If you have ever watched a vendor demonstrate their software, they start off showing the functionality that ticks off all your requirement boxes. For the big finale, the salesperson perks up and excitedly asks the executives in the room to pay particular attention to what is coming next. The demonstrator pulls up beautiful reports with charts and gauges. Your team is excited and sold on the reporting capabilities alone. Then, the deployment starts, and you find out that all those beautiful reports were for demonstration purposes, and you have to build the reports yourself. IT is left with their pants down because now they need a specialty resource to create the reports, and there is probably no budget for that.
In many cases, niche vendors sell a reporting module that is very specific to their product. While you can generate some nice looking reports, they do not have the breadth of features a report designer needs. For example, the reports may look good on a computer screen, but there is no output option for mobile or there is no way to push reports out with emails or via notifications.
Another challenge with these niche vendors is their reporting module only works with their data. In most cases, organizations need reports that can capture data from multiple systems. Let’s say for example that you purchased a niche project management application. The application has some excellent reporting capabilities, but the only way you can get value out is by reading data from your financial system. It is likely that the project management application’s reporting module will not offer any functionality to read your financial system’s data.
The Excel dilemma
With all these challenges (and more), the next obvious solution is to use Excel. In this situation, you have a person (or people) working every day, week, or month getting data extracts from various sources. This person now has to compile all the data and build the reports management is looking for.
Data from different systems always store information in slightly different ways. Before distributing the report, the data usually requires hours or even days of cleansing. You get the basic issues of dates in one source being a string with dates in another source being in a valid date format. Then you have the most complicated scenarios where data has to be linked using vastly different types of unique identifiers or needs to be pre-processed and aggregated before the information is useful.
These Excel power users you are paying for have dozens of macros with hundreds of lines of code and double that in specialty formulas. If just one little number or field is off, nothing will run, and there is probably little documentation. Worse, the macros may do their job, but then the actual data is wrong, and then bad decisions are made because of incorrect data that no one caught.
The real challenge
I can list all the issues related to designing reports and dashboards, but there is a bigger problem here, and this is one every CIO needs to remember:
A business application is nothing without a solid reporting foundation. If people are scrambling around gathering data every week, the company is wasting precious resources. The resources should be spending their time evaluating the data to make better business decisions.
Enter self-service BI
Having information readily available and easy to use is still a dream for many organizations, but that is changing, and quickly. IT pros now have so-called self-service BI (business intelligence) software products that promise to commoditize reporting and dashboards for application owners and business users. These products include (but are certainly not limited to) QlikView, Tableau, and PowerBI.
These self-service reporting products have the promise of changing how businesses can make use of their data. Need your sales forecast from Salesforce? Drag-and-drop the data onto a canvas. Need to look at the past financial results of those customers from SAP? Drag that data onto the screen. Want to see the data by region? Drag a map onto the screen, and it will automatically color-code the map with the underlying data.
If you pre-build some dashboards for the business users, you can package them as mini applications that users can access on a phone, computer, or tablet. The business owner can interact with the data in a visually rich and (mostly) user-friendly front-end. If data changes, the application owners can receive notifications and see how their business is affected in real time.
It is this type of self-service BI we need, but that too is not without its challenges.
Self-service BI challenges
Self-service BI sounds great, but there is still work to do. Just like the Excel power users, someone needs to build the data connections and, where applicable, link it to other IT systems. IT will need database designers and developers to update and refine the reporting data continuously.
Many software vendors still do not document their databases, so IT must require providers to have an open and accessible way of accessing the data. IT must also invest in bringing that data together into the self-service BI reporting tools.
Help is on the way
Fortunately, with these new self-service BI solutions, vendors realize they can essentially package up data access and dashboards as applications for QlikView, Tableau, and PowerBI. Some third parties are even beginning to sell report packs and dashboards. You simply find the reports you want on an online store, purchase it, and install it into your self-service BI application.
Remember, though, these solutions may not always tie your other corporate data into the pre-made dashboards, so you will have to be thoughtful in this sense.
A call to action
If you are investing in an expensive software solution, it must have a reporting element in the project. IT needs to demand vendors make their data accessible and usable for their chosen BI platform. The business users need training, so they can learn the BI tools and build out reports with the datasets IT provides them.
Business owners need to recognize the data they are gathering is not just meant for the weekly management meeting. The business owners should invest in data analysts and data scientists who can interpret the data, enable data-driven decision making, and drive better performance.
Self-service BI should not just be for the data analysts and scientists. Anyone (or as many people as possible) should have access to data pertinent to their jobs. If IT can consolidate the data, there should be an expectation anyone in the business can and should be able to create reports.
Excel still has its place, but it is time to embrace the age of data-driven decision making by opening up the data locked in our IT systems.
Photo credit: Pexels, Qlik, Pixabay