Group items tagged

Integration BI infrastructure: All tools in the platform use the same security, metadata, administration, portal integration, object model and query engine, and should share the same look and feel. Metadata management: Tools should leverage the same metadata, and the tools should provide a robust way to search, capture, store, reuse and publish metadata objects, such as dimensions, hierarchies, measures, performance metrics and report layout objects. Development tools: The platform should provide a set of programmatic and visual tools, coupled with a software developer's kit for creating analytic applications, integrating them into a business process, and/or embedding them in another application. Collaboration: Enables users to share and discuss information and analytic content, and/or to manage hierarchies and metrics via discussion threads, chat and annotations.

Information Delivery Reporting: Provides the ability to create formatted and interactive reports, with or without parameters, with highly scalable distribution and scheduling capabilities. Dashboards: Includes the ability to publish Web-based or mobile reports with intuitive interactive displays that indicate the state of a performance metric compared with a goal or target value. Increasingly, dashboards are used to disseminate real-time data from operational applications, or in conjunction with a complex-event processing engine. Ad hoc query: Enables users to ask their own questions of the data, without relying on IT to create a report. In particular, the tools must have a robust semantic layer to enable users to navigate available data sources. Microsoft Office integration: Sometimes, Microsoft Office (particularly Excel) acts as the reporting or analytics client. In these cases, it is vital that the tool provides integration with Microsoft Office, including support for document and presentation formats, formulas, data "refreshes" and pivot tables. Advanced integration includes cell locking and write-back. Search-based BI: Applies a search index to structured and unstructured data sources and maps them into a classification structure of dimensions and measures that users can easily navigate and explore using a search interface. Mobile BI: Enables organizations to deliver analytic content to mobile devices in a publishing and/or interactive mode, and takes advantage of the mobile client's location awareness.

Analysis Online analytical processing (OLAP): Enables users to analyze data with fast query and calculation performance, enabling a style of analysis known as "slicing and dicing." Users are able to navigate multidimensional drill paths. They also have the ability to write back values to a proprietary database for planning and "what if" modeling purposes. This capability could span a variety of data architectures (such as relational or multidimensional) and storage architectures (such as disk-based or in-memory). Interactive visualization: Gives users the ability to display numerous aspects of the data more efficiently by using interactive pictures and charts, instead of rows and columns. Predictive modeling and data mining: Enables organizations to classify categorical variables, and to estimate continuous variables using mathematical algorithms. Scorecards: These take the metrics displayed in a dashboard a step further by applying them to a strategy map that aligns key performance indicators (KPIs) with a strategic objective. Prescriptive modeling, simulation and optimization: Supports decision making by enabling organizations to select the correct value of a variable based on a set of constraints for deterministic processes, and by modeling outcomes for stochastic processes.

The balanced scorecard is one popular methodology for linking strategy, technology, and performance management. Other methodologies, such as applied information economics, combine statistical analysis, portfolio theory, and decision science in order to help firms calculate the economic value of better information. Whether you use a published methodology or develop your own approach in-house, the important point is to make sure your BI activities are keyed to generating real business value, not merely creating pretty, but useless, dashboards and reports.

Mistake 1.  Not putting a strong interdisciplinary team together. It is impossible to put together an analytics platform without understanding the needs of the customers who will use it.  Sounds simple, right?  Who wouldn’t do that?  You’d be surprised how many analytics projects are wrapped up by IT because “they think” they know the customer needs.  Not assembling the right team is clearly the biggest mistake companies make.  Many times what is on your mind (and if you’re an IT person willing to admit it) is that you are considering converting all those favorite company reports.  Your goal should not be that.  Your goal is to create a system—human engineered with customers, financial people, IT folks, analysts, and others—that give people new and exciting ways to look at information.  It should give you new insights. New competitive information.  If you don’t get the right team put together, you’ll find someone longing for the good old days and their old dusty reports.  Or worse yet, still finding ways to generate those old dusty reports. Mistake 2.  Not having the right talent to design, build, run and update your analytics system.  It is undeniable that there is now high demand for business analytics specialists.  There are not a lot of them out there that really know what to do unless they’ve been burned a few times and have survived and then built successful BA systems.  This is reflected by the fact you see so many analytics vendors offer, or often recommend, third-party consulting and training to help the organization develop their business analytic skills.  Work hard to build a three-way partnership between the vendor, your own team, and an implementation partner.  If you develop those relationships, risk of failure goes way down.

Mistake 3.  Putting the wrong kind of analyst or designer on the project. This is somewhat related to Mistake 2 but with some subtle differences.  People have different skillsets so you need to make sure the person you’re considering to put on the project is the right “kind.”  For example, when you put the design together you need both drill-down and summary models.  Both have different types of users.  Does this person know how to do both?  Or, for example, inexperience in an analyst might lead to them believing vendor claims and not be able to verify them as to functionality or time to implement. Mistake 4.  Not understanding how clean the data is you are getting and the time frame to get it clean.  Profile your data to understand the quality of your source data.  This will allow you to adjust your system accordingly to compensate for some of those issues or more importantly push data fixes to your source systems.  Ensure high quality data or your risk upsetting your customers.  If you don’t have a good understanding of the quality of your data, you could easily find yourself way behind schedule even though the actual analytics and business intelligence framework you are building is coming along fine. Mistake 5.  Picking the wrong tools.  How often do organizations buy software tools that just sit on the shelve?  This often comes from management rushing into a quick decision based on a few demos they have seen.  Picking the right analytics tools requires an in-depth understanding of your requirements as well as the strengths and weaknesses of the tools you are evaluating.  The best way to achieve this understanding is by getting an unbiased implementation partner to build a proof of concept with a subset of your own data and prove out the functionality of the tools you are considering. Bottom Line.  Think things through carefully. Make sure you put the right team together.  Have a data cleansing plan.  If the hype sounds too good to be true—have someone prove it to you.

Schema on write This is the traditional approach for Business Intelligence. A model, often dimensional, is built as part of the design process. This model is an abstraction of the complexity of the underlying systems, put in business terms. The purpose of the model is to allow the business users to interrogate the data in a way they understand.

The model is instantiated through physical database tables and the date is loaded through an ETL (extract, transform and load) process that takes data from one or more source systems and transforms it to fit the model, then loads it into the model.

The key thing is that the model is determined before the data is finally written and the users are very much guided or driven by the model in how they query the data and what results they can get from the system. The designer must anticipate the queries and requests in advance of the user asking the questions.

How Do I Learn to Ride a Bike if I Don’t Own a Bike? I get it.  A lot of businesses went out and bought the bike.  In fact, some of them bought a bike that would be the envy of a Tour de France cyclist.  Now, they’re trying to learn how to ride it.

Hey, that’s fine.  Smart business people don’t live a linear life.  You get the tools and implement the tools – all while you’re learning the tools. 

There’s no “on button” for Big Data.  You need people who will put their hands into it, manipulate it and find the valuable insights.

Summary: Check Search whole cells and uncheck Regular expressions

Great design, high-quality code, strong business sponsorship, accurate requirements, good project management, and thorough testing are some of the obvious requirements for successful analytics systems.

As a professional in the field, you must be able to do these things well because they form the foundation of a good analytics implementation.

Successful analytics professionals should follow a set of guiding principles.

The greatest benefit of columnar is to avoid disk I/O.  By choosing “select *”, you run the risk of losing that benefit. 

BI tools are here to stay, and they really help make visualization of analytics easy.  When working with Infobright, always take an extra second to review the generated queries.  The extra few seconds could mean seconds or minutes in saved query times.

You cannot accept problems as handed to you in the business environment. Never allow yourself to be the analyst to whom problems are “thrown over the fence.” Engage with the people whose challenges you’re tackling to make sure you’re solving the right problem. Learn the business’s processes and the data that’s generated and saved. Learn how folks are handling the problem now, and what metrics they use (or ignore) to gauge success.

Solve the correct, yet often misrepresented, problem. This is something no mathematical model will ever say to you. No mathematical model can ever say, “Hey, good job formulating this optimization model, but I think you should take a step back and change your business a little instead.” And that leads me to my next point: Learn how to communicate.

In today’s business environment, it is often unacceptable to be skilled at only one thing. Data scientists are expected to be polyglots who understand math, code, and the plain-speak (or sports analogy-ridden speak . . . ugh) of business. And the only way to get good at speaking to other folks, just like the only way to get good at math, is through practice.

Big Data has been all the rage. But fast data, even if it is small, can be more valuable than complicated masses of information.

Here's why: All the focus on "bigger is better" has overlooked the fact that most Big Data segments have not been validated with a business application or value.

Those kinds of analytics can help you find the right streams to access and work with, and also can help you build out robust programs that identify valuable customers.

J4L FO Designer 1.6.3