Beyond Dashboards: Business Intelligence Tools for Program Analysis and Reporting

Submitted on Wed, 8/8/2012 - 4:29pm
Business Intelligence tools, or "BI" tools, take the dashboard idea to the next level. More than simply graphically displaying static data, they offer trend analysis, forecasting and drill-down capabilities that can dramatically expand your insight into program performance.

Strategic, mission-critical decisions require a clear understanding of program performance, but for most nonprofits, making decisions on facts rather than gut feelings is easier said than done. As a way to better understand program performance and become more responsive to changes, organizations are increasingly turning to dashboards—custom utilities that gather, organize and present information in an accessible way—that let them more effectively measure, monitor and manage the way they meet their missions.

There are a number of ways to create dashboards, including using such common tools such as Excel or Access, or proprietary systems embedded in databases. But these approaches can lead to dashboards that are not very user-friendly and difficult to update. 

Business Intelligence tools, or "BI" tools, take the dashboard idea to the next level. More than simply graphically displaying static data, they offer trend analysis, forecasting and drill-down capabilities that can dramatically expand your insight into program performance. With a good BI tool, you can combine data from multiple sources, view it from different perspectives and distribute it more easily. Beyond simple reporting, BI tools allow a more comprehensive analysis of your organization's data1

Not every organization needs a BI tool, but if you have a solid strategy for program evaluation and monitoring and already capturing the raw data you need, but struggling to analyze data and make use of it, one of these tools might be a good fit. We'll look at a few of the better known options—but first, let's take a closer look at what they can offer, and how they work.

The Benefits of Business Intelligence Tools

Business Intelligence tools have grown in popularity with nonprofits for a number of reasons. The most prominent are better data integration, more flexibility, easier distribution of data and better visuals. 

Many organizations have data that is compartmentalized in separate information silos—you might have donor information in one database, client information in a second database, and financial information in a third. BI tools let you integrate data from numerous sources, and many can even pull data from Excel spreadsheets and Access databases, or any database that has an Application Programming Interface, essentially a way for programmers to access and export your data.   

Most BI tools move the data into a cache—a virtual storage space—or a separate data warehouse, effectively creating a separate database. This allows the data to be manipulated for analysis without affecting the data in the original databases. Data can be loaded manually, or set to load automatically at predetermined times or when certain actions occur. 

If you want more flexible reporting than you can easily get from your existing systems, BI tools can offer even more flexibility than something like Excel or Access, and can display data dynamically from a variety of perspectives and in near real-time. (Because the data is loaded from the original source into the warehouse, it's not fully "real time," but it's close.) For example, it's possible to see a report on the number of people admitted to your program, view the success rate at different stages, then see a trend analysis of the success rates over time.  

They also offer more diverse ways of accessing and distributing reports and dashboards. You can email staff up-to-the-minute program data on scheduled days and times, and easily view and manipulate dashboards from a smartphone or tablet or anywhere you have an internet connection. Many also let you set up custom portals for certain audiences (such as your board of directors) that grant unique access to view predefined performance measures and reports.

All the data in the world won't help if it cannot be displayed in a way that is easily interpreted. BI tools go far beyond dials and bar graphs, and can produce sophisticated graphics like scatter plots that move with time, spark-lines that show thousands of data points, and forecasts that assume different user-inputted scenarios. Not only can these graphics display more information, but they can also be customized with colors and themes that match your organization's brand.

The Considerations 

Though BI tools vary in cost, most organizations can find an affordable system for their price range that satisfies their specific requirements. However, there are other things to consider before you invest any time or money, including your organization's evaluation strategy, data collection, and personnel.

Your organization needs to have a solid strategy for evaluation in place. Do you already know how your organization wants to analyze its programs, and what metrics to track? No BI tool will give you these answers, and spending time outlining the requirements early in the process will save a lot of time and money later. These metrics need to be very specific. Outline the data fields you want to use, and how you want to analyze them. However you choose to define them, make sure you know the basics of what you want to analyze and how you plan to do it before you begin shopping for a BI tool—even if it means consulting with someone who can help define an effective and appropriate plan for measuring your data. 

After implementing a BI tool, many organizations struggle with "operationalizing" the data to build a strong performance culture—in other words, finding ways to improve performance based on what the data show. Your organization may want to change processes once it begins analyzing program data from different angles, but if the data in the BI tool is not organized the same way as in your primary system, these changes might be difficult to enact. Data warehousing creates multiple data models, and can result in making key data points difficult to operationalize—if this is one of your goals for purchasing a BI tool, you may want to first consider reorganizing your original data model. The greater the difference in the data models, the more difficult or complex the project will be.

You also need to weigh the human side of the equation. What skills do your organization's staff already have? Every new tool will require learning some new skills, and training is a consideration. A consultant can help get your organization up and running, but be ready to invest in IT staff training for configuration and ongoing maintenance—doing so during the implementation process can engage them and help them learn from the consultants.

A Few Good BI Tools

There are a number of BI tools on the market, with new ones being developed all the time. More and more of these systems are adopting web-based platforms or using in-memory technology, which means they offer increased accessibility and responsiveness over installed systems, but can suffer when scaling to very large datasets, and require a fast, consistent internet connection. These tools tend to be less expensive than other types, but these are all complex systems that will require someone with data expertise to set up and maintain.

GoodData's cloud-based platform offers great visuals, a very fast data engine, and data pivoting—a quick way of summarizing data based on different variables—from multiple data-sets updated in near real-time. The tool offers canned reports, which might not apply to all nonprofits, and adds a reports-and-dashboards sharing feature for easy collaboration among colleagues. GoodData's tool is very quick to deploy, but setting up and administering the data model can require more technical expertise, and accessing the data using a cloud-based user interface requires a solid and consistent internet connection. If you're looking for a cloud-based solution that allows quick ad-hoc analysis and have some more technical people on staff, this is a great option.

iDashboards is a very slick data visualization tool that allows organizations to monitor and analyze their programs while also creating "what if" scenarios for strategic planning. The user interface does more than just look good; it allows staff to personalize individual dashboards with very little training. A built-in wizard helps facilitate its connection to multiple data sources and automatic refreshes every minute. If your organization wants staff to create personalized dashboards with great visualizations for analysis and planning, iDashboards might be a good fit. (The vendor offers discounted licensing and training for nonprofits.)

Birst is an extremely powerful and flexible tool with a responsive interface that stores data in-memory or on-disk, which allows it to scale quickly while easily performing sophisticated calculations with very large datasets. Birst comes pre-packaged with standard reports; these might not meet the needs of most nonprofits, but the tool does allow end-users to make quick Excel-like calculations on their own. Setup is comparatively easy due to Birst's new graphical interface for logical modeling and "data warehouse automation technology," which automatically creates a data scheme. Birst is a great option for mid-sized to large nonprofits with dedicated IT staff who are working with a lot of data and seeking scalability and flexibility in deployments. 

QlikView is a tool that is quick to deploy and relatively easy to administer. The tool is designed for exploration of data by leveraging their "associative model" that allows users to click on a piece of data and see all relevant data associated with it. This type of visualization is perfect for analytical end users who might not have a pre-defined question. QlikView is also very fast; the tool uploads and compresses your data in its memory to allow for very quick analysis and drill-down performance. Organizations should expect to send their IT staff to training as the complex tool does use propriety scripting, a unique interface design tool and management console. If you're looking for a tool that is quick to deploy, looks great, is cost-effective and very fast, and have a strong IT staff prepared to attend a week of training, QlikView is an excellent option. QlikView provides the software free to nonprofits, including training and consulting. (http://www.qlikview.com/us/company/community-service)

Tableau can run straight from your database with no additional modeling required, assuming that all the data you want to report on is in a single database. This allows for very quick deployments—just five clicks—if your current data structure is sufficient. Organizations can also choose to re-build their data models in-memory. Because everyone knows that "a picture is worth 1,000 words," Tableau used that concept as the basis for its product; the tool is extremely visual, allowing organizations to quickly see their data when answering questions. The tool's proprietary language, VizQL, quickly translates your data into graphics that can be manipulated in the drag and drop interface. Tableau provides all of its training materials online for free, and hosts regular webinars to keep everyone up to date. Tableau is an excellent fit for organizations that do not have large IT departments, but are looking for a tool that offers a strong visualization suite and quick deployment. Tableau offers reduced cost licensing for nonprofits.

Wrapping up

Business intelligence tools can relieve a lot of the headaches associated with monitoring and evaluating programs and outcomes. Your organization's need will depend on what (and how many) systems it's currently using and how sophisticated you want your analytical capabilities to be. Not every organization needs a BI tool, but if you have a solid strategy for program evaluation and monitoring and are already capturing the raw data you need, but struggling to analyze data and make use of it in your organization, these tools might be a good fit. 

Every tool offers connectivity and visuals, but the devil is always in the details. Setup, customization and scalability can greatly effect ease of use and cause even more headaches. Make sure your organization has thought through its data strategy, evaluates the tool completely, and purchases the one that fits the needs of future data analysis and the current capacity of the IT department. 

This article was originally published at http://www.idealware.org/articles/beyond-dashboards-business-intelligence-tools-program-analysis-and-reporting and is reprinted with permission.


1 For more on APIs, see NTEN's report: "How Open APIs Can Change How Nonprofits Manage Data". NTEN, 2007.

Thanks to Danielle D'Antuono at the Center for Employment Opportunities for providing recommendations, advice and other help.

Patrick Yurgosky is an independent IT and performance management consultant. He previously worked as the Director of Business Analysis at the Center for Employment Opportunities, a leading workforce-development organization based in New York City; and with Caribbean governments on management and environmental issues for The Nature Conservancy. He has founded multiple for-profit companies, and manages a successful Pilates company in his free time. His expertise includes information technology, operations analysis, strategic planning, and business development.