News and Views: Drive Smart Decisions with Cloud Analytics, Machine Learning and More

Recent Posts

Data Visualization

Understand User Behavior Using Oracle Analytics

How would you like to know what your software users are thinking and how they might react in the future? Oracle Analytics Cloud supports the accumulation of Usage Tracking (UT) statistics that help in monitoring system usage and performance to understand and predict user behavior. Knowing in advance what choices a user is likely to make helps increase efficiency and reduce errors. These statistics can be used in a variety of ways such as in system or database optimization, aggregation strategies, or internal billing of users/departments based on the resources that they consume. Usage tracking is particularly useful in determining user queries that are creating performance bottlenecks based on query frequency and response time.  Previous versions of Oracle Business Intelligence Server can track usage at a detailed query level. When Usage Tracking is enabled, it collects data records for every query that is executed and writes them all to database tables. Both logical and physical queries are tracked and logged in separate tables, along with various performance measures:  The following video gives a high-level overview of how to set up Usage Tracking on your Oracle Analytics Cloud environment.     Prepare to Enable Usage Tracking We're going to dive into this process, but first you will need to meet the following conditions to enable Usage Tracking in your environment: Usage Tracking requires a full metadata repository database (RPD) to be enabled. It will track usage on any queries, even outside the RPD, but it only requires an RPD for its configuration. For example, UT cannot be enabled when only a Thin Client Modeler (TCM) such as a web-based admin-tool model is active on your Oracle Analytics Cloud. UT requires access to a database to write data back into tables. A user login with create table privileges on the database schema will be needed during the configuration. Configure Usage Tracking UT configuration involves the following two steps:  Database connection in RPD: Using Administrator tool client (the editor for the RPD repository), define a database connection in the physical layer of your RPD. This connection should point to a database where Usage Tracking tables will be created and maintained. This database can be anywhere on the cloud as long as it is accessible for write access by Oracle Analytics Cloud.  System Settings in Oracle Analytics Cloud: Configure Usage Tracking parameters, such as connection pool name, physical query table name, and logical query table name.   Define a Database Connection in RPD Open the RPD which is uploaded in Oracle Analytics Cloud and create a new database in the physical layer. Provide an appropriate name (e.g., UsageTracking) and choose the database type as Oracle 12c. Under this database, create a new Connection Pool with an appropriate name (e.g., UTConnectionPool). Provide the connection details to the database where UT is to be configured and login credentials to the schema.  Note: This database user needs to have table privileges on the schema.    Next, create a physical schema in the RPD with the same name as the database schema to be used by UT (e.g., UT_demo).  Once these definitions are complete, the physical layer should look like this.    Save the RPD and upload it onto Oracle Analytics Cloud using the Replace Data Model option within the Service Console. System Configuration On Oracle Analytics Cloud, open the System settings screen from the Console Tab.  Scroll down and modify the following properties. 1) Toggle on the Enable Usage Tracking button 2) Usage Tracking Connection Pool: Enter the connection pool name in the format: <Database>.<ConnectionPool>  Example: UsageTracking.UTConnectionPool 3) Usage Tracking Physical Query Logging Table: This is the table where details about the physical queries executed by Business Intelligence Server against the underlying database are tracked. Enter the name in the format:  <Database>.<Schema>.<Table> Example: UsageTracking.UT_demo.PhysicalQueries 4) Usage Tracking Logical Query Logging Table: This is the table where details about the logical SQLs executed by Business Intelligence Server are tracked. Enter the name in the format: <Database>.<Schema>.<Table> Example: UsageTracking.UT_demo.LogicalQueries 5) Usage Tracking Max Rows : Specifies the number of rows after which the UT tables will be purged. The default value is 0 which implies there is no purging. You can set it to any desired value and the records will be purged once that threshold is reached. After entering these details, restart Oracle Business Intelligence Server. Once the server is restarted, Usage Tracking is enabled, and the two UT tables are created in the database. UT Tables Explained Open SQL Developer, log in to the UT database, and observe that the two tables have been created.   To generate some usage tracking data, log in to Oracle Analytics Cloud and click around some Data Visualization (DV) projects, both Extended Subject Areas (XSA) as well as subject area based. Also, open the BI classic home and open a few dashboards to generate some queries. Now observe that the tables for logical and physical queries are populated with usage tracking information.     Important Columns in Logical Queries Table END_TS End time of query execution ERROR_TEXT Error message if the query has errored out ID Primary key QUERY_TEXT Actual query text RESP_TIME_SEC Total response time of query in seconds ROW_COUNT Number of rows returned by the query SAW_DASHBOARD Dashboard name where query is getting generated SESSION_ID Session ID from the user firing the query START_TS Start time of the query execution SUBJECT_AREA_NAME   Subject area used for the query USER_NAME User ID who has executed the query     Important Columns in Physical Queries Table ID Primary key LOGICAL_QUERY_ID  Foreign key from the logical queries table QUERY_TEXT Query text TIME_SEC Time taken by query to complete ROW_COUNT Number of rows returned by the query START_TS Start time of the query END_TS End time of the query   The join between the 2 tables can be performed using the join condition:      LogicalQueries.ID  =  PhysicalQueries.Logical_Query_ID.    Note that not all logical queries will generate a corresponding physical query. For example, some logical queries may hit cache and return results from cache. When this happens, logical queries will not generate a physical query. Analyzing UT Data Once UT is enabled, the system usage can be analyzed from DV. In order to do this, create a DV connection to the UT database, create datasets for the Physical Queries and Logical Queries tables and analyze them within a DV project.  Following is a sample analysis built on the UT tables that shows number of sessions, number of queries, most frequently used subject areas, most frequently used dashboards, and so on.  Conclusion Usage Tracking provides a mechanism for administrators to keep track of the usage of the Oracle Analytics Cloud system. These statistics can be leveraged to take decisions to scale up, scale down, restrict access during certain time periods, pause/resume the system, and so on. To learn how you can benefit from the latest features in Oracle Analytics, visit Oracle.com/Analytics, and don't forget to subscribe to the Oracle Analytics Advantage blog and get the latest posts sent to your inbox.

How would you like to know what your software users are thinking and how they might react in the future? Oracle Analytics Cloud supports the accumulation of Usage Tracking (UT) statistics that help in...


Learn Modern Data Visualization with Oracle Analytics

Are you a business analyst curious about what Oracle Analytics can do? We recommend a new online course designed to provide you with the essentials of augmented data visualization. It's called Modern Data Visualization with Oracle Analytics, and you can find it online here. If you want to check it out, please enroll—it's free. I know it's good because I volunteered to help Product Management build this online series as a Massive Open Online Course (MOOC) for Oracle Analytics. What will you learn? In this course, you will broaden your understanding of modern—or augmented—data visualization concepts through hands-on training with Oracle Analytics. And we designed this course so that you can jump right into a technical, hands-on product experience. You know how sometimes you volunteer for something at work and then almost immediately regret it? This isn't one of those times; in fact, quite the opposite. And why, you might ask, would you care that I have trouble turning down new work projects? Because you get something really useful out of it--a solid hand-on product tour of Oracle Analytics for analysts and business people. It was an intense project to get it all done and published, but as we built it, we came together as a team. I also learned a lot about MOOCs and now have a shiny new skill myself. Then the enrollments started coming in, and we're having a great time seeing who is taking the course, and what feedback they're sharing with us. We're creating a community, and that's so cool. I'd like to invite you to be part of that. Now down to business. Why a MOOC? Why this particular MOOC? How does it work? What's in it? Answering the "why" question is simple. With the breakneck pace of change in the analytics world, you need access to training that's self-paced, accessible, low cost, detailed, scalable, hands-on, and available anywhere: Helloooo, MOOC on Udemy.com. "Why this particular MOOC" became a subject of some heated debate, as we all realized that if we included everything we wanted, it would be far too long. Many discussions were had to finally nail down what topics to include.  We hope we hit the nail on the head with topics ranging from machine learning-driven visualizations, data blending and mash-ups, augmented data enrichment, advanced maps and charts, and how to narrate and present your analyses. "How it works" is simple as well. Just click here and enroll.   And look what's included in the course: 4 hours of learning videos and screencasts 45 lectures 5 business use cases 4 bonus projects Also, did I mention? It's free! We're targeting this first course for analysts and business people who want to see what Oracle Analytics can do first-hand. Every section is packed with both video and screencasts to showcase each analytics capability. There are also demo files and scripts to download so that you can try it yourself, for real, with the actual product, no marketing fluff. You'll master everything from built-in functions for advanced visualizations, machine learning-based visualizations, and how to collaborate and share your discoveries. Check out the course outline for the full list of topics. Since we can't guess which use case you're itching to try out, we've packed the course with different projects like sales analysis, school donation analysis, HR attrition analysis, as well as advanced projects such as machine learning models for predictive analysis. Curious about some other application for analytics? You can try it with your data too. I'll freely admit the full course is a time commitment. To help you get through it, we've split it up so you can do it in sections. Plus, Udemy has this terrific "offline" capability, perfect for long airplane business trips. Just think how well-honed your analytics chops will be once you're done. I hope you enjoy the course and gain many useful skills from it. This was a fun course to build, in a slightly insane kind of way, and I'm looking forward to creating more content in the future. Once you get started, let us know what you think and how we might improve Modern Data Visualization with Oracle Analytics going forward and visit the Oracle Analytics Cloud website.

Are you a business analyst curious about what Oracle Analytics can do? We recommend a new online course designed to provide you with the essentials of augmented data visualization. It's called Modern...


Analytics Helps Power HR Transformation

Human Resources managers often have a difficult time finding, keeping, and developing talented employees, but the tools they can now use to transform their business have never been more enabling, according to a principal with advisory services firm KPMG. While HR managers use their usual toolsets for employee acquisition and retention, data analytics tools are helping companies keep on top of market trends, discover underlying factors for worker success, and understand how future worker opportunities can be developed into revenue streams.  We asked KPMG's Todd Randolph to join us for a discussion about improving business through analytics. Randolph is a huge proponent of analytics in HR and has written several articles on the subject. Feeling the drive to be competitive, many organizations are moving their core transactional systems to the cloud. In the HR department, KPMG’s 2017 HR Transformation Survey found that 75 percent of surveyed companies reported a successful implementation of cloud technology after undertaking HR transformation. Randolph says it all starts by understanding leading practice HR processes. "Our group does a lot of what we consider HR analytics or human capital management (HCM) analytics," Randolph says. "This involves exploring ways to improve the overall employee experience using analytics. And what we see is the idea of the whole, e.g., self-service or directed analytics. With self-service, it's giving employees the opportunity to help themselves. Rather than waiting for someone to pull data from a number of different data sources to deliver a report, you allow people to go out to a dashboard or even be sent information on their cell phone or mobile device, greatly improving the employee experience. We advise workers or managers and even executives to know what they should be looking at. So, they are reviewing data from areas where the company needs to focus and can take meaningful actions." Often times, Randolph says companies start their analytics journey with a dashboard supplying data from a source or sources that are transparent to them. They may review key data points, compare competitive information, and be guided to indicators that recommend a change or that action be taken. If action is necessary, the user should be seamlessly routed to the transaction system to initiate or approve the required action.     Randolph says that new technologies like Oracle Analytics Cloud are driving his clients to look at how they're using data, specifically in three major categories: embedded analytics, strategic analytics, and advanced analytics. With embedded analytics, Randolph says the business decision to move to the cloud allows embedded analytics to thrive outside the IT department and become a real competitive advantage. "It creates an opportunity to start looking at how you pull data together into a single source of truth for operational reporting," Randolph says. "Strategic analytics—with maybe a product like Oracle Analytics Cloud where we can have that single source of truth pull data from multiple data sources—gives executives and managers the opportunity to look at the organization as a whole from a reporting and analytics perspective. We're also seeing a lot more advanced analytics over the last few years as well as the tools are getting better and easier to use, giving people who are not necessarily data scientists the ability to create some more advanced analytic functions or embedding things like artificial intelligence and other advanced functions." By using embedded analytics, strategic analytics, and advanced analytics, companies can avoid keeping information separate (in silos) and even develop more data transparency. This can include transactional data that can be reviewed, analyzed, reported, and shared so that the data helps validate the right business decision in a very intuitive way. As Randolph explains, "You go into an organization because there’s an issue or some sort of problem.” When that’s about reporting or analytics, it often involves either moving to the cloud or some other ERP type implementation. The client knows they're having issues or can get better at doing their reporting and analytics function.  A lot of times, we find that the problem is siloes. Even within one function such as HR (with its sub functions such as recruiting, payroll, and so on), things can be extremely siloed. You can look across finance or supply chain or any other parts of the organization and see the same thing. There are often different source systems, and you have someone manually pulling data from all these solutions to try to get to that single source of the truth. Then it's typically offline, with people developing some sort of Excel and/or PowerPoint reporting and then delivering that out to stakeholders. So much of what we do is streamlining the process of getting to that single source of truth for our clients.” To hear the conversation with KPMG's Todd Randolph in its entirety, click on the photo below to play the podcast. To find out more about how Oracle Analytics Cloud can best complement your company's business process strategy, visit our website.

Human Resources managers often have a difficult time finding, keeping, and developing talented employees, but the tools they can now use to transform their business have never been more...


Solving the World Bee Crisis with Oracle Analytics

Bees and other insects that pollinate flowers are being driven toward extinction, according to the World Bee Project (WBP), so the global community launched a new initiative in partnership with Oracle using analytics to help slow the decline. At the "Nature of Data" customer event during Oracle OpenWorld, WBP Founder Sabiha Malik discussed the increasingly inhospitable conditions for bees, driven by loss of flower habitats, intensified farming methods, climate change, and increased use of pesticides. "The more we understand the relationships between pollination, food, and human well-being, the more we can do to protect bees and pollinators—and help protect our planet and ourselves," Malik said. Consider the following: Bees are responsible for pollinating one-third of the global food supply Of the 100 crop species that feed 90 percent of the world's population, bees pollinate 70 percent England's honey bees are vanishing faster than anywhere else in Europe, with a 54-percent decline between 1985 and 2005 "Imagine a farmer whose crops yield one-third less than they did before. That is what we are faced with if bee colonies continue to decline." “The World Bee Project Hive Network will remotely collect data using a network of connected beehives,” Malik added. The data will then be fed into an Oracle Analytics Cloud, which will use analytics tools, including artificial intelligence (AI) and data visualization, to give researchers new insights into the relationships between honey bees and their environments. WBP's partnership with Oracle and the University of Reading School of Agriculture, Policy, and Development (SAPD) will allow researchers to 'listen' to the honey bees and analyze intricate acoustic data captured inside the smart hives, including the movement of bees' wings and feet. Combined with other precision measurements—including temperature, humidity, and honey yield—researchers will be able to closely monitor bee colonies, detecting patterns and predicting behaviors. This will enable conservationists and bee keepers to act to protect colonies, such as preventing swarming at the wrong time of year or removing predators like the invasive Asian hornet. The WBP Hive Network launched in the United Kingdom with expansion expected in the United States and Africa next. The value of the data is in informing beekeepers of the various different states of the colony throughout the year to aid colony management. Check out the following video to see how the project came together. Are you ranking higher than your competition? They utilize data analytics--Should you? Take a quick 2-minute assessment and find out your results.

Bees and other insects that pollinate flowers are being driven toward extinction, according to the World Bee Project (WBP), so the global community launched a new initiative in partnership with Oracle...


7 Steps to Building an Actionable Marketing Dashboard

Every forward-thinking business today is focusing on being more data-driven. This holds true even for Marketing departments which are often assumed to be more creatively-focused. In fact, according to Forbes, businesses that have more data-driven marketing are up to six times more profitable year over year than business that have marketing teams that don't run by data. That's a huge incentive to care more about your data. But what good is gathering all that data without a way to compare and contrast the information and come up with valuable insights required to be competitive? "The North America Field Marketing team tackled this head on, we've moved from insights from weeks to seconds," says Dave Ewart, Director of Digital Marketing in Oracle's North America Field Marketing team. "We wanted to gain better visibility and understand the levers that run our business to help us make smarter decisions and more quickly optimize Marketing's investments to drive incremental revenue for the Sales organization." Even a company like Oracle encountered these marketing challenges when it came to becoming more data-centric around our lead-flow engine. First off, our traditional dashboards were limited and difficult to customize to meet different stakeholders' needs. Secondly, because of the limited customization functionality on our dashboards, the team ended up exporting massive amounts of data on a weekly basis, making our data reports up to seven days old. Finally, I relied on a team of up to six people to regularly pull lead reports for different stakeholders, making this effort manual, time-consuming and resource-heavy. Since having access to our business intelligence tools, Oracle Autonomous Analytics Cloud has completely changed the way our marketing organization looks at lead data. Our data is now visible in real-time, customizable and completely low-effort, freeing up human resources. Instead of constantly building reports, we can now extract more insights and inferences. "The beauty of these dashboards is that they equip us with predictive insights as to which marketing programs or channels are performing best—and therefore, should be actioned first by the Inbound BDCs," says Sami Halabi, Oracle Lead Development Specialist Here are the seven views we included in our MQL dashboard in order to make our marketing dashboards more actionable and provide better insights: Track lead volume over time Being able to track lead volume on an even day-to-day basis is crucial for our business. Daily lead volumes can dramatically impact a sales representative's productivity if either too many or too few leads are being sent over. Having this visibility helps marketing managers better pace their campaigns. Breakout lead channels Identifying which channels are driving the most leads - and the top converting leads—is great feedback for marketing. Having the insight to tell if the social media tactics are driving demand, or if our paid channels aren't optimized, helps marketers react more quickly and adjust tactics. Top programs sourcing leads Marketing programs can be costly, so being able to see which programs are hitting the mark and which ones aren't on a regular cadence gives marketers greater control to turn off initiatives that aren't working and double-down on the ones that are. Regions and cities with most leads Seeing where on the map most of our leads and top converting leads are coming from is not only interesting, it can inform where we may choose to host our next event or spend more of our advertising budget to target. Track converted leads over time Most marketers have pipeline and opportunity targets to meet throughout the fiscal year or quarter, so a regular pulse check on where we are towards hitting our goal is critical for running our business. Seeing trends over time can help us foresee if there are roadblocks ahead and help us course-correct. Track conversion rate by channel While lead volume is very important, equally important is understanding what channels are driving more high quality and high converting leads. Top converting channels and programs Finally, visibility into what channels and programs are sourcing the highest volume of converted leads and pipeline for our business is data every marketer should track regularly. Including spend data could also help us calculate ROI and prioritize our programs that way. This dashboard is just one of many marketing analytics dashboards we're building for our Field Marketing organization using  the data visualization capability in Oracle Analytics Cloud. While according to Econsultancy, only 33 percent of marketers have the right technologies for data collection and analysis, that's not the case at Oracle. Start your Oracle Analytics Cloud trial today and get one step closer to becoming a more data-centric marketer.   Guest author, Veronica French is a digital marketing manager with Oracle

Every forward-thinking business today is focusing on being more data-driven. This holds true even for Marketing departments which are often assumed to be more creatively-focused. In fact, according to F...


Oracle Analytics Cloud Gets a Visualization Tune-Up

The latest version of Oracle Autonomous Analytics Cloud includes a refreshed data visualization a capability that empowers business users to explore, discover and visualize their data.  This marks its fourth year since becoming generally available with Oracle's initial cloud analytics offerings (Business Intelligence Cloud Service and Data Visualization Cloud Service).  Oracle Analytics Cloud's data visualization feature addressed that primary shift to user self-service with a data-driven approach to creating compelling analytics.  Let's face it, data visualization in its most rudimentary form has become table stakes with many vendors (like Tableau, Qlik, Microsoft) offering equally good tools.  Every data visualization tool provides the basics of self-service.  Uses can connect their data sources and create great looking charts, or whatever their mood dictates that day.  There is virtually no compelling differentiation between the vendors on this level.  Oracle has taken the next step by heading into autonomous analytics.  Going beyond the common capabilities of centralized business intelligence provided by on-premises core BI systems, then going further beyond the now common self-service capabilities provided by current cloud analytics services.  Autonomous analytics provides capabilities that assist the end users with finding, compiling, and creating compelling stories powered by machine learning that then describes their findings.  Some of these autonomous capabilities have been released into the product already. So, what's coming in this latest version of Oracle data visualization capabilities and what's different from any other data visualization tool on the market?  Oracle Analytics Cloud release version 5 (v5) has many updates in every aspect of the product making it far more flexible with even more capabilities to customize and tune the visualizations to better fit your visual needs.  For instance, we introduce great enhancements to our mobile application, entitled "Day by Day."  We improved search capabilities and newly recognized query terms including top, bottom, first, last and most for queries like "Who were my top 10 performing customers last quarter?"  We also introduce new built-in visualizations that include dynamic heatmap, multiple data layers, 100 percent stacked bar and picto charts (as seen in the images below).                       Data Preparation Enhancements New with v5, the data flows section of Oracle data visualization capabilities introduces smart data types.  This autonomous capability allows the tool to recognize certain data types and intelligently make recommendations.  For example, if "city" appears in your dataset, other related attributes such as population, GPS location, county associated with that city will be recommended, thus autonomously enriching your data.  This data could comprise demographics, weather, and commodity prices that analysts would otherwise usually manually blend from their own personal data sources.  Another aspect is autonomous recommendations for obfuscation of sensitive data columns.  Credit card details for example, should never be display in clear text.  In some regions, this may breach privacy laws.  Therefore, these data columns like credit card numbers or social security numbers are detected and autonomous recommendations will be made to obfuscate those details (i.e., for credit cards) only the last 4 digits remain (as seen in the image below).  There are many such new smart data types available that not only enrich your overall dataset but by removing the otherwise tedious and manual process makes the overall time to decision faster, more efficient.  Along with Smart data types, functions are shipped to the most optimal place to execute them.  If you have a data lake then users can continue to create data flows to enrich their data from their big data source, but will experience optimal performance by having all processing executed by the most appropriate big data engine in the most appropriate location.  Incremental data processing means that existing data flows can be rerun on data sources and only new data will be processed and joined with existing data.  No need to completely rerun the whole process.  Data flow branching allows users to split their data into subsets that can produce a group of related output results or load that data to multiple different locations (as seen in the image below). Automatic Oracle application data replication provides the capability for data visualization to take copies of data sets automatically from your favourite Oracle cloud applications and create an analytic sandbox to enrich it and then perform your analytics without any disruption to the main application or the core operational data. I've only exposed the tip of the iceberg on the changes already set for Oracle Analytics Cloud release 5.  Check out the video below to see these changes in action.   Visit us for more information about Oracle Analytics Cloud.

The latest version of Oracle Autonomous Analytics Cloud includes a refreshed data visualization a capability that empowers business users to explore, discover and visualize their data.  This marks its...


Fuel Innovation with an Analytics Cloud

If you've ever had to chase down data for a report, you know it requires multiple steps, perhaps multiple trips to IT, and possibly multiple software programs to come to a single source of the truth. Successful data analytics doesn't need to be so complicated. It turns out that all you really need is a single, cloud-based platform, whether the data is in a spreadsheet or a data warehouse. "Most business people don't wake up one day and decide to do analytics, they have other jobs to do," says Matt Milella, VP of Development Oracle Analytics. "But they need analytics to do their jobs. We want to make the steps easier and easier, so you can take advantage of the really important information." Milella recently spoke during a webcast entitled, "Make Innovation Your Business with Oracle Analytics Cloud," which outlined the benefits of a cloud-based analytics platform that makes it easier to get to actionable insights. The presentation also included a demonstration of the latest version of Oracle Analytics Cloud. Historically, your path between all your relevant data and the valuable insights you are looking for has been full of obstacles such as outdated or complex data, blending disparate data sets, business modeling, aggregating the information, and publishing among your peers or even customers. Cloud-based analytics platforms overcome these obstacles by allowing for personal and proactive experiences thanks to data visualizations and improved data modeling capabilities. Cloud implementations for analytics platforms continue to outpace on premises business intelligence systems. Researchers estimate the market size will double from its current levels of $57.4 billion to $100.3 billion by the end of 2025. While North America leads the way in implementations, European and Asia-Pacific businesses are also building their analytics in the cloud to glean insight from social media, mobile devices, and other forms of unstructured data. And while business people have tried to take a do-it-yourself attitude with standalone desktop analytics programs to augment their spreadsheet reports, more successful organizations rely on cloud-based analytics platforms to inspect, prep, and deliver relevant data in real time. "The 'one version of the truth' dilemma still exists and it is really important that data sets can be accessed and curated to improve time to insight, deep pattern profiling," Milella says. Instead of multiple vendor solutions cobbled together, Milella suggests a single analytics platform is beneficial in three ways. First, you absolutely need a proactive and actionable analytics platform where the system automatically alerts you to changes, has context at every level, and is accessible and actionable. Second, the analytics platform needs to provide self-service insights. The user should have the freedom to explore with reliable data quality. And third, this platform should have embedded machine learning and advanced analytics that connect to all forms of data and allows for improved exploration and data discovery. "One of the benefits of an integrated analytics platform is that it does a lot of the heavy lifting for you," Milella says Register here to learn more how Oracle Analytics Cloud makes it easier than ever to use data so you can take the guesswork out of innovation.

If you've ever had to chase down data for a report, you know it requires multiple steps, perhaps multiple trips to IT, and possibly multiple software programs to come to a single source of the truth. Su...

Data Lab

Business Analytics' Next Big Thing: Inline Data Prep

Data preparation—also known as data enrichment—isn't new. In fact, I can almost guarantee that every analytics deployment out there has their users doing some kind of data preparation to support their visualizations.  They might be using a specialized tool, like Alteryx, or a bespoke home-grown process that leverages several tools, Microsoft Excel, coding, and sometimes even Microsoft Access. However, the fundamental problem stems from the fact that source data systems don't provide the information in a form to support the required analytics. Also, joining data from multiple sources is almost always required to gain a complete view.  Using multiple tools to extract, transform, enrich, and visualize the data introduces multiple points of potential failure and much greater potential to introduce human error. This results in numbers that remain somewhat inaccurate, stale, or inconsistent between different users or departments. Ultimately, this leaves management sceptical of reports created and they resort back to gut feel decisions to drive their business.  Not exactly a modern or data-driven approach. The trend today is to incorporate data preparation capabilities inline. Simply put, it involves a unified tool and interface to source the data, enrich it, and visualize it. As such, it removes the multiple tools, which reduces cost massively, and increases productivity, efficiency, and accuracy.  Numbers seen in the visualizations are traceable back to their source (data lineage), not lost in confusing Excel-based processing that only some users comprehend or could decipher.  Any analytics vendor worth noting is providing inline data preparation. Qlik has it as part of their most expensive package Qlik Sense, which requires you to contact them for their price—always a bad sign.  Tableau recently released Tableau Prep, a new tool launched this year and only comes with the most expensive option, Tableau Creator. Microsoft with Power BI openly acknowledges that this is the next big thing and are actively looking to invest in this space (i.e. they don't have anything, yet). For the longest time, Power BI users considered that Excel is the data preparation tool for Power BI. Granted, Excel is a pretty powerful data enrichment tool, but that just brings us back to all the issues Excel creates with bespoke home grown hard to manage processes.  Silo tools like Alteryx should begin shaking in their boots.  Who will they work with when all business intelligence tools have built-in data preparation capabilities.  So, what is the correct approach? Well, for one, we understand that our analytics should be a unified system from extraction through enrichment to visualization. And secondly, we don't want to be spending double the money to do something we most likely already have in place.  Only Oracle Autonomous Analytics Cloud, with embedded machine learning, provides access to varied data sources, enrichment and data preparation, along with visualization and mobile access in the same product.  Visit our Oracle Cloud Analytics site to learn more.

Data preparation—also known as data enrichment—isn't new. In fact, I can almost guarantee that every analytics deployment out there has their users doing some kind of data preparation to support their...


Take Business Analytics to the Next Level with Machine Learning

The convergence of big data, analytics, data science, and cloud is creating a need by business managers to optimize their investments with a comprehensive way to derive value from data. Successful companies know how to strategy that includes applying machine learning to their processes, so they can automate and speed their time to decision. For example, companies like yours use machine learning in the following ways: A large bank used machine learning to analyze its collection activities and learned it could eliminate more than 40 percent of customer calls with better outcomes. A global retailer used advanced machine learning to forecast customer demand cutting forecast error in half. A telecom company found that its machine learning yielded a 75x reduction in "false alarms" for churn and instead focused its resources on those truly at risk of leaving. To better illustrate the business applications of machine learning and how it affects you, Oracle sponsored a webcast entitled: "Where Will Machine Learning Take You? See Your Future with Oracle Analytics." Rich Clayton, VP Product Strategy, Oracle Analytics and Mike Lehmann, VP of Product Marketing for Oracle Big Data and Machine Learning lead you through the process, where you can hear how Oracle Analytics uses machine learning to help you understand more, faster and understand how to get started on that path immediately. Every industry has been transformed by applying machine learning to its data analytics strategy, including automotive, healthcare, media, energy, communications, and government. Likewise, ML can be applied to all business divisions such as HR, finance, sales, marketing, and IT. Whether it's investigating customer churn, text sentiment analysis, forecasting and modeling, data discovery and auditing, or transactional data extraction and transformation, Clayton notes that machine learning enables better business data visibility. "Traditionally what we see is that people not being able to work together," Clayton says. "What adding machine learning to Oracle Analytics Cloud does is ultimately help them organize their work, build, train and deploy these data models. It's a collaboration tool whose value is that it accelerates the process and allows different parts of the business to collaborate, giving you better quality and models for you to deploy." One of the barriers to has been the multiple layers that data must pass through before it is processed, and its value derived. There is an extraction layer, data blending, modeling, aggregating and publishing long before there is a discovery mode. Clayton argues that by automating and embedding machine learning allows for faster time to decision. "There is a foundational opportunity by taking some of these components and embedding it into a value chain from an analytics perspective," Clayton says. A typical finance department is routinely burdened by repeating a variance analysis process—a comparison between what is actual and what was forecast. It's a low-cognitive application that screams for assistance of machine learning, Clayton notes. "By embedding machine learning, finance can work faster and smarter and pick up only where the machine left off," Clayton says. To hear the entire conversation, register for the webcast and see where your future will go with machine learning as part of your data analytics strategy. For more information about Oracle Analytics Cloud, visit our website.

The convergence of big data, analytics, data science, and cloud is creating a need by business managers to optimize their investments with a comprehensive way to derive value from data. Successful...


Data Scientist Kirk Borne Discusses Business Impact of AI and ML

Adding artificial intelligence, machine learning, and other cognitive interactions to traditional business processes and applications enables greatly improved user experience and productivity. These technologies are already impacting all levels of business including finance, marketing, human resources, and sales. Proponents point to incremental changes in the way we work with AI and ML. From the auto-complete or autocorrect on your phone, to chatbots, and virtual assistants like Siri and Alexa. But larger implementations are also taking hold. The New York Metropolitan Transit Authority (MTA) is using machine learning to understand rider patterns and will be changing color coding of their maps to help commuters and tourists get to their destinations faster. However, these visible AI and ML projects are more the exception than the norm. Industry experts point to massive interest and investments over the next few years: 61 percent organizations surveyed by O'Reilly Media identify machine learning as the most significant data initiative for 2019 $57.6 billion in spending on AI and machine learning will happen by 2021 compared to $12 billion in 2017, estimates analysts with Deloitte. $3.5 to $5.8 trillion in potential annual AI-derived business value across 19 industries is very possible, according to the McKinsey Global Institute research paper. To help put these trends in perspective, we invited Kirk Borne to the Oracle Analytics Advantage podcast to discuss the value of AI and Machine Learning for business decision-making. Borne is often touted as a worldwide influencer on data science and one of the leading thinkers on the topic of large scientific databases and information systems. Borne currently consults clients for Booz Allen Hamilton. He's a researcher, blogger, data literacy advocate, TEDx speaker, and author of several books. He's even a project scientist for NASA as part of its Space Telescope Science Institute. During his interview, Borne explained that the successful evolution of AI and ML in business should be linked to solving the questions at hand "I like to remind people that machine learning is an algorithm that learns from experience—it detects and recognizes patterns in data," Borne says. "So, if you are classifying a disease or if you are classifying a customer, or information in your weblog, once you see that pattern, you can take the appropriate action." While the promise of AI and ML are encouraging, one strong barrier to adoption is a cultural one, according to Borne. "If you are going to have AI help you in decision making, that means letting go of some authority of your own," Borne noted with a nod to top executives in contrast to mid-level managers who may see AI and ML as a threat to their jobs. To help mitigate cultural changes, Borne advocates for a group of data experts instead of a single decision-maker when it comes to implementing AI and ML. "Whether the person is the chief data officer, chief data scientist or chief algorithm officer—which is a term I heard the other day—this person should have the view that people should be empowered to speak up when they see something in the data that can help improve the business they are empowered to bring that forward – it's not just the executive suite or the data scientists." Thankfully, many companies are already setting themselves up for success by adopting a cloud-based analytics infrastructure, which he says is needed to keep costs down.  "If all you need are a few minutes to process data and some of these major cloud providers are providing the tools that also use the cloud—then you don't have to incur any more costs," he says. Borne is also eager to change the meaning behind the acronym of AI. Whereas most people identify AI as artificial intelligence, Borne suggests businesses think of decision-making with the "new AI" where the letter A could mean Accelerated, Actionable, Adaptable, Amplified, Assisted, or even Augmented. "I think taking full advantage of artificial intelligence is a stretch even for the major companies that announce they are 'AI First,'" Borne says. "Even the ones that are fully using AI tell me they are still on the growth curve. Your average company has typically not even started with AI, which means there are a lot of opportunities to get started." To hear the full interview, check out the Oracle Analytics Advantage podcast by clicking on the podcast photo below and visit the Oracle Analytics Cloud website to see how to apply AI and machine learning to your data strategy.

Adding artificial intelligence, machine learning, and other cognitive interactions to traditional business processes and applications enables greatly improved user experience and productivity. These...