underlooked is the fact that when you do start a practice like this like in 19 we did not have tableau experts we had BI specialists we not have tableau experts or even people who even used tableau so the transition there was a little bit rough especially since I was on the on the clock for getting this stood up to support our men and their and their teams so we use our bi specialists but we complemented it with a strong external presence in this case was a consulting company with m2 who helped us to be able to form a lot of different things we needed such as in training modules and just being there for things like doctor sessions and workshops just to help us get started and to move forward so the tableau Center practice so what I'm going to show you now is as you saw previously that whole brain dump there was a couple of key things that came out of it that we built on some pillars but one thing that I was really passionate about is I wanted a brand I wanted something that I could put all over communications I wanted something that stood out and so what better does that and then a logo right we all love logos right so we had a sort of a small little logo contest and this was the winner mixing ours with tableau we didn't infringe they loved it so but around this we built the pillars so three distinct pillars that support our foundation of the tableau Center practice as you saw we have things like onboarding and consultation training and communities pour in enablement and branching off of these are some of the key things you probably do already or want to do now what I'm going to also show you in this particular slide is I'm going to fast forward to some of the milestones I'm going to give you a sneak peak of the milestones first and then we'll reverse engineer a little bit about the will break down these to give you some insight about what what we're doing with each one of these pillars so first off since we started this in March of this year we've had a hundred and forty engagement requests and just to put that into context an engagement is one of our customers one of our internal customers coming to us who may represent five 510,000 but they're coming to us to ask us they want to get to under tableau we need to help them and we go through a whole process of listening and then providing some guidance so that turns out to be we had one team every two days coming to our my small little intimate team we had to create a communication channel in order for people to learn and to share and so we used our own internal intranet site and I'm going to be showing that in just a moment we've had 15,000 views since March of people accessing these views to be able to get help we've had over 800 attend webinars sponsored both by tableau as well as our own internal champions providing insight to our groups these are also attending our tableau user group forums these tog sessions are also extremely helpful in not only socializing but collaborating across regions and cross business lines then we've had some very special events happen over this year just by a show of hands I'm just curious how many have actually run a tableau day or been a part of a tableau day in their companies oh that's excellent so for those of you who have and maybe those who you don't tableau days are just small little tableau conferences right they run the very similar type of theme and we were very very excited about running one but as you see here we actually ran three this year and we ran one in New York City where me and Armen helped out with that and then we did one in Singapore and one in Frankfurt over 500 employees attended in general across those and there was some serious excitement caming out coming out of these teams and the businesses because they never had that they didn't see this coming to no other business intelligence was coming in front of them talking about how they can be how they can improve and then our training this is something close to my heart so while we've had the surge of people wanting to learn or get tableau they needed learning it's a self-service tool but make no mistake there is learning a learning curve that has to be applied and the cost of every employee or everyone who wanted to go out and purchase a training module wasn't cost-effective so we had to learn how to help them internally and give them opportunities to get right on their tableau journey quickly so let's take a look at some of these these pillars I'm breaking down for you a little bit so the onboarding and consultation we have a know your customer engagement this is a 30-minute call that we do with every customer that wants to on board to our tableau platform whether they're doing a desktop or a server they get they go through this 30 minutes and we basically want to know what they're using want to use tableau for and in some cases tableau is not the right tool for them and that is also helpful as we identify and then later on we may bring them back if they're if their situation requirements change we also provide a quality gate onboarding protocol to ensure there's some governance principles applied they understand what the process is for moving from a development or sandbox into a production environment we do roadmap discussions mostly those product roadmaps are done with our senior teams they help to understand where the strategic path is for tableau within Deutsche Bank we do live demos and then we periodically also do these health checks where we actually listen in to what groups have done and help them to get you know steered in the right direction or maybe they're just doing really good and then we use that to help other teams who aren't our next bubble is training so we with the help of a - as I mentioned earlier one of our consulting companies we identified these computer-based training modules that we worked and partnered with them these are self-paced training modules that allow customers at their own time to be able to go through some desktop one or desktop twos kind of pseudo training modules and this has been well-received by our global community then we've had instructor driven and virtual and in-person training advanced training on topics such as extending your tableau capabilities with JavaScript or with api's then we provide offline training materials which will be stored on our intranet site and then on demand tableau doctor sessions so the doctor sessions are is an interesting one because again we're talking about close to 8000 permission users in growing calling our hotline for help with a doctor session we've been very difficult to do with just my team of me and two other people so in order for us to accomplish this we had to expand out and identify our champions in these regional locations and then we brought them in gave them a some formal training with the help of our consultants and now they're becoming the champions to address some of their own regional or business specific questions that are in these doctor sessions and then finally the community support and enablement this is where we do the collaboration events tugs tableau days this contests all for the purpose of helping people to talk to each other at the same time learning what they can do to improve their own individual workbooks and dashboards we use this opportunity to set standards and best practices provide a centralized content management that's our intranet sites and then welcome kit so the Welcome kits are really kind of pseudo community support as well as engagement because we send out these welcome kits when people are interested and it tells all about what we are what we do how they can help links to direct them back to the internet sites and some helpful hints about ways that they can improve so now I want to give you a look at our intranet site so as we talked about we needed a communication channel we needed somewhere and it if it was up to me I would have loved to get a Java developer a web developer a design thinker all in the same room to build this massive beautiful looking web application that's intuitive interactive just not there yet so what I had to use what I had and that was an intranet site so with some design thinking and with some help from members of our team we put together this site and this is the launching pad that you're looking at here this is our Charter stands our logo and then below there are links to various aspects of the pillars that we were just referencing here so quick access the TLC everyone needs TLC right and it's and this is something that our teams in our businesses across Joyce uh Bank have grabbed on because they need they love training they thrive for learning so we started to create a page that directed attention to things like self training modules libraries and video links even either back to tableau or those that we've actually brought in-house and stored on our own site this is another another page within our tableau Learning Center and what's interesting about this is that we've created this training resources page that helps people to kind of gauge when next training sessions are coming and what to expect we also have these little submit request buttons that actually send a request through a ticketing system we use to help us to keep things in queue when it talked about requests such as workshops or even these doctor sessions and when we do get these requests we identify where they're at what region they are and then we respectively push them into the individual champions groups for them to be able to contact them I also I made a promise to one of my developers this little design you see in the corner here so Luna kanungo one of my guys here he's actually going to be retiring soon so I'm very sad but on his side project he uses tableau to create art so everything you see both in that little corner and the big one was done using a mathematical mathematic forget the word he's using an algorithm to put together this particular images and he's done a bunch of these so I think he knows what he's gonna be doing when he retires he's gonna open up his own web shop and sell these maybe but but yes so training is extremely important and what something we've done very seriously in fact one of the new things we're going to be doing here is creating this same page across our franchise regions in Asia back in poon a and in London as well as in Frankfurt Germany the tableau journey so another site that we've dedicated to helping intermediate introductory as well as to our advanced even to our Jedi stat
s we created a page that will help them to go across these different disciplines so you would get started you would find some best practices how do you connect to data and then beyond the basics all of which have links that take you to other information maybe it's a wiki maybe it's a library of some sort and then our as I said the is key assets for beginners to experts and the material is a mix of links practices checklists so one thing I found in here is that people need a specific checklist to things like performance optimizing your performance what is a good dashboard and these are simple checklists that you can find on tableau as well as any places but it needs to be somewhere where they can get it very quickly and then beyond our tableau basics we start to talk about some of the more advanced things are in Python are becoming very popular and one that we will most certainly be addressing both in training as well as in our workshops tableau champions so the idea behind this live discussion board is to use the power of the community to help answer questions as you can see from the tagline I don't know if you guys can see it really well but where each question is a great question that is a true point we want people to ask questions but obviously my small team can't answer every question within a appropriate time so what we do is we have them ask their question it gets answered extremely quickly as we've gotten thousands of people watching this and being able to participate and they're learning from this as well it's also our first line of defense because many times in the beginning when I launched this I was getting constant phone calls constant messaging constant emails and things that maybe I considered simple but we're very difficult for them and in order for us to break free of that we had to use this discussion board so that they can ask the question getting answered by their own peers some some of them were sitting right next to them they just didn't ask them right so they were getting answered through here but this is an excellent way to also identify who your champions are so one of the things that we're using this for is when I see a reoccurring person answering questions for whoever they are I call them up and I'm like you're doing really good I said would you like to be a part of the tableau champion what is that and we talk about the benefits then they get a nice accolades where they sent to their manager and so forth that's how we've been building up this network of champions by seeing who's helping and who's willing to help now what I like to do I know you saw this briefly Armen had a small snapshot on the on one of his slides there but I'm gonna give you a live demo if it will work here let's see let's go all right I'm on three okay oh we're gonna get there every one second I guess it doesn't want to show yeah all right well it's not popping up here give me one moment okay we're almost there there we go all right so we're using pages to to kind of generate the effect of the growth pattern here and again this is a live demonstration of the growth and what you'll see as at certain points here in early 2016 mid 2016 we adopted as part of the BI platform but you'll see a fairly progressive growth as we would expect tableau to do but it wasn't until March of this year when we took on and launched the tablets in our practice that you're gonna start seeing huge jumps right I mean this number right now is already past world we're close to I think we're close to 8000 I have my admin back there stream of us are we close to 8,000 trini 8100 Phaleas it's changing as I'm speaking right but what this simple little slide should help you appreciate that is the the growth is exploded due to the fact that there is some personal attention being done by your internal team with your business and allowing us to be able to help in it in this case we've seen it across various regions and populations specifically as you see in our top five ranging from the UK all the way down into Singapore so this is just I just wanted to kind of give you it was my way of cool thing I wanted to showcase this this as well let me just go back and let's finish off here so what are we doing to keep moving forward okay so we've started it we've got a great adoption what do we do to keep it moving so a couple of things that we want to I want to talk about and kind of share with you so the first thing I want to do is enhance efficiency through automation so one of the things we're looking at is improving automation on our administrative side creating very near real-time or very proactive administrative views to track the health performance capacity because the growth like this can easily a platform I mean we're running around 24 cores right now we're looking to growing that especially in light of what we heard from Arman we're talk about 25,000 users so we need to be prepared for that so we need some automation to assist us in tracking our Postgres system and get that out to us quicker certified project owners a program that I will be kicking off in 2019 we need to help rather strike that we need them to help us the project owners who go through our engagement sessions need to understand that self service does not mean you put your hands up and you just let them go and do whatever they want right we need proper governance we need them to watch what they're doing to help enforce best practices enforce that there they're following proper standards and that way then we have less issues and less problems with our shared platform so we look into expanding our TCO P strategically so my small team of three we are what we did is we built a run book where we identified a recipe for this tableau center practice and now we're meeting with different groups in these regions to talk about how they can do the exact same thing we just did a proof-of-concept with Singapore and they are now running a tableau center practice in Singapore answering questions doing a lot of what we do but there's the overhead is really still on our shoulders we're handling the engagement request for handling the licensing or doing all the administrative stuff they're just managing the business and at particular regions so we're looking to expanding that out to those other areas as well we would look to it like to improve our procurement process we've got some bottlenecks that we're going to be hopefully improve so that people can get licenses faster and then upgrade so we're currently sitting on 10.4 and I think we're gonna be jumping to 2018 dot one or two if it was up to me I'd like to go all the way for the next one but we have to be very careful there's a lot of testing we have to do and ensure for a global bank to make sure that it's it works but we're gonna be talking about how we can do that more appropriately how do we scale for example you guys all heard at the keynote the tableau the tableau prep editions they asked data all of these do have an inherit effect on back end on the background errs on the service we need to think about that so that we can be proactive to prepare for upgrades rather than waiting and then we get long delays for getting it out to the public and then we want to build stronger tableau champions we want to expand the train the trainer program so that more champions can come up and teach other people how to train and how to start the center of practices across the globe so here are just a few takeaways that I wanted to give you some of you can after you can ask me questions as well but one of the things you first thing you want to do is identify the strengths and weaknesses that you may have inherently with starting up a center practice or a center of excellence and take an agile approach my worst fault for me was I wanted to take on the world right in the beginning and I found myself failing too many times so when I thought about it we needed to come back with tableaus help take a little by little introduce a tableau user group maybe a little later once you start figuring out who your community is but yeah take an agile approach to it build strong partnership with your business friendlies with your senior management get them involved show them the benefit hypothesis right what is it that tableau can do for them make it simple and direct and you'll see great results from doing so and don't forget to communicate and communicate does not mean sending out a slew of emails to every known person that was one of my faults I had emails going out and I kept hearing people say stop communicating to me stop sending me emails so what we did is we moved it more to our intranet platform and allowed them to be able to consume it rather than sending every new update via email there is still some emails we have to do like newsletters and so forth but we've cut it down and leveraged the success of your peers so myself others you're gonna see here at the tableau conference are here to help right we're not on an island we should all be working together to building a tableau community and assisting of where were where we're selling and where maybe we need some help from and then focus on building your tableau champion Network I can't stress enough that if you're a large organization especially if you have a global footprint you need these champions you cannot do it yourself you need to encourage them make them visible make them feel proud to be a part of advocating a product that is helping when you do that you're gonna see them doing even more and that has been the that is the foundation for where I'm seeing our our growth and success across the global footprint so I think I did pretty good on time I did pretty good so I wanted to open it up now to questions either for myself Arman is here to answer questions and we're just going to get one sec we're gonna get a roving mic here get it out there you got it Oh yeah so let's see let's start right here in the front here gentlemen we get somebody to help with a roving there you go Mike thank you gee what's your approach to I suppose realize the benefits that there's both the insights that the the different team developed and how do you make it visible to senior management so you can continue a tabla new students continue being funded I'm a funding one is it's a challenge in our organization like in every other so it's a fight like we have to fight for you know and define you know what we think the benefits are gonna be and we go through that kind of on an annual basis again my organization mother piece of my organization is now much smaller than the total organization in tableau so I presume others are going through that fight but a little bit of it sells itself you know the way we did it and it sort of goes into your second question I think so can can can can add to it but the way we've done it it's sort of our senior managers we just take things that are in Excel or broken or slow and we'll take that information and give it to one of our champions and turn it into tableau report in 24 hours because not that hard to do from an excel file and then we just give it to senior man and say this is what the the simple power of it is that the some one report they may see every day and that becomes the carrot yep yeah and while the businesses owned a lot of the assets when it comes to getting their licensing and so forth the exposure of come you know successful launches - you know polished workbooks that are excelling in one part of the bank and others are just getting started how you get that is we're using them we're using the intranet sites that we have we display them when we can write some of it does have some some confidential information on it but then we make leverage of a newsletter we have a quarterly newsletter where we get you know senior leaders to quote on the success we showcase the accomplishments through their visuals we give them the a picture of what the visual is and what the benefits have been derived from it so it's a good way for us to share but yeah yeah we use a lot of analog based you know distribution of information so we actually hand out the newsletter we print it out and put it on people's desks and because there's just too many mails right and you know yeah we don't do that for the entire globe oh you should admit that in turn they matter per second but it ultimately works that's what works is people pick it up and read the newsletter so yeah hi my name is tal ha I'm with LK manufacturing when I hear your story is actually very similar to ours we're probably earlier in the journey it started off with in one part of the business and then I'm with an IT but we quickly adopted it made it our enterprise standard one of the things that we're struggling with is how do the traditional bi tools the data warehouse coexist with tableau is there a methodology or an approach that you've adopted to say well this kind of analytics is a perfect candidate for the data warehouse and this for tableau yeah so I mean right off the bat you're gonna find that from a data modelling or data warehouse methodology it's for tableau specifically it's good to look at that aggregated views right not to say that you can't use a you know a deep puddle in the secure data lakes right to do so but the methodologies we've seen successful are at the aggregated sections of the of the data warehouse right something also to keep in mind that for for Deutsche Bank is probably for many of you tableau had a very specific purpose to remove and user developed applications from the community or from the organization these are excel based thick you know driven applications and the content the data wasn't a lot but they needed to get it pushed into a data warehouse right then you have a more centralized portal for you to be able to extract and build visualizations on so the idea is that many of the data warehouse methodologies outside and we don't really touch OLAP models much but most of them work very well with tableau question here hi this is umbrage margin from Prudential my question is around your journey from being sort of an early adopter to what you are right now and typically on this journey the first hurdle you come across is risk and compliance and then the second one is ID and you talked a little bit about ID but I just want to get an understanding of how those conversations went as you went on your journey typically risk and compliance are worried about this is the right data who should see this data or shouldn't and an IT has their own concerns about roles and responsibilities and what you do and what they do so could you give a little bit of insight or those conversations will give you some insight on the on the compliance part you talk about the technology side bump Alliance respective same pushback right initially it was this data hasn't been cleansed how do you know if this is right data and there was a lot of conversations around that we ended up having to do is again this goes back to very simple mechanisms is we ended up like putting large disclaimers on top of it almost sort of a you know watermark type of analysis to say this is in draft form and this is something to be used just for you know management you know analysis at the same time and then we started to really make the argument like if somebody consent around an Excel spreadsheet all they want around the organization right and that's not considered golden source data and so this is just another tool that is better than Excel but it really took the explanation with compliance but then going back to getting senior management to say I really want this I want to see this report every single day and that's you see quickly you know compliance can can adjust their their thesis yeah and from an IT perspective we had to partner with that same thought process of keeping data secure so security was key for this whole journey right because again there are groups like the finance of groups in compliance and even in HR so we would encourage a dual layer of security going back to the data warehouse encouraging that a lot of that handling role based there but then we would help them with regards to this second tier of within tableau to provide that kind of security that way then it helped you know help or the the risk tha
they were going to take to be able to ensure that there wouldn't be any issues with tableau specifically I have a question here yep Doug Ranahan IQ via we're starting on the journey of creating I'm starting on the journey of creating a center of excellence and one of the challenges that I'm running into is we have three tableau servers spread across the globe in different business units and we're going to try and be coalescing those and everyone's got their special case of why they're special and special er than everyone else how have you addressed that you know with have your isolated pockets and bringing them together yeah I mean right now as you heard I we have one us-based instance as 24 core that's actually supporting a majority of our global footprints we do have a separate dedicated instance and it might help for me to explain a little bit about dedicated and shared right so we have a very diversified portfolio of businesses across the globe right and from HR to different groups they're all using one server over here then we have a dedicated server specifically for compliance and risk would they want to be separated they want to dedicate kind of like what you described multiple servers right we are not encouraging these dedicated servers for a lot of apparent reasons we would like there to be a part of a more we grow out clusters to be able to support more of them plus the cost model cost model is extremely hard to justify you know whatever the amount is for one server for one particular group when that cost could be charged back across all users across different domains in different groups so that's why the justification for us would be to try to build out these unique servers unless you come across the only other challenge we'd have is like data privacy laws in some areas crossing data over so in those times we may have to make some adjustments yeah I think and we're still not perfect no we had a group in London that was you know that affects Crippen yeah as in their own world and we're refusing to kind of be a part of the center of excellence and be part of the architecture we created and they're now moving over but it took me calling the head of the business me like what are you guys doing you're in left field weird them some of those things that the team you know South tainment created and said you guys are losing the benefit of all this so but it's not perfect it's so it's it's just a lot of you know push and pull yes Alan Armen with five minutes left to your right two more questions yes God hi I'm Ramesh from University of California San Francisco thank you so much for sharing your journey and your learnings and I have two questions you know on the training side I know you have put the roadmap and then you organize the training for the tool how do you dream people on the data because you know without knowing the data just the tool is you know it's not sufficient so if you could help us understand you know did you organize the training on the data and the second question is on your onboarding process when when you engage people that offers conversation how do you decide whether the person is asking to be a developer or an explorer or a viewer you know so yeah so again I think the first question was on how we would do training with groups who would like to use their own data general data yeah so what we typically do the training is is done from both again from a desktop one desktop two methodology they're using some generic data right now for for the basic principles and fundamentals of building a dashboard right when they get to the next stage we encourage them to use their own data sets now in some cases we can be an employee can can see their data some cases we can so it's not perfect right it's very difficult but we seem to get around it we allow them to mask them the data from before we can come in and sit with them whether it's a doctor session or a workshop but that typically helps accelerate training faster than it is to use some superstore model which is good for the basics but not when you want to start learning how to put insights together yep and the second question was on the roles of creator Explorer viewer yeah so we we do that during our engagement model we identify who they are what roles they're looking to doing are there is there going to be like one developer and then you know a thousand create you know interactors we get all that big picture that way then we could help scale them and start talking governance about what they should be doing what roles they should be looking into accomplishing I had a last question right in the middle here yeah I know she's been raising her hand and missing there thank you hi Tori sparks with Mondelez international thanks for your time a couple of questions one quick out of curiosity what platform is your user community built on and then secondly I sit an IT but I'm business engagement with our strategy insights an analytics team and some of the feedback we've gotten I'm based in our global headquarters in Deerfield Illinois but some of the feedback we've gotten from the regions is that it feels like the global team is really pushing a lot of stuff to the regions without that two-way communication and when you were talking about really identifying your tableaux champions you said it was really important to get people in the regions how did you do that what was your process for identifying your tableaux champions and then you also talked about the benefit of being a champion what are the benefits at DB for being a tableau champion thank you you know the brain always remembers the last word so yeah so let's through the champions first so the champions are identified they were identified we had a group when I first launched he had five but they were nowhere they were all in Cary North Carolina and we didn't know how many we actually had one Quinton who is in Germany who actually was part of the GCT team and built a lot of that so I needed to figure out how to get more champions and what I did find out is that the data scientist community the strong analyst community were already eager to be a part of and they were already using tableau I just didn't know that yeah so as I mentioned previously we also had the tableau discussion board and there I started find out people who were answering a lot of questions both difficult and easy I contacted them and got him into our repository we're growing so we started getting names of individuals right we sent out a survey I probably should have mentioned that as well when everyone's here surveys are very important because in the survey I asked would you like to be a tableau champion and we got 40 more people want and then we grew that out gave them appropriate training what does it take to be a champion what does it mean to advocate what does it mean when you're gonna be called on to do a doctor session but none of that would be helpful if they didn't know the benefit of being a champion so for us the benefit was you know exposure at a very high level about what your accomplishments are and how it impacts the bank - we give some recommendations to their management as it helps them with their career path and the third one we're working on right now is a more of an incentive program where they get high marks which would lead to you know whatever we can do it don'tcha Bank to give to them but something either you know a prize that we get from swag from tableau a water bottle goes a long way in the heart of a tableau champ you'd be surprised I'd like to but they wouldn't want I think also philosophically you'll see that the community really strives to help each other out you know - because everybody struggles with their data right and so it you can talk to any organization or any part of the part of the world and it's just one of the most frustrating things in the world when you can survive access to data and then you start to get a piece of it with tableau and so we find energized folks across the globe and it doesn't it's not we can't it's hard to pick them right you'd think it's whether it has to do with youth or whether it has to do with what part of the organization they sit in there's no rhyme or reason to it frankly we get people I have a traitor on my desk who's when he's old grumpy traders he's like forty five years old and he is really one of these like really curmudgeon he kind of guys and he's one of our biggest creator of reports he just loves it because he can get data himself and doesn't have to deal with lots of people so I think just getting the word out but you'll be really surprised how once you sort of expose the fact that it's social how many people kind of will come to you yeah yeah I think that's it for our time so again thank you all for the platform your community was built on sound the platform for what your community was built oh the Internet oh that's a great question I'll get back to you on that yeah come to me outdoors I'll think about exactly what that platform I have my admin over there he can ask that question thank you thank you and please complete this survey as Sal head up on the previous slide yeah Thank You Sal and Arman for your time
