Data Visualization and Analytics 📈
Organizations hire me to unlock the power of their data and identify issues. I do this by creating interactive visualizations and custom reports.
For decades, enterprise systems have shipped with loads of standard reports. With a few clicks of a mouse, one can see a firm’s P&L or balance sheet, latest sales, payroll ledger, and the like. As for ad hoc tools, some are better than others. Brass tacks: It’s not uncommon to find employees regularly scurrying every month to cobble together reports from disparate data sources. Even small errors can lead to disastrous business decisions.
Less technical employees in large organizations typically open tickets with support desks. While smart, those folks often don’t possess the business knowledge necessary to correctly write the report. Almost invariably something gets lost in translation. Make no mistake: the IT-business divide is alive and well.
Over the years, I’ve spent more hours than I can count writing reports for my clients. A large part of my consulting work actually fell into the buckets of data visualization and report writing. Most of the time, I was able to automate my clients’ report-generation process. In the process, I saved them a great deal of time, improved the quality of their data, and allowed them to make better business decisions. My dashboards allowed even the least tech-savvy employees to run reports with ease.
Fifteen years ago, I’d write reports in Microsoft Access, Microsoft SQL Server, Crystal Reports, ReportSmith, and a bunch of Cognos tools. These days, I’ve moved primarily to Tableau— and even teach a class on it at ASU. As I write in my book The Visual Organization, Tableau excels at interactivity and allows users to easily discover new insights in their data. Make no mistake, though: I still know my way around Access database.
My wheelhouse remains HR and payroll reporting. To this day, I know the Infor/Lawson HR Suite and database tables like the back of my hand. Depending on my clients’ needs, though, sometimes I write inventory, sales, financial, and logistics reports.
The complexity of my reports and data visualizations varies from project to project. Some are straightforward. Others…not so much. Without going into too much detail, some of the reports that I have had to write hurt my brain. One extremely time- and resource-intensive report took three hours to run from start to finish. That one was a real doozy.
WE HAVE CONTRACTED PHIL A NUMBER OF TIMES BECAUSE HE IS VERY KNOWLEDGEABLE, BUDGET-CONSCIOUS, HONEST, AND PRODUCTIVE. HIS WORK IS FIRST-RATE.
—KAREN ZUILL, CONTROLLER, HARTWICK COLLEGE
Engagements and Philosophy
I’ve learned a few things from doing this over the years. For one, my most successful client engagements tend to be ongoing—especially at first. That is, an organization had me build a number of critical reports and/or data visualizations as part of a medium- or long-term contract. As with any technology gig, it can take a while to do the following:
- Procure the requisite credentials and get set up on virtual machines
- Locate and provide access to data dictionaries and other relevant documentation
- Coordinate schedules
- Understand the difference between what people say and what they really mean
After I had established a rapport with my contacts, then we had more success with one-off reports and visualizations.
Next, I’ve seen firsthand the folly of attempting to boil the ocean. As a result, I follow Agile development practices I detail in Analytics: The Agile Way—specifically Scrum. In short, I present my work in short iterations with closed loops.
Don’t get me wrong: I am used to working independently. Still, there are times in which I’ll need to solicit client input on a number of different fronts. People who think that they can can simply hand off a list of vague reports in Excel to developers are bound to be disappointed with the end result. I strive to avoid that.
Finally, I won’t use e-mail as a project management tool. I prefer Slack for general communication. For individual data visualizations, I like using Trello or an equivalent tool. Keeping feedback in individual containers is far more efficient than parsing through dozens or hundreds of messages.