These days, data is more critical than ever, with customers, employees, and decision makers alike demanding more transparency than ever before. However, as the need for data increases, so does the amount of data being collected. While more data might seem like a good thing, it actually creates a lot of confusion for the intended audience. Moreover, it makes it difficult for data analysts to do their jobs effectively. The point of data sharing is to drive informed decision making, but data analysts are finding that the more data they are given, the more difficult it becomes to create reports that drive action. For business owners, this is a major concern, as the last thing they want to do is invest in BI efforts that end up costing more resources than they provide in benefits.
At iDashboards, we love data more than most, but we also understand that more data doesn’t necessarily mean better data. Each report developed and each piece of information shared should input business decisions. If it doesn’t, it isn’t necessary. By taming your data, you can reduce BI costs and more importantly, ensure that valid data is acted upon instead of lost amidst a sea of interesting but pointless insights.
Identify What’s Relevant, and Scrap What’s Not
“Just Because Something Can Be Reported Doesn’t Mean it Should Be.” – Advice from a frustrated Reddit user
As data nerds, we understand just how hard it can be to not include every piece of data you collect in your dashboard presentations. However, if you want your audiences to care, you need to trim down your information to what is strictly relevant to them. In order to do this effectively, choose one metric to focus on and shape your data story around that. This is a matter of getting people the data they actually need as opposed to what they think they need.
Unfortunately, too many data users don’t fully understand what they need. This is where a good data analyst comes in handy. Instead of asking users what they want or need, ask what problem it is that they’re trying to solve or about particular challenges they face. You can then pull data strictly relevant to those challenges, highlighting one key metric that directly addresses their concerns. All other pulled data should either serve to support that key insight, or be archived for later use.
For higher-ups that need insight into several areas of the company, “at a glance” reporting is key. In these instances, detail can be more cumbersome than helpful, and a succinct and accurate summary should suffice.
Identify Key Data Sources and How They Talk to One Another
In addition to having too much data, many companies have too many data sources. Moreover, they don’t know how those data sources work together. Most businesses pull KPIs from a number of sources, including customer-relation management systems, sales records, mobile sites, apps, Salesforce, and even Excel workbooks. However, once they have all this data, they don’t know what to do with it, so they throw it all into a report and call it a day. Doing this is highly unhelpful, however, and results in data overload. Data blending can prevent this from happening.
Data blending refers to the act of extracting value from multiple data sources and uncovering correlations between different datasets. By understanding how your data sources work together, you can create dashboards with actionable insights necessary for better business decision-making. Moreover, by understanding where your valuable data comes from, you can refine how you organize and segment your data, making for a more streamlined data-mining process.
iDashboards allows you to connect all of your data sources and to easily build charts using a mix of sources in a single dashboard. Better yet, dashboards identify valuable patterns for you. This way, if you do miss a key connection between, say, your sales records and mobile site transactions, dashboards will reveal the repeated trends so that you can act on them.
Customize Your Data to Your Audience
Finally, many data reporters have the added problem of having to share their findings with several different audiences. For example, reports that get delivered to the regional level will vary from those that corporate receives, which will differ from those that the division level receives. While the same information pertains to everyone, it may not necessarily be relevant to everyone. It is your job, as a data analyst, to figure out who needs what information, and to deliver that information in a clear and concise manner.
In my last blog post, How to Optimize Data Reporting, I urge data reporters to “think like a journalist,” and “instead of delivering awesome headlines, deliver relevant data.” The best way to accomplish this, he explains, is to understand stakeholders’ goals, what they need to know, and why they want to know it. Go even further, however, and anticipate future questions. Once your audience reviews your reports, what questions will they have? Concerns?
Once you have a firm grasp of who is going to see your reports and what they’re going to want to see, you can begin building your data story. Like a journalist writing an article, start with the most important information first, and then follow up with details and contextual information. By prioritizing your data, you can effectively capture your audience’s attention with information that they want and need to see first. As they get further into the data story, what would otherwise be seemingly random information becomes important, as you’ve successfully made the correlation between those additional datasets and the leading dataset.
It can be difficult to identify relevant data and to accurately prioritize it, but according to Scott, reporting follows a natural hierarchy once you answer these questions:
- Who is your audience?
- What do they need to know?
- Why is this data important to them?
- When do they need the information?
- What’s the best way to deliver this information?
And hey, if all else fails, start with KPIs, as more often than not, stakeholders’ concerns begin and end with department performance levels.
Dashboards Can Help Tame the Data Beast
At iDashboards, we understand that data can be overwhelming, but that’s no reason to throw your hands up in despair. The next time you’re feeling overwhelmed by the amount of data thrown your way, or by the sheer number of data sources you have to sift through, take a deep breath and look to iDashboards for help. We offers the solutions you need to turn seemingly random sets of data into cohesive information that your organization can gain tremendous value from. With features like automated alerts, ad hoc and scheduled reporting, interactive intelligence, and much more, this customizable data visualization software is your secret weapon against the data beast.
Get the Guide Psychology of Data Vizualization
Take a primer in cognitive psychology, the science of perception, and neuroaesthetics and learn how to make dashboards even more effective.