Data Visualization Storyboard: Documenting Goals
Gaining a clear understanding of your audience’s goals and existing pain points will help you determine what to include and — more importantly — what not to include in the storyboard. The easiest way to do this may be to hold a small planning session that includes the executive sponsor (if there is one) and representatives of each job role in each line of business.
If the sales manager is requesting the data visualization, for example, it’s important that you have that sales manager present, as well as one or two members of her reporting team to ensure that all parties agree on the identified goals. Although in-person meetings are preferable, most global organizations find it challenging to do planning sessions in person. You can request a simple conference call instead.
Avoid sending e-mails to accomplish this task. You’ll only receive multiple, conflicting goals that take extensive time to track down and align with each job role.
Although goal gathering can appear to be a simple task, chances are that it will reveal conflicting agendas and priorities across different job roles. This is why it’s critical to have the executive sponsor or most senior job role present to dictate and align the goal of the data visualization. If a senior-level sponsor is missing, your audience members may be confused and may distance themselves from the rest of the process. This scenario is your worst nightmare. Don’t overlook this requirement.
To guide the goal-gathering process, ask each person present the following two questions, and be sure to document their individual answers:
What are your problems and pain points today?
Ask the audience members to focus on problems that can be fixed with a resolution that can be measured quantitatively. Here are examples of quantifiable and nonquantifiable problems:
Quantifiable problem: If the sales manager states that sales are declining because too much money is being spent on old marketing campaigns that aren’t producing results anymore, an opportunity exists. You can review historical trends, identify declining campaigns, and reallocate the spending to boost high-performing marketing campaigns, thereby increasing sales.
Nonquantifiable problem: If the sales manager states that the sales have declined by 5 percent to 10 percent for the past four years due to a lack of motivation among the sales reps, you should avoid trying to measure this decline. Avoid statistical measurements that require heavy data modeling, such as regression and T-models
What are your goals, and what does success look like?
These questions usually invoke a wealth of responses. You want to be sure to get each goal down to a simple statement and keep the count to the top three or four most important goals.
In addition, the goals need to include quantifiable responses that can be measured with a defined target. Each goal should tie directly to solving one or more of the problems identified in the responses to the first question, which is the only way you’ll be able to measure the return on investment of your Big Data visualization project.
Here are two examples that show you the difference between good and bad goals:
CFO (Chief Financial Officer): “We want to increase our company’s revenue by 10% in the next 12 months. This will require that we bring in an additional $500 million in revenue across all divisions.”
This is considered to be a good goal because it has a clear target with a set timeframe that can be measured.
Sales Manager: “We hope to influence when our product hits the sales shelves, to drastically improve our ability to sell more, and therefore to hit our target of a 10% revenue increase in the next 12 months. However, because we are distributors and have no actual control over the shelving process in the stores, our sales rep will be required to visit the store managers twice as much each month to build the relationships, hopefully influencing our products hitting the shelf sooner.”
This is considered to be a bad goal because the Sales Manager is seeking to increase his revenue by reducing the time it takes for products to hit the shelf — an action that he today has no influence over. This is a typical case of users wanting to view data that is not intelligent (unactionable).
When you’re able to document one to four solid goals, your aim is to gain consensus for each goal among the entire group to prevent confusion going forward.