Business Intelligence Tutorial

welcome to the technologyadvice business intelligence webinar we'll be going over all the essential components of business intelligence and looking at how it can help your company achieve greater efficiency we've got a lot to cover so let's dive right in the first question you probably have is what is business intelligence and essentially business intelligence refers to applications that transform data into meaningful information which helps businesses make better decisions and now we're going to go over a brief history of business intelligence the term business intelligence actually came into use around the 1950s and it grew out of technology called decision support systems this is actually a pretty fitting name as the technology is still used by companies today in order to support and enhance their decisions over the past few decades business intelligence systems have grown more powerful and comprehensive this is largely due to the increase in the amount of data that companies can easily can easily collect on their customers and on their own internal processes this is due to the rise in things like smartphones wearable devices the internet general computer usage etc also companies can store data cheaper and in greater quantities than ever before companies have access as I mentioned to vast troves of data in the form of smart phone metadata internet usage records social media activity etc business intelligence platforms can sift through this data to find patterns and trends among purchases or their own internal manufacturing processes by 2018 the business intelligence market is expected to be worth 20 point 1 billion now there are three main types of data they're structured data semi structured data and unstructured data unstructured data is information that can't be easily read by computers it includes text documents videos or really any kind of information that doesn't have isn't clearly sorted can't be organized into databases doesn't come and rows or columns or with a clear description of what it is and this is by far the most common form of data on the Internet a good rule of thumb is that 80% of all data produced is unstructured for instance when you when businesses collect information about Facebook usage any kind of messages any kind of comments on walls things like that are all unstructured data now structured data is data that resides in a fixed form and it's labeled so this could be name collection boxes on websites or the address fields for shipping information has a header you can put that into Excel and you can query it or search it with a computer you can analyze it so how do companies store and manage all this data then well company data isn't always in one location in fact it's rarely in one location instead it usually resides across different systems you'll have customer or lead information in a CRM program you might have information about how your marketing efforts are going and a marketing automation system you'll have information from your customers such as consumer sentiment or reviews in social media platforms and those are just a few of the places where you could find data so the first step for companies looking to implement business intelligence is in taking inventory of that data finding all the different sources and figuring out how you can analyze them all how you can cross-reference them one of the main ways the companies do this is data warehouses data warehouses are used to consolidate disparate data into a central location using a process known as extract transform and load ETL data warehouses standardize data across those different systems that I just mentioned they pull it all into one place and then you can query it you can ask questions and you can get those results pretty quickly there's also data marts now data Mart's are simply smaller more focused to warehouses and instead of taking all of the data across your entire company you might just say we're just going to take all the data that the marketing team produces put it in a data Mart it limits the complexity it's cheaper and it's a little bit easier to implement you don't need maybe a full IT staff you might just need one person and so you might be asking how does this information get to these central locations though and that's a good question the answer is extract transform and load usually and this is a again a method for standardizing and centralizing data the first step is extract the raw data is taken from a source program such as that CRM program or an enterprise resource planning software or you know through a plugin into you know Twitter so getting that raw data extracting it from that program and sometimes this is where any kind of unstructured data that you want to import into your warehouse you'll tag it with metadata or you'll put a description on it so that it's easier to find and query once it's in the warehouse the second step is transform this is when the data is normalized in order to properly analyze data it all needs to be in the same format think apples to apples so you need that in you know as in columns or at least in some kind of form that you'll be able to query and will return the same results finally load this is when data is transferred into the central warehouse or data Mart and this process can be done every week every day every hour it can be done pretty much in real time if you have the capacity the more often it's done the more up-to-date all of your analytics reports will be and when you run that report from from the interface of the business intelligence program that you're using you know it's accurate but it also the more often you do it the more network bandwidth that your company will need and this can eat up a lot of resources especially if you want to analyze a lot of data in real time now you might have heard the term a dupe and you might be wondering exactly what that is it's talked when people talk about big data and business intelligence Hadoop comes up a lot and Hadoop is essentially an infrastructure for storing and processing large sets of data across multiple servers in many ways a dupe is an alternative to these data warehouses instead of centralizing all the files Hadoop uses a cluster system it allows you to keep files on multiple servers so you can have some on one server some on another and you can issue one query that searches all of those different servers and brings back the answer it's a very flexible system it's inherently flexible because it uses a cluster system but that also makes it complex to implement and to run you're probably going to need a data scientist or at least some dedicate IT people who have a good understanding of these systems and it isn't well suited for just ad-hoc queries it's best suited for companies that produce massive volumes of data or working with incredibly large files Facebook uses a custom Hadoop array eBay uses a dupe as well they produce terabytes of data every few days MapReduce is the processing arm of dupe and MapReduce allows data to be queried and processed on the server where it resides instead of transporting the data across the network to be analyzed on the computer this is a pretty huge breakthrough as traditionally if you wanted to analyze data you had to get the data wherever it was being stored transfer it all the way across your network to your computer and the software would analyze it and then transfer it back with a dupe only the question the query is transferred across your network all of the analysis is actually done on the server then that information is reduced to just the answer and transferred and the answer is brought back to the computer so obviously the question itself and the answer take up a lot less space than the data so this can save huge amounts of network bandwidth and resources now let's talk about analyzing big data because regardless of how companies store normalize or process their data the analysis portion of business intelligence is what drives the field that's what companies are investing in and the insights from these analytics reports influence company direction product lineups even hiring decisions here are some of the key terms and concepts data mining sometimes called data discovery involves automated and semi-automated analysis of large sets of data in order to find patterns and correlations data mining can be used to group sets of data find outliers or draw connections between disparate data sources you could really argue that all of the intelligence in business intelligence software results from data mining it's sort of the foundation of all of these different types of analysis there's also text analytics text Analects software which is also sometimes called text mining homes through unstructured textual data like an e-book or research paper to find patterns this can be used for linguistic analysis or for sentiment analysis on social media posts or online customer feedback so if you want to track how your brand is mentioned across social media or you have a lot of customer reviews and you want to sort them to figure out the general tone text analytics is what you're going to need now business analytics is used to draw connections between data and companies use it to predict future trends gain competitive advantages and reveal unknown inefficiencies in their systems there's three main forms of analytics there's descriptive predictive and decision and the nice thing is that all of these names are very intuitive descriptive analytics essentially describes data that you already have and it's the base upon which other types of analytics are built so if you have if you take your internal data you've mapped where your data is coming from you've centralized it or you've implemented Hadoop or you have some way to analyze it you can use descriptive programs and software to look at all that data and identify the trends and relationships that are inside of it and it can be used to group raw data into easily digestible pieces such as the number of unique pageviews or the sales numbers for a specific department and that's going to allow you to really take a look at what your company's been doing if there's anything it's really inefficient or any unexpected correlations but you might also want to go a stage beyond that and say well that's what we've been doing this is where our problems are but what can we see about future patterns what can we extrapolate from past trends and that's where predictive analytics comes in predictive analytics searches for a correlation between a single unit or factor and features that pertain to it across different data sets and the goal is to find the same correlation across your different the different areas where you're gathering that data and that's going to allow you to predict future patterns this is probably the fastest growing form of analytics in terms of the features that you know legacy vendors are adding to their software and a lot of the startup companies right now in the field specialize in predictive analytics decision analytics is right on the cutting edge it's been around for a little while but it's kind of just gaining a lot of popularity and this software builds on predictive and it helps companies make decisions it not only analyzes your past data and then extrapolates that into future trends but it looks at external conditions so you can load information such as manufacturing trends or what the market is going to be like in a few years or the predicted shortages of resources maybe that you need to manufacture something and then it will recommend the best course of action for your company so it's looking at your data external data of the market and then trying to map out the safest path for that to work you really need to have good clean data as you're dealing with a lot of unknowns so now that we've looked at the different ways you can analyze data and the different analytics programs that are out there let's look at what you can do with that data once you've analyzed it or how you can turn that data essentially into presentations powerpoints how you can report how you can report business intelligence the main way that you can do this is through data visualization and data visualization is one of the big growing fields in BI right now it's essentially just a graphic display of the results of your data mining efforts or your analytics you know the queries that you've run and this can actually update in real time these are different charts or graphs that come out of these queries if you have the bandwidth that can all update in real time and that'll be through a dashboard now dashboards are subsets of visualization and they are the interfaces that represent specific analysis so you can see the screenshot we have of good data and their executive overview dashboard and these kind of things will just show you you know how many qualified leads you brought in who your top salespeople are maybe the new opportunities that you found and you know where you found them and often you can drill down into that so you can click on a salesperson see where their leads are coming from click on the leads see what regions of the country or you know what sources what industries they're in and so it's a very powerful tool but they're also very easy to interact with and they're particularly helpful for those who don't wish to interact with software through essentially a command-line interface or who don't want to write queries themselves and look at the raw data return you know this puts it into charts pie charts graphs and it makes it so the business users can really start to interact with the data you're bringing in and can sort of analyze it on their own and draw their own conclusions from it without having to consult with an IT department or a data scientist every time that they want to get these results so that's very powerful and we see it as a huge mover right now and hugely important in what companies are looking for in current business intelligence software it's really driving a big part of the market which brings us to the state of the market where as business intelligence headed and who are the main companies our market research shows that more companies than ever before are implementing bi programs and according to the latest data the majority view it as a business opportunity they see the potential however getting access to clean high-quality data remains difficult for some companies and of course that's always a concern when you're using business intelligence software as high quality clean data essentially data that has been normalized properly that you've stripped it of any errors whether that just be typos or misplaced decimal points etc you need to make sure that everything is in line because when you run those queries if there are mistakes in the data are there are errors somewhere in your data warehouse or in your spreadsheets the results that you will at the best just be incomplete you won't be seeing the whole picture at worst you'll be getting a result that's wrong or that will lead you to the to wrong conclusion and when you're predicting future trends you really need the base of those predictions to be as accurate as possible so let's look at TD W is latest survey on business intelligence of the companies they surveyed 57% had already standardized one or more bi applications throughout their business so that's huge that's a large jump and only 38% of companies said that they weren't using business intelligence at all so that's the Varia minority amount of businesses are not using it at all and I think they'll continue to shrink as the software becomes more available and as consumerization consumerization makes it friendlier for end-users a whole 89% of companies see big data and business intelligence is an opportunity there's a lot of market enthusiasm right now only 11% see it as a problem that's something of an outlier I think that will shrink a different survey by information week give some insight into why companies are adopting business intelligence 44% list predicting customer behavior as the driving factor behind their interest that is a very common one that we've seen as well when we've been advising people and working with them to find business intelligence solutions predicting customer behavior is huge and you can do that through analyzing their interactions on social media looking at if they're opening your emails you know what form of communication is getting to them what their purchasing habits are and then you can use that data to extrapolate what you should do moving forward new kinds of tactics that might be successful so there's a lot of potential there for optimizing your selling marketing strategies 58% of respondents here said that accessing timely reliable data stood is their biggest obstacle it's not surprising to you know as we talked about to get that reliable data you do need something of a team or at least someone who's going to go through normalize it you know the extract transform load process make sure it goes well get it into some kind of data warehouse or get a hadoop system set up if it's a distributed system and then you're going to need the bandwidth to access that and refresh your answers and your queries you know every day or at least every week or month depending on the industry are in and sixty-seven percent of respondents to information week said they're interested in using VI applications so again that matches up pretty well if the TD WI the overwhelming majority of companies are very interested in this technology there's a lot of enthusiasm in the marketplace and which brings us to the current trends one of the big ones right now is in-memory processing and in memory processing uses RAM memory solid-state memory instead of traditional hard drives or spinning disks to execute search queries and RAM memory is a lot faster and this vastly increases application performance doesn't take as long to query a lot of files doesn't take as long to store them and move them and as the main barrier to using solid-state memory or RAM memory has been how expensive it is and getting enough of it to store the massive amounts of data the companies bring in as solid-state memory prices continue to drop we expect this will become the standard way to store and process queries second usability and visualization we talked about the move towards greater usability through dashboards and business intelligence interfaces and essentially business intelligence software is making its way out of the IT department visualization dashboards and intuitive user interface design are allowing these programs to be used across departments and the companies that implement user-friendly software throughout their workplace that let their end-users you know explore their data dive into these dashboards look for trends they're going to benefit massively from the insights produced from these end users because they're going to be looking at things in a way that a data scientist or traditional IT department might not be looking at the problems and they're going to be able to collaborate in a data-driven way now let's look an example of business intelligence in action Oh to Ireland which is a large cellphone carrier and a telefónica Europe subsidiary so what's happening with o2 Ireland was that they notice that a lot of their customers were buying prepaid SIM cards for their mobile phones and leaving Ireland a few days later and never purchasing from them again usually these are businesspeople going on a short business trip and OH to Ireland didn't want to spend valuable marketing resources on these customers because they had almost no they presented no opportunity for repeat business it was a one-time transaction but the company didn't have the ability to identify or predict who these customers were what the habits were you know what the identifying markers were so that they didn't target them in their marketing efforts their solution was they contracted teradata to centralize their information - warehouse - you know map out where Oh - Ireland's data was coming from where they're getting information on the customers extract transform load all that standardized it get into a data warehouse then they implemented Cognos business intelligence software so they could run queries on that analyze it and using the centralized data and their new business intelligence system Oh - Ireland was able to identify in segments the 65% of their customers who stayed in Ireland after purchasing SIM cards and who are likely to be repeat customers and so that resulted in in a much higher return on investment for them as they were no longer wasting resources on customers who had no intention of using their services past a few days so hopefully that gives you a good idea of how business intelligence can be used in enterprise environments of course it's just one example but we do have additional case studies and vendor reviews on technologyadvice.com thank you for watching our business intelligence webinar join us next week may 28 2010 experts from companies such as looker and board analytics among others and make sure to visit technologyadvice.com you can get a free copy of this guide on our website you can see reviews of some of the vendors that we've mentioned here and you can find additional case studies on our blog thanks again for watching

Channel: TechnologyAdvice

Views: 221,544

Keywords: Business Intelligence (Industry), Business Intelligence Tools (Industry), Business Intelligence Tutorial for beginners, business intelligence concepts, business intelligence analyst, business intelligence tools, business intelligence dashboard, business intelligence and business analytics, big data, technologyadvice, technology, advice, Software (Industry), product review, expert opinion, information technology

Id: jkCCnwvO_fg

Length: 26min 3sec (1563 seconds)

Published: Fri Jul 11 2014