For the second day of dashboard week, we took a step back and reflected holistically on our craft. Specifically, we explored what factors and considerations go into effective data communication. This subject is well within the scope of previous articles that I’ve written, such as here and here.

 In this article, I will document some of the notes that I took throughout the day. In particular, I will highlight principles of effective data communication and how they ultimately interact with the larger idea of business intelligence analysis through data visualization.

 We started the day by working through two broad stages that underlie our work: exploratory data analysis and explanatory engagements (presentations, etc.). In the former, this work is mostly done for the sake of our own understanding. With this, we try to understand the underlying structure of our data and ascertain what interactions the data may have (or may not have) with the business objectives that we are given. For example, we would want to understand:

 What does a row of data mean?

 Where does the data live? In other words, what hands does the data pass through and how does the changing of hands affect the data that we are ultimately given?

 Are there any missing fields? Do we need to bring in more data?

 What is the distribution of the data?

 Are there any outliers or discernible patterns?

 In its current form, is the data actually able to address the business objectives? If not, does the data need to change (or rather, would the business objectives need to change to better cohere to the current data?)

 While these questions scratch just the surface of what the exploratory stage entails, the “hands-on” work (what can be done with Tableau, Excel, SQL, and so on) of this stage lays the foundation for the explanatory stage that is to come. That is, it is only by learning from the data (and communicating with our stakeholders and audiences throughout this entire activity) that we can understand the data sufficiently enough to be able to glean insights from it.

 It is in the second activity, the explanatory stage, where we actually communicate the insights that we have gleaned from the first activity. It is in this stage where the bulk of our work gets measured up and scrutinized. It is also where we receive direct feedback with which to inform further iterations of our prior exploratory work. Put another way, I see the exploratory stage as the planning and construction phase of our data products while the explanatory stage is when we actually share our conclusions, recommendations and larger concerns about the data from our data products.

 In the next section, we delved into some fundamental consulting questions and considerations. For example, we talked about the differences between stakeholders and audiences and how the two can sometimes be the same person or group of people. My biggest takeaway from this section was how we should try to anticipate and ultimately showcase the priorities that our stakeholders and audiences may have. Another key piece of this section involved stakeholder management. In particular, we were reminded and encouraged to always relay the business objectives that our stakeholders ask of us after any meetings and throughout our work. Taking the time to document any changes in business objectives is instrumental in maintaining trust with stakeholders, work-life balance and ensuring clear direction and work progress throughout our partnerships with our clients.

 Moving on, the rest of our discussion centered on:

 Why do we visualize data?

 What core themes do our charts convey?

 How do we organize (sort) the data in our charts?

 From these above considerations, we explored some of the most basic and foundational charts and what their different use cases could be. For example, we discussed different types of bar charts and line charts and talked about how they convey data differently (such as how bar charts are more effective at conveying magnitude and ranking than at conveying spatial information). At the end of this section, we worked through different business prompts by drawing some charts on whiteboards and talked about why we chose the charts that we did. Something that really stood out to me in this section was how we, as data consultants, have to tread the fine lines of contributing to the creation of business-sensitive data products on the one hand as well as recommending data visualization best practices to improve business operations on the other. It is not that we do only one or the other, but that we do both and that how we do these two things have to constantly be top-of-mind in the questions we ask, the feedback that we receive and the insights that we share.

 That is, ultimately, our work isn’t simply to build, maintain or remake dashboards that are useful but we additionally have to promote dialogue between our different stakeholders and audiences to support them in their efforts to become data-driven and data-literate. In other words, our work isn’t simply to engineer or to train; it’s both and we have to navigate these two broad tasks in a business-aware manner. We want to provide genuine value and support for our clients by working with them rather than dictating that they abide by our standards and conclusions. In practice, to me, this means always having a conversation as to why something about the business is the way that it appears. Why does the stakeholder want a Sankey chart over a bar chart? Why could making very niche and complex data products potentially be a problem and actually undermine the larger aim of data-literacy within the business?

 Ultimately, for me, today was incredibly insightful and gave me plenty of space to really come to terms with the crux of my training here at the Data School. While we had similar discussions like this sprinkled throughout our four months of training, this discussion in particular further reminded and clarified to me the core expectations, skill-sets and operational awareness that are involved with being a competent data consultant. That is, while we may not have all of the general business acumen of seasoned industry researchers and business analysts, we are sufficiently data-literate to be able to communicate to our business partners almost all of the ways their needs can be met by incorporating data visualization best practices. Moreover, rather than simply offering recommendations or providing educational training, we can actually work with different business departments to meaningfully translate their unique data needs into holistic, effective and responsive data products that can be used by them and their peers to meet their business requirements.

 To me, what effective data communication looks like is when stakeholders and audiences are better able to learn from and develop their data in order to apply their newfound understanding(s) in positively bringing about real progress to their operations.