everyone my name is Edson I'm I'm the co-host I'm a voluntary co-host of the Korean talk of today I'm going to introduce to you uh direct Bill Gia from analytics Studio uh here's the current CEO of analytics Studio which is a Canadian consulting firm providing expert expertise and guidance in data analytics to clients in a variety of Industries and he holds a economics degree from Charleston University and an MBA in finance and marketing from UC and he has the CP he has he's a CPA and here in th
e parade management professional designation for the TMI Institute so his career has include assignments in more than 15 industries from organizations across Canada and organizations and organizations according to him is increasingly hiring data data scientists to improve operations accelerate revenues and support information but to many times data science breaks delivering satisfactory results fortunately business analysts are equipped to play a critical role in ensuring that data science proje
cts selected are where the investment so this session will examine why data science breaks fail and explores the various ways that business allies can drive success so now I'm going to on share my screen thank you for that very kind introduction and for inviting me to speak today I've got a lot to cover so uh as he said I'm going to ask you to put your questions in the chat and I'll do my best to answer them at the end of the presentation and should you wish to connect with me after uh please se
nd me a message on LinkedIn I'll be happy to connect with you so um this is kind of an outline of everything that you just was told um give you some hints on what I've been doing for the last 30 years a good part of my work is involved business analysis but I've made contributions in other areas including financial management project management and strategic planning as he's told you my education began with an economics degree at Carleton University in Ottawa and ottawa's where the snow is plent
iful but the skiing is not fun so after a trip to Whistler many years ago I abandoned Ottawa for Vancouver and I've been here ever since my work here started as a financial auditor to try on track to become a chartered accountant and that since that required a deeper understanding of business I signed up for the MBA at UBC and since then I've kept plugging away earning a variety of designations just like all of you and this certainly helped me prepare this presentation for you today So currently
I work for a small consulting firm known as analytics Studio the company focuses on data analytics I created it in 2016 when I realized that data analysis was really my only real passion so since then I've been lucky enough to attract some very talented individuals from both UBC and Simon Fraser who've helped me to satisfy some pretty demanding clients before putting this presentation together I wanted to get a deeper understanding of your experience with data science so for those 29 people who
responded to my questions I want to thank you so these next four slides are all about you the first Insight I found is that 80 percent of you are already dealing with data science in one way or another though only a small proportion of you are currently seeing significant benefits so this tells me on on the right path the second Insight I found is that for two-thirds of you data science was important but for a third of you it was not currently a factor in your work this next Insight really moti
vated me is ninety percent of you are interested in increasing the Reliance of your organization okay I'm just going to go back and show you this is this is what I covered and you're going to have access to this deck later on thank you so my apologize apologies for messing up on this so um the next Insight really motivated me uh and and you'll see why here it's because you all want to get going with data science in your organization and finally I saw that more than half of you have already been
building your skills and knowledge in this area so with all of that in mind here are six questions we're going to explore in the coming hour have a quick read so to start I want to talk about data about data science and why your career as a business analyst is so heavily tied to data and the huge amount of techniques and Technologies used to manipulate that data so for starters let's enumerate some of the ways that data science can help your organization here we see that data science is going to
reveal new and valuable insights about your business it's going to lead you to good Solutions and products it'll solve Mysteries like why things are suddenly failing it'll make your customers happier it'll let you incorporate new data from other sources it will provide validation for your work it'll create long-term value for your organization and finally it's going to provide data-driven decision making when we think about our organization our own organizations we can think about how as employ
ees we work to connect suppliers or providers of inputs to our customers or clients those consumers of our outputs from a functional perspective many of our organizations are structured into three large categories with departments inside each of these categories here's a typical structure now not all organizations are going to have these elements but this is a helpful way to understand how we organize to serve our customers and clients what you'll find in many organizations is that each of these
departments and subunits collect data in their own fashion according to their own rules and there may not be a lot of consistency across the organization even in naming items that are critical to the business what date what business analytics and data science attempt to do is span all of these separate domains in the case of data scientists they're also likely to be looking outside the organization to suppliers customers government and others for data that's going to help them Drive insight for
the organization another way to think about this is to examine the flow of cash through the organization every organization receives funds it's often for a variety of reasons so it's important to understand where the money is Flowing from and why evaluating this can be categorized as first of all customer engagement analytics so whether it's cash coming from a retail customer a Wholesale customer or even a government agency there's data to be analyzed and we can generally call this customer eng
agement Analytics similarly we consider the risks associated with cash whether it's credit granting or managing receipts there's data to be had to support your risk management Analytics and thirdly we can think about the necessity of optimizing inputs and outputs of various kinds that influence the movement of cash from going out to coming back in and this is generally referred to as optimization Analytics now I'd like to turn your attention to a little framework that's helpful for understanding
the difference between business analytics and data science here we organize everything we know into four categories in the upper left hand corner we can put all the things that our organization understands and that is known to our customers or clients going down to the lower left hand corner we have all those things that are proprietary to the organization those things we know but our customers probably don't know or we don't want them to know in the upper right hand corner we have those things
that our customers know but we haven't yet figured out and down in the lower right hand corner we have those things that neither our customers know or that and that we don't know so you might call that the unknown unknowns so while that box might appear to be empty in fact with a little digging and scraping we might actually be able to extract insights of real value so let's take a look inside each box in the public box this is the domain for business intelligence our work in business intellige
nce also would involve the proprietary or confidential information that's private to the organization so these are two boxes you're likely very familiar with however in the third box which is labeled blind this is where we would put customer analytics uncovering new information that we can see before may involve business intelligence but more likely will be the result of data science techniques and when we get down to the last box of unknown items this is where we apply big data analytics so the
job of data science is to help move the boundary between the known and the unknown now since the beginning the beginning of time humans have had to make decisions in the past if we study carefully that instinct May well have been the most popular method if you were deeply involved in your business understood the ebb and flow of demand and Supply then you had a pretty good grasp on what to do when decisions were needed you could reference the results in your financial statements and make mental
connections back to the actions that got you that profit or loss you could make a good guess and then make some plans based on your best gut instinct but if you didn't feel like trusting your gut then you might bring in your Senior Management or your best outside advisors and after some discussion perhaps put the matter to a vote or possibly what would happen is that everyone in the room had a different idea so that decision would be made by the highest paid person in the organization we know to
day that these approaches to decision making well they're certainly fast and convenient they don't always give you the best results now to slightly change the topic here baseball's always been a professional sport that celebrated statistics so in 1997 the Oakland A's baseball team was struggling with a very tiny budget and they hired a man called Billy Bean as the team's head coach Billy Bean had to solve the problem of how to win more games with much less money than competing teams like the New
York Yankees Billy Bean's eventual success was turned into a movie called Moneyball starring Brad Pitt so I want to show you a movie preview that literally spills the beans there are rich teams and there are poor teams then there's 50 feet of crap and then there's ours that's a dollar man what welcome to Oakland I need more money we're not New York my players or the money that we do have I like Perez got an ugly girlfriend ugly girlfriend I mean no confidence you guys are talking the same old n
onsense like we're looking for Fabio we got to think differently who's Fabio your goal shouldn't be to buy players your goal should be by wins in order to buy wins you need to buy runs who are you I'm Peter Graham's first job in baseball it's my first job anyway we're gonna shake things up and she'll walk me through the board I believe there's a championship team that we could afford because everyone else on their values him like an island of Miss pitoris we want you at first base I've only ever
played catcher not that hard Scott filmwatch it's incredibly hard you can't grow but what can he do you want me to speak he gets on base we are card counters at the blackjack table we're going to turn the odds on the casinos I'm heading there text me the play-by-play wait what I'm watching games Billy Bean has tried to reinvent a system that's been working for years it was a nice Theory just not working out how long has Billy Bean going to last he's proven himself right out of the job and their
minds is threatening to gain certainly the way that they do things do you think you lose your job what where'd you hear that I want to go on the internet sometimes don't go on the internet watch TV or talk to people you're discounting what Scouts have done for 150 years goodness what is happening at all but it defies everything we know about baseball [Music] will change again you've got a word I'm just kidding so what happened this table tells the story but that was the raw data here's a much b
etter way to visualize the results with this visualization you can quickly see the advantage that was created now not so long ago nearly all data held by organizations was captured by hand in fact my grandfather's on both sides were very skilled at handwriting and you know that was crucial for their careers and then during their lifetimes typewriters became quite common and the speed and Clarity of captured information increased considerably but the amount of data captured was absolutely tiny co
mpared to today a year ago I purchased an iPhone with a one terabyte capacity which means that I can carry around in my pocket a thousand times more data than my grandparents could ever have scribbled down over their whole lifetime every time you touch your portable phone you're creating more data and quite often this data is traveling outward in many different directions and is being stored in a huge variety of locations what this means for you as a business analyst is that compared to even fiv
e years ago the amount of data available to your organization is many times greater in volume and complexity yeah so given the endless and massive generation of data every second of every hour there are many steps involved in handling every piece of data that you and I are continuously creating everyone is involved with data one way or another we create records of assets events and action we use it to make decisions we manipulate it for new insight we consolidate it for our reports we analyze it
to gain new understanding we work to protect our data and we share it with others in our organization but perhaps the biggest challenge of data science is the never-ending struggle of dealing with data that's poorly controlled these issues include the following people can forget to record assets or events they can use it to justify poor decisions sometimes they manipulate it to hide poor performance they may erase it accidentally or on purpose they can waste valuable time analyzing it for meani
ngless reasons or they can fail to protect it or as the U.S government found out recently they can share it with Outsiders without permission so keep in mind that these are only some of the actions that create havoc in the world of data scientists next I'd like to talk about what makes data unique as an asset to be managed if you're an accountant it's your job to know exactly how much money there is at any point in time if you're a human resources manager you're responsible to know exactly how m
any people are working and what they're doing if you're an asset manager in a factory you need to know where the tools and equipment are located and if any of them are missing but if you're a data manager you're dealing with a very unique asset for starters data is invisible if you talk to those accounting people they'll tell you that they put no Financial value on data and in many organizations you'll be stymied to find out who is the designated owner of the data all too often data is just a po
or orphan so what makes data so unique and different as an asset unlike physical resources such as oil or gas data can be collected and reused without being depleted data can be constantly generated captured and processed without losing its value data can be shared among multiple users without being consumed or diminished the same data can be used by multiple individuals or organizations for different purposes all at the same time data doesn't have to have a physical form and can be stored in tr
ansmitted electronically this makes it easy to access share and analyze across many different locations and devices data has no physical presence which makes it difficult to quantify in value assigning a financial value to data can be challenging and often depends on the context and the intended use of the of the data the value of data can vary depending on the user's perspective context and the intended use what may be valuable to one user may be totally irrelevant to another despite its intang
ible nature data has tremendous value to businesses and organizations it can be used to make informed decisions improve operations and again a competitive advantage data is everywhere and is constantly being generated by various sources such as sensors social media and even online transactions this makes it an abundant and accessible resource for businesses and organizations and even more your data when graphed can assume an endless number of patterns but now before we get into some of the detai
ls of data science let's start with a quick review of your responsibilities as a business analyst you need to understand the problems facing your organization and the goals that have been set out for success the key problems in your organization are typically identified by certain measurements this means that they're evaluated on a scale and likely recorded on a regular basis similarly the goals for your organization are very often expressed numerically meaning that key targets have numbers you
are also tasked with identifying what the organization needs and doesn't have today again this is something that is very often involving measurable characteristics or measurable requirements as well you need to discover solutions that are going to benefit the organization and those Solutions often involve the manipulation of data as a ba you need to be devising strategies that are dependent on those Solutions and will take the organization to desired outcomes that are measurable finally you need
to make sure that all the players in your game work together in a Cooperative fashion and fully understand each other this means that they need to speak a Common Language so there's no confusion about what will be managed and measured now let's explore the problems that data science can address you know that data science can do a lot of heavy lifting first here's the general context whether you're in a commercial profit driven organization or serving the public in the non-profit sector your org
anization faces the same three General challenges the amount of money and the number of Staff in the organization are never infinite how you manage those limited resources can determine success or failure in reaching strategic goals secondly staff at all levels contribute to the success of the organization or they don't when staff fail to deliver when processes break down when results are mediocre the organization is at risk of collapse thirdly all organizations face risks that are both internal
and external that are manageable if addressed early enough or can be avoided by changing direction in time these risks never end and need to be identified and managed if the organization is to throw survive and thrive so it's data science that can help organizations address these three major challenges many organizations use data science and data analytics to find ways to reduce cost as an example if you have hundreds of suppliers each generating thousands of invoices then data analytics can qu
ickly solve the problem of identifying which suppliers are overcharging or in more complicated environments it can be through the use of data science that organizations find better ways to combined inputs needed for product creation this may involve changing business processes training staff with new skills or implementing new technologies to make work easier faster and less costly these kinds of exercises can help generate better products and services at a lower cost so while cost reduction can
be critical the long-term success of any organization often depends on Innovation and related performance improvements that lead to higher or more stable Revenue as an example data science can be critical in identifying what new products should be offered and how best to create and deliver them similarly organizational performance often depends on having a very clear and precise understanding of customer needs and preferences data science is excellent at accelerating the delivery of these insig
hts finally value can also be enhanced by reducing risk buried in those mountains of data that every organization collects are indicators of things that shouldn't be happening or things that if not properly controlled could result in a disaster data science can be a tremendous help in pinpointing the most serious risks early before they start to destroy the organization so there are many ways that data science can deliver value your data scientists dig out those critical insights that add value
they help the team come to better decisions they generate predictions that can be devalidated they provide the path for automating processes they help to drive innovation and what this adds up to is they enhance our experience to deliver on these objectives there are a wide variety of techniques that can be applied here are just a few now well it's not my job today to teach you how these work I would suggest that if you are working with data scientists then you'd be wise to develop a general und
erstanding of these and those other techniques at Play now which Industries benefit from data science well the answer is that really there are no industries that do not benefit from data science to make this relevant for you I worked on identifying those major industries where you as business business analysts are working to deliver value I'm going to run through my understanding of some of the many ways the data science might just benefit your organization since many of you work in retail I'm g
oing to start there if you work in the retail world you already know that managers have endless questions the data science can help them answer here are just a few of those questions where should we locate our stores or branches which suppliers should we use how can we optimize our inventories what do our customers want and when what are they willing to pay what skills do our staff need who should we hire and how can we best equip them so let's take a quick look at an example I imagine that ever
yone here knows about our provincial government's biggest retail business the BC liquor stores now I'm pretty sure this picture was not taken in BC but it illustrates the idea that retail preferences are very different depending on location well you might be able to sell an 80 to 80 bottle of wine up here there's almost no chance you're going to find a buyer here so to make the point clear have a look at this this is what's offered for less than 10 bucks in more than 145 stores across BC and thi
s is where you can find your thousand dollar treasure but just in one or two stores assuming that you get there on the right date so the next time you want to spend 620 plus tax on a bottle of cognac you have to go to Richmond similarly there are many ways that data science can help in the world of higher education these are only a few of the questions where data science can provide guidance I'm not going to read off this list but you can access it later by downloading the slide deck and again i
n the world of Health Care where the lives of millions of people are in play data analytics is a crucial factor in delivering value and allocating those limited resources again go to the deck if you want to start your list keep in mind that there are dozens more questions for each of these industries here again in the world of banking Investments there are many potential benefits of applying data science to optimize operations and client satisfaction and for municipalities they're involved in a
wide range of activities from tax collection to road construction and maintenance policing managing real estate development setting environmental controls and much more data science can add a lot of value so here are the key takeovers sorry takeaways that I'd like to offer first of all all organizations of all sizes in all sectors of industry and government they benefit from data science and the more you do the more you get but there's more it's also worth pointing out that data governance which
is the focus of management on properly managing data to ensure high quality appropriate security proper focus and Secure Storage it's essential to optimizing the benefits of data science now sadly way too many organizations only pay lip service to data governance and they pay a price the problem is that too many managers just can't see the value since data governance by itself doesn't make money the final point I'd like to make is that your value to the organization depends in part on your unde
rstanding of the world of data scientists and data engineers so that brings us to this question what should we know about data science in this section I want to expand your understanding and highlight some of the key issues and opportunities every organization I've ever worked with was competent at reporting on past events some have been much better than others at summarizing and dissecting those historical events but the basics of telling what happened is referred to as descriptive Analytics go
ing further takes us to diagnostic Analytics telling us why those things that happened turned out the way they did for most organizations data science involves a progression from basic reporting and Analysis of historical events toward the ultimate goal of using data science to make critical decisions about the future when you know why things are happening then you have a foundation for predicting the future what we refer to as Predictive Analytics now once we have that mastered then we're equip
ped to rearrange our assets to change direction using the insights of data science we refer to this as prescriptive Analytics a key Point here is that this is a mountain that needs to be climbed each step up the mountain requires more skill and effort but the payoff becomes greater so you can't deliver the benefits of prescriptive Analytics with Oda solid foundation except of course if you have a crystal ball that was a dad joke mom Davenport is an academic he's written extensively on data analy
tics and he is the founder of The International Institute for in Analytics Davenport developed this organizational maturity assessment for Analytics using these metrics you can determine what level of maturity your business can claim and then you can learn what it takes to go up to the next level at the lowest level level one your organization is mostly ignoring the data as and making decisions based on what we just said before punches and gut instincts at the highest level your organization is
using data extensively as a competitive weapon for every major decision in between you have increasing levels of adoption and sophistication in terms of your tools and expertise this is one way to keep the data science scorecard for your organization so what is it the data scientists know and do there are three broad areas of competence needed to be a successful data science scientist the first and most prominent is statistical skills some of the skills are listed here secondly they need to know
how to harness programming languages and analytical tools to build and test the statistical models and their other creations and thirdly they need to know the contextual realities of the organization that they're supporting to have impact they need to clearly and accurately communicate their insights to the players who are going to leverage their work it's in this third area of competence where you as a business analyst add value so this intersection point here is where data science lives so th
is leads to one of my main points you're very often the translator between the management team and the data science team helping to make the communication between these groups walk work flawlessly so in your role as business analyst you most likely share a lot of characteristics with data scientists you look for recurring patterns of behavior you use a variety of tools and models you focus on business process Your Role isn't traditional you spend a lot of time on computers handling data you know
that pictures often outperform words you spend time with the leadership of your organization and you both care about getting accurate content and a clear understanding of the issues but then what makes you different in contrast you do lean in a different direction from most data scientists not a bad thing first of all you're stronger domain knowledge and that's what the data scientists need you focus on discovering needs their focus is data analysis you focus on explanation and Design their big
focus is scientific method you're searching for Solutions they're searching for explanations or proof you have limited Reliance on statistics and that's their core competency you have a standardized presentation format typically they're really good at creative visualization you have a longer history in your organization you're widely accepted for the value you can add and frankly data scientists and many organizations are not well understood because they haven't been around that long and they'r
e not often that good at explaining why they're there so what each side can give but these differences in mind let's review what each side of the equation can provide business analysts can share their deep understanding of what drives the organization and how management views their world they can help setting the context for new initiatives they can assist in setting priorities when demand exceeds the available resources which is a very common issue and as well they can validate exploration proj
ects for relevance and feasibility in return data scientists can deliver ideas never before considered they reinforce their work with scientific insights and the rigorous methods and given their training and mental dexterity they can reveal opportunities that may never have been thought through about in the past so each side has the valuable contributions to make and when brought together these points of view can add considerable value to your organization so in making this collaboration work su
ccessfully it's critical that you build a cohesive team starting with your data scientist you're also probably going to need a talented programmer who can do the necessary engineering work to operationalize the creations of your data scientist and of course your work as the business analyst is equally important and you will likely want to have a specialist in data visualization and communication this person will help to deliver those key messages that come from all that work one of the tools tha
t will help your team to deliver on their promise is the data dictionary a data dictionary is used to standardize the definition of each data element in your data universe it enables the various players in your organization to share common understanding of every data element used by the organization the data dictionary is used to document standard definitions of data elements their meanings and allowable values the data dictionary contains definitions of each data element indicates how those ele
ments combine into composite data elements and they're used to standardize usage and meaning of data elements between Solutions and between different stakeholders data dictionaries are sometimes referred to as metadata repositories and they're used to manage the data within the context of a solution as organizations adopt Data Mining and more advanced analytics a data dictionary can provide the metadata required by those more complex scenarios a data dictionary as you may be aware is often used
in conjunction with an entity relationship diagram with the support and guidance provided by your data dictionary your team can engage in serious data mining data mining is used to improve decision making by finding those useful patterns and insights buried in your data dating data mining is an analytic process that examines large amounts of data from different perspectives and summarizes that data in such a way that useful patterns and relationships are discovered the results of data mining tec
hniques are generally mathematical models or equations that describe underlying patterns and relationships these models can be deployed for human decision making through Visual dashboards or reports or for automated decision making systems through business rule management systems or in database deployments now recently the world of artificial intelligence has been dominating the headlines hundreds of cloud-based AI applications have been released to the public in the past few months probably the
most prominent one has been chat GPT if you're not using it already you've probably heard of chat GPT and before we get into it I want to ask you to participate in a little survey just to get an idea of where we're at with chat GPT as a group your response is by the way are confidential and will not be shared we ready for that so to put things in context I want to go back more than 100 years my grandfather was a teenager when the first modern aircraft was flown at the time it was a notable and
innovation but for many people the idea of going way up in the air without a strong ladder was mostly a curiosity I'm pretty sure that nobody living in Vancouver at that time imagined the thanks to airplanes they would one day be able to spend the weekend in New York and then be back at work on Monday but over time our perceptions and our abilities have been totally transformed so in my view the introduction of large language models to the general public represents a similar inflection point in
Human Experience but we are at the beginning of the major transition that AI will bring to everyone over the next few years it's notable that in just five days more than one million people began playing with chat GPT and within two months a hundred million compare this if you will to the 24 months it took the first one million people to begin playing with Twitter today's social media Giant now chat GPT has created both praise and Terror in some quarters likely you've already been exposed to some
of the noisy debate about the impact of AI on jobs on people in the world in general some folks are focused on the positive aspects and some others are very worried about the harm that AI can do to our whole civilization so we have both hopes and fears associated with the introduction of chat GPT and all the similar other AI applications of which I know only about 100 to 200. on the positive side it's easy to see how it can make people more productive leading to faster results new and approved
products and services greater versatility at all levels and overall a better world but on the negative side there's a fear that up to 50 percent of all white collar jobs might just disappear there's also concert considerable concern over the potential loss of privacy confidential information and control over things that we own including intellectual capital could also fuel public outrage on social media fears have also been expressed about the new leverage that could easily give AI to criminals
and foreign adversaries there's also a fear that control over AI will become concentrated in a few hands this is to happen it would further exacerbate the existing division between rich and poor people all around the world and at the very extreme Edge there's a fear that the intelligence of machines will rise above human intelligence levels and could potentially lead to Global destruction when those AI machines lose patience with the crude thinking of human brains or just develop thoughts that a
re wholly destructive Please don't panic but this leads us to the next item on our agenda what are the business risks that come with data science in the past I've worked with clients where much of the data was entered by hand Often by clients or customers who were not much good at data entry or just hated the work and made a mess here's an example of what can happen now do we have one city or 16 different cities this is clearly a data quality problem one that's obvious there are way too many of
these problems with data and too often they're not obvious and there's and as a result they are seldom resolved quickly if they're even identified in time since data is the foundation of data science more advanced applications can take huge resources to deliver results but also be spurious in what they tell us so data quality is a serious problem that's too often overlooked and it can have a serious impact on the productivity of our data scientists to illustrate this point I'll give you a high l
evel View of the process that most data scientists often follow to start with they collect the data that they need then they process it they clean it once they're happy with the data they start exploring it and that leads to creating models that they use and when they're happy with the work of the models they produce an output in some form and they may also build some kind of a data product that actually can use the insights to deliver value and this goes back to the world now the nasty reality
is that too many players in the data science world are complaining about data quality over and over they tell us that they often need to spend 80 or more percent of their working hours organizing and cleaning up that data before they can even start analyzing it considering the data scientists are typically very well paid doesn't seem like a good idea to me that they're spending 20 or less of their time doing what they were hired to do another severe risk for many organizations is that they own a
vast amount of sensitive and private information it can cause serious damage if the data is allowed to flow beyond the boundaries of the organization well this responsibility is often assigned to a separate Security Department involvement of data scientists with this same data needs to be carefully managed on this point further to my earlier comments about chat gbt and other external AI Services unique you need to take into account all these Associated risks so don't share sensitive information
with chat GPT it can act like a parrot squawking your private information to the whole world as an example Samsung's semiconductor division allowed their Engineers to use the service to help them fix some problems with their source code however the workers unintentionally entered sensitive data thus making the business vulnerable and heavily reduced their competitive advantage since any data shared with chat GPT is now stored on open AI servers it's impossible to pull it back to delete it and t
hat would leave your organization vulnerable to potentially disastrous consequences while generative AI is undoubtedly a valuable tool for improving workplace efficiency it's important to ensure that you have full control over how your data is stored and accesses accessed now one way to deal with this is build your own chat tool and that way you have the means to ensure that your sensitive data is secure and protected from potential breaches making poor technology choices can trip can your data
science efforts so another risk to point out is consider the technology politics that often arise in many organizations Some people prefer to leverage their existing skill sets they lean heavily towards what they know already but what they love is not always the best choice other technology people submit to an urge to stay on the Leading Edge of Technology they're always jumping on those latest offerings not always but it can lead to dangerous consequences particularly with recently released cre
ations so one way for you as a business analyst to reduce these risks is by making sure that the requirements you deliver are sufficiently detailed and comprehensive a final point to consider is that the execution of data science projects can be poorly managed there are a number of pitfalls to consider your objectives are not clear and measurable your data science team is focused on the wrong problem the real world constraints facing your organization are being ignored the outputs of the data sc
ience team are not properly tested and the results are not being communicated fully and clearly a final point to consider sorry [Music] um in this final section of my talk I want to pass on suggestions for your personal development as with all important aspects in your life it helps to construct a map for the future this map will lay out the things that you learn and develop over time I can suggest two things about your roadmap first keep it up to date and expand it over time as your learning ad
vances secondly share it with your data science buddies and incorporate their suggestions those that you think will make you a more effective business analyst I expect that many of you are on the path to iiba certification if you're not already there if you're studying for certification you're going to find a variety of references to data analytics world in the techniques section of the business analysis Book of Knowledge or babach as we call it I suggest that you spend some time on these items
on another dimension what I have to tell you today depends to some degree on where you are on the spectrum of organizational size and complexity some of you may be in a relatively small organization in extreme cases you'll be able to name everyone in the organization but in somewhat larger firms you'll still be able to recognize a large proportion of the staff for these organizations there will likely be only a handful of key applications that are critical to the organization as well you might b
e on your own or be part of a small team of business analysts in these settings you'll be carrying a lot of responsibility for defining critical and strategic projects this is going to have an impact on what you learn and how fast you learn now some of you may be at the other end of the spectrum working for a very large organization in this case you'll be doing well just to know those people in your department and there's no likelihood or possibility that you would know everybody in the organiza
tion in these cases there are dozens or hundreds or even thousands of critical applications necessary to reach the Strategic objectives of your organization in these settings you are likely part of a very large group of business analysts and if you're just starting out on your career you get the advantage of being able to call on the senior players for advice and guidance now no matter where you are on the Spectrum what I have to say should be useful to you in those larger organizations a deeper
understanding of data and its implications for organizational Health are generally better understood you will be under some pressure to get up to speed but you'll be working with others who can help you along for smaller organizations the scope of your projects may be quite wide and quite deep as an example you may be called upon to manage the project for which you have developed the necessary requirements or in other cases you may even be called upon to do some data Engineering in order to val
idate your requirements now there are many relevant books on today's topic and all are widely available I earlier mentioned Tom Davenport this perk by Davenport and Harris was first published in 2007. it brought the the ideas already in use by major companies into the mainstream business press and popularized the Notions that are appropriate in the use of Analytics combined with solid business strategy and the ability to implement discoveries that give companies a Competitive Edge and this is no
w conventional wisdom in most large organizations reading this book or other more recent studies will help you to solidify your understanding of the data science world now this is the fine lady I think of when I query chat GPT I invite you to ask her any and all questions about your work if you frame your questions well you'll be amazed at how much useful and highly organized materials she can provide I agree with many of the commentators who say that chat GPT and other public AI services will t
ake over much of the grunt work that consumes a large chunk of your day so to get ahead of the pack or maybe just to keep your job I'd suggest that you start today begin to investigate the numerous ways that regenerative AI can enhance your productivity from my personal experience I uh I would say she's very good at blowing kisses your way so with the knowledge you've built up there are some things you should do to help your team of data scientists data Engineers data analysts and everyone else
you should collaborate with the team on defining critical business problems you should involve them early in the process you should make sure that they have important work you should ensure that they have full access to all the necessary data you can you should ensure that they're up to speed on how things really work in your organization and finally you need to stay involved and keep providing unnecessary feedback now whether you recognize it or not the data in your organization has a large imp
act on your role as a business analyst if you're not paying full attention to the role of data you can easily get into trouble not having responsible owners for the data is often a recipe for disaster data quality is a key to this success of both your organization and your personal future creating or aligning with an effective data strategy will help you to success in your projects with this in mind there's some things necessary to make it all work smoothly first make sure that your Communicatio
ns are regular complete and clear secondly you can assume a leadership role by modeling the behavior that you want others to use thirdly you can be sure to get everyone with a stake in the game to be at the table and fourth once they're gathered make sure that the necessary training for execution and implementation is defined and fully in place then as things develop make sure you're monitoring the progress and making all the necessary adjustments along the way and finally once you're done make
sure you recognize the people that made it all happen this of course is good advice for any project whether data science is critical or not so finally to wrap up my presentation here are five key things to remember data science and analytics are driving innovation organizations accumulate exponentially increasing volumes of the of data and it needs to be managed organizations that can harness their data are going to win data scientists work better in teams and they need your business leadership
so I'm done at this point I'd be happy to look at your questions and do what I can to answer them
Comments