Good Business Intelligence will provide answers to your questions. Great Business Intelligence will answer the questions you wish you had asked. So how do you ensure that your Business Intelligence is great Business Intelligence? The first step is to develop a definitive picture of what you want the BI to deliver. Never lose track of core idea: BI should provide answers to your questions.
Initially, I would advise that you produce a list of dreams and desirables. Ignore whether data is actually available in this first step. A good Business Intelligence professional will often be able to derive information from existing data where it would seem impossible - let them decide what is achievable and what is not. It's extremely common for a BI implementation to highlight the benefit for new streams of data to be collected and stored.
Joe's Garage has expanded beyond all expectations through diligence and the use of Business Intelligence. As a result of this success, Joe has expanded the types of data he records and now, as well as customer and completed job records, the company also retains information on employee time keeping, job timings and costings, and a host of other areas of business. As more information is collected Joe becomes increasingly excited about new opportunities. Start here...
With wish-list in hand, confer with anyone who may have an insight into the information your company is storing. Now is the time to track down any information that anyone may be tracking on their personal spreadsheet etc, it can all be reported on and analysed, and can all be of benefit. If your company has Database Administrators, they are likely to be more aware of what data is available than anyone else. Most IT, administrative and management roles involve elements of data collection, recording and monitoring, so prepare to be surprised by the scope of your existing data.
With your increased knowledge it is at this point that bringing in a Business Intelligence consultant is highly recommended, as expert guidance is crucial in making the most of the opportunity. In my previous article 'blah blah blah' I discussed the options available for hiring in expert help, and gave some handy pointers on pricing and negotiating contracts.
Joe has an in depth look at the data he and his staff have gathered over the past few months and makes some rough notes on areas of interest. The main core of Joe's focus is staff efficiency and work standards. Whether you decide to use an in-house developer, or bring in a specialist consultant to execute your project, you will need a general understanding of reporting to be able to communicate your goals and requirements effectively.
Types of Reports
There are a few distinct categories of report styles, and the type of report you choose will determine how successfully the report will meet its objective. The following list describes the basic types, which will give you an idea of the features and advantages of each. Please note, there are no hard and fast rules, and reports can be a mix of these types tailored to your specific requirements:
* Detail Reports - Essentially a list of all the information of interest. These are great for identifying particular records or problems. Additional formatting can be used to highlight records that may be of interest.
* Summary Reports - These provide totalled summaries of the data that is of interest, such as hours spent on a particular task in a week. The range of options for the type of summary total is extensive - average, sum, percent, count - to name a few (Crystal Reports has nineteen different types of summary). Results of Summary Reports are often displayed in charts for clarity.
* Drilldown Reports - This is like a Summary Report, but has the added feature that when you click on any summary total, the report displays the detailed data that makes up that summary.
* Dashboards - These are a collection of Summary Reports, usually viewed in graph form with a minimum of numeric values displayed, and are designed to give the viewer an immediate overview of a particular set of related information. These options, either individually or used in conjunction, will meet virtually all business requirements no matter what the business sector.
Joe wants the freedom to look closely at all his data and so opts for drilldowns on all his reports. This allows him to investigate the cause of any summarised results that catch his attention. He particularly likes the idea of being able to compare employee work rates, and then look more closely into the data to see whether any patterns emerge.
Date Ranges
Every report must specify the time span from which data is to be reported. It is important to make the right choice, and very much worth taking the time to give it careful consideration. In a lot of cases your decision will depend on what the information is needed for. For example a report which shows data from the previous six months in summary, with the current month's data shown on a day by day chart and in summary, is an invaluable tool for a manager to track their progress daily during the month.
However, if you want to measure the performance of a third party provider, measuring in four week blocks rather than months (which vary in length) is the only way to view performance consistently. Another common requirement for this type of report is to differentiate between work carried out during a standard working day and that which is resolved "outside of hours".
Comparing seasonal sales can also be of great importance depending on the type of business you are in. The software used to create Business Intelligence reports is capable of producing any variation of date ranges imaginable, so some thought is required to take maximum advantage of all these possibilities. A good BI consultant will provide guidance and suggest viable options.
Frequency
How often do you want to view the information? This can also have an impact on the date ranges used for the report and should be considered when setting up a schedule for when reports are run. Initially Joe wants a one off set of reports to show how his company is currently performing. However, he plans on making several changes based this analysis and wants to monitor how they alter profits.
Joe re-runs the reports weekly for the first month to monitor the initial impact of his changes, and decides that a monthly run will give enough information thereafter.
Distribution
Once a report is developed it must be run to collect, manipulate and display the requested data, and the results communicated to the people who need to view it. This can be done as a manual process, running the report and then either saving it to a suitably accessible drive, or emailing to the intended recipients. However, specialist software is available for this purpose, and reports can be scheduled to run automatically at specified time interval, and results sent to recipients in a convenient format (usually Excel or PDF).
The size and scope of your project will determine whether specialist software is a worthwhile investment. I discuss Distribution in greater detail in later articles.
The Development Process
Business Intelligence implementations are a different animal to other IT development and should be treated as such. Traditionally, a design document is written to client specifications and once this is signed off as correct and complete, the software is developed and the project is considered finished. Any change to the software after this point incurs additional cost.
Developing BI reports however, should be a more fluid and iterative process. All efforts should be taken to identify what is needed before work commences, but the first draft of the developed report should be treated as just that, a first draft. Many consultants will try to set requirements in stone as early as possible. This is usually done with the best of intentions, but by accepting a rigid template from the outset, there is a high likelihood of a compromise in the quality and scope of the finished reports.
There are many factors to consider, some subjective, or which seem a good idea in theory, but in reality do not achieve the desired results. The result of Joe's initial meeting with Patrick generated more ideas for reports than Joe could realistically use or would be affordable in the short term. Joe is toying with the idea of staff bonuses and so decides to focus exclusively on the times associated with different jobs. This is to encompass the performance variations between jobs completed on different days of the week, morning or afternoon, employee undertaking the job.
After viewing the first draft reports and further discussion with Patrick, Joe decides to have the various information presented wit one report for ease of reference, but knows this is as much work as producing three separate reports.
When is development complete?
Ideally, a report should only be signed off once its target audience has viewed, and ideally used the report with real data. Minor requests are common at this stage. Often small adjustments are required to take account of factors which were not apparent at the time of the initial design discussion, but become obvious when viewing the report for real. For example, perhaps two columns which tend to be compared are illustrated at opposite ends of the report, thus requiring repositioning.
Many consultancies are not fond of this approach as it gives the customer the freedom to make changes their brief and requirements after work has already been done. Personally though, I find that if the original requirements analysis is done correctly, any alterations needed later are minimal and worth the effort to fine the report exactly to the customers needs. Patrick supplies a set of "prototype reports" which Joe looks over. Other than a couple small changes, the reports are exactly what Joe had envisaged.
However, after using them for a couple of days Joe decides that one report needs to be altered. Patrick looks at the report and they agree that it does match the initial requirements exactly. Joe is sure that, after viewing the other reports, he wants the information in a different format. He decides that he will keep the report in its original format as well, and so it is clear that the new format report is an extra requirement.
As this new report is derivative of the original it does not take a lot of work to develop, so Patrick is able to produce it at an affordable price.
So, in short...
To recap, here is a list of the basic steps involved in a Business Intelligence implementation.
Identify the Questions You Want Answering
Look Over Your Company Databases and See What You Have
Utilise Expert Help to Set a Solid and Agreed BI Implementation Scope
Evaluate the Developed Reports and Test Their Suitability
With All Alterations in Place, Sign-off the Implementation as Complete
And finally, reap the rewards of great Business Intelligence!
Friday, October 7, 2011
Wednesday, October 5, 2011
The Evolving Landscape of Business Intelligence
There will always be a need for business decision makers to access business data in order to understand what is going on in their organisations. The Business Intelligence (BI) tools available for this need have changed significantly within the last few years. This article looks at 3 issues that can be seen to be contributing towards the changing BI Landscape. First of all, there has been a shift of power from IT to the end business user. Secondly, there is a move from traditional BI to Business Discovery and lastly the changes in how businesses now actually operate and make decisions can be seen to be making an impact.
To start with, let's look at the new business user. Business users are increasingly becoming more vocal and demanding. The way that we consume, explore and share information has been redefined by the way we search on the internet, the easy to use consumer apps we access, new mobile technologies and social networks. Users now expect business software applications to be as simple, intuitive and user friendly as the applications they use outside the office. Business users are now seeking out tools themselves with or without the help or consent of IT.
Next let's look at Traditional BI. Originally BI was IT driven. Business users would request reports from IT and they would do the necessary work involved to extract, manipulate and present the required data. The Data requirements were pre-determined. IT could often take weeks or months to create the dashboards and reports required.
The new Business Discovery tools that are available put users back in control of data. Business Discovery allows you to explore data rather than drilling down through the data. Data does not have to be predetermined. It works the way your mind works, associatively. One question gives rise to a second question, which gives rise to a third question and so on. In essence, it enables you to ask the last question first and enables train of thought analysis. It allows you to uncover hidden trends in data by not only highlighting associated data, but by also highlighting what is not associated. (For example; how is it that we have sold product x to this customer, but not product y?)
In essence, the new business discovery applications have similar characteristics to the tools and applications that business users adopt outside of the workplace. Business Discovery is fast, simple & intuitive like internet search tools such as google. Business Discovery applications also include a global search facility which allows you to explore the entire data set. You can then manipulate and explore the data from a variety of different angles and visualise it in numerous charts, tables and graphs. The associative search feature enables users to define an experimental journey through the data. Business Discovery applications can pull in data from any source and can be accessed on the go via internet browsers on mobiles and tablets.
Lastly, let's look at how Traditional BI was managed. In the past BI Projects tended to be managed centrally and filtered out to different parts of organisation. For example, a management exec in finance would maybe approach IT to create a dashboard or report on specific financial metrics. The approach was hierarchical and specific to that particular department or area.
However, if you think about it, this doesn't actually fit how most businesses actually communicate and make decisions these days. Businesses actually communicate and make decisions in a much more devolved networked way. Decision making is more widely spread throughout the organisation. The new Business Discovery tools available match this framework. They empower, enable and encourage the end business users at all levels in an organisation to access and explore data to improve decision making.
To start with, let's look at the new business user. Business users are increasingly becoming more vocal and demanding. The way that we consume, explore and share information has been redefined by the way we search on the internet, the easy to use consumer apps we access, new mobile technologies and social networks. Users now expect business software applications to be as simple, intuitive and user friendly as the applications they use outside the office. Business users are now seeking out tools themselves with or without the help or consent of IT.
Next let's look at Traditional BI. Originally BI was IT driven. Business users would request reports from IT and they would do the necessary work involved to extract, manipulate and present the required data. The Data requirements were pre-determined. IT could often take weeks or months to create the dashboards and reports required.
The new Business Discovery tools that are available put users back in control of data. Business Discovery allows you to explore data rather than drilling down through the data. Data does not have to be predetermined. It works the way your mind works, associatively. One question gives rise to a second question, which gives rise to a third question and so on. In essence, it enables you to ask the last question first and enables train of thought analysis. It allows you to uncover hidden trends in data by not only highlighting associated data, but by also highlighting what is not associated. (For example; how is it that we have sold product x to this customer, but not product y?)
In essence, the new business discovery applications have similar characteristics to the tools and applications that business users adopt outside of the workplace. Business Discovery is fast, simple & intuitive like internet search tools such as google. Business Discovery applications also include a global search facility which allows you to explore the entire data set. You can then manipulate and explore the data from a variety of different angles and visualise it in numerous charts, tables and graphs. The associative search feature enables users to define an experimental journey through the data. Business Discovery applications can pull in data from any source and can be accessed on the go via internet browsers on mobiles and tablets.
Lastly, let's look at how Traditional BI was managed. In the past BI Projects tended to be managed centrally and filtered out to different parts of organisation. For example, a management exec in finance would maybe approach IT to create a dashboard or report on specific financial metrics. The approach was hierarchical and specific to that particular department or area.
However, if you think about it, this doesn't actually fit how most businesses actually communicate and make decisions these days. Businesses actually communicate and make decisions in a much more devolved networked way. Decision making is more widely spread throughout the organisation. The new Business Discovery tools available match this framework. They empower, enable and encourage the end business users at all levels in an organisation to access and explore data to improve decision making.
Monday, October 3, 2011
Business Intelligence and Mobile Applications
The modern version of an office workspace is not of one which is restricted to four office walls, a desktop computer, and a small area of carpeting. Office workers and executives from every corner of every business and industry are demanding more intelligent mobile applications as they are required to have access to analytical data regardless of their proximity to the office. Business intelligence solutions for mobile devices create an office everywhere you are. To take your company into the next phase of business computing means investing in mobile business intelligence. Mobile BI increases a company's ability to make educated business decisions in real-time, based on agile interpretation of various data streams.
The market for mobile business intelligence solutions is growing exponentially but until now has only been a very small niche in the overall BI market. As a result, advancements in the technologies designed to support and provide BI to users have been slow and inconsistent. Recently though, wireless and remote devices have had ever-increasing access to faster and faster web services. As mobile devices become more powerful many also become less expensive. The fall of two of the largest barriers (speed and accessibility) the door to mobile business intelligence is opening wider, allowing more users more access to important data while on the move.
Developers interested in building BI applications for mobile devices typically have two development options. One could develop native business intelligence apps which is an attractive method if you need apps with the best possible user-experience for one particular mobile platform. If you're developing only for Windows devices native apps can provide tremendous results. However as some companies begin instituting bring-your-own-mobile policies, developing apps to function over many devices is the only way to provide access to everyone who needs it, and everyone is going to need it. There are different methods and languages which can be used to write apps for use across platforms each with individual attributes, some positive and some negative, depending on how it's going to be implemented. What is most important is that mobile applications that provide business intelligence through smartphones and tablets is becoming a big part of the standard operating procedure for managers and decision makers involved in virtually every area of industry all over the world.
As the conclusion to 2011 approaches, the use of business intelligence in mobile application development is growing in popularity but remains in the early stages of utilization. It may be true that mobile BI is new on the scene, but there is proof that it is not just a fleeting trend, proof that can be observed all over the development landscape. Tablet ownership is up, as is tablet use in the office and among executives, and companies are implementing mobile data sharing strategies at rapid rates. The development of mobile apps to provide detailed, real-time data on tablet and smartphone devices grows alongside the numbers of users. Companies not already heavily invested in mobile have been making plans to use the convenience and quickly updated nature of mobile BI solutions to carry themselves into 2012.
The market for mobile business intelligence solutions is growing exponentially but until now has only been a very small niche in the overall BI market. As a result, advancements in the technologies designed to support and provide BI to users have been slow and inconsistent. Recently though, wireless and remote devices have had ever-increasing access to faster and faster web services. As mobile devices become more powerful many also become less expensive. The fall of two of the largest barriers (speed and accessibility) the door to mobile business intelligence is opening wider, allowing more users more access to important data while on the move.
Developers interested in building BI applications for mobile devices typically have two development options. One could develop native business intelligence apps which is an attractive method if you need apps with the best possible user-experience for one particular mobile platform. If you're developing only for Windows devices native apps can provide tremendous results. However as some companies begin instituting bring-your-own-mobile policies, developing apps to function over many devices is the only way to provide access to everyone who needs it, and everyone is going to need it. There are different methods and languages which can be used to write apps for use across platforms each with individual attributes, some positive and some negative, depending on how it's going to be implemented. What is most important is that mobile applications that provide business intelligence through smartphones and tablets is becoming a big part of the standard operating procedure for managers and decision makers involved in virtually every area of industry all over the world.
As the conclusion to 2011 approaches, the use of business intelligence in mobile application development is growing in popularity but remains in the early stages of utilization. It may be true that mobile BI is new on the scene, but there is proof that it is not just a fleeting trend, proof that can be observed all over the development landscape. Tablet ownership is up, as is tablet use in the office and among executives, and companies are implementing mobile data sharing strategies at rapid rates. The development of mobile apps to provide detailed, real-time data on tablet and smartphone devices grows alongside the numbers of users. Companies not already heavily invested in mobile have been making plans to use the convenience and quickly updated nature of mobile BI solutions to carry themselves into 2012.
Subscribe to:
Posts (Atom)
