“Without big data analytics, companies are blind and deaf, wandering out onto the Web like deer on a freeway.” I love this quote from organizational theorist Geoffrey Moore because of the implications it portrays with such simplicity. We have all seen the consequences of deer roaming highways. As digital leaders, we'd rather not see our organizations become roadkill. That is why we invest large sums into advanced information systems that collect, store and analyze our organizations’ data.
But collecting data isn’t the goal; it’s the means to a specific end. The real opportunity is customer intimacy. How well can we get to know our customers? How will we leverage this knowledge to improve their experience and satisfaction?
Getting To The Right Data: Quality Over Quantity
In our quest for numbers, there must be some method to the madness. If it's to be of service to our organization, data must be more than just prolific; it needs to be useful. It should provide opportunities for strategic insight and support data-driven decisions. This is where the quality of data comes into play.
I'm the co-founder and president of a data-driven company, and from day one we were measuring everything, from number of views to number of clicks, and this helped us tremendously. We started taking the same approach for improving our sales and customer success processes, but we weren’t able to rely on the data in these areas at first because of lack of standardization.
This is why clean data is of paramount importance. Without it, leadership can't trust they're making sound, strategic decisions. Once an organization has a dirty data problem, the mess that follows isn't pretty. Poor data quality inevitably leads to dissatisfied customers, poor order to cash and inability to forecast earnings.
Organizations are right to invest in master data management (MDM) solutions for the purpose of data cleansing, transformation and integration practices. Some data scientists even suggest that they really act as “digital janitors” (paywall), cleaning up data before it can be analyzed. However, relying on MDMs — and even data analysts — is like hiring cleaners to come mop your muddy floor rather than just removing your shoes at the door. Where is the dirty data coming from? How do we cleanse it and make it useful?
Taking A Closer Look At The Systems Handling Data
Enterprise organizations invest in a variety of information systems. All can be categorized into two main groups:
• Information systems that examine existing data, but the data is created automatically, such as via a customer journey funnel on a website
• Information systems in which the data is created by the user
In the first group are analytics platforms, such as Google Analytics or any number of embedded programs, as well as business intelligence (BI) systems. These systems observe, offer analysis and garner insights based on data that already exists in various systems. These platforms are able to extract and cross-reference data from different systems and present the findings in a visual dashboard.
The second group is made up of platforms that serve as data-production sites — your CRMs, HCMs and ERPs. A CRM has a component of user data. Salespeople input customer data into the system, which generates reports for tracking KPIs, identifying trends and forecasting earnings.
The Problem With Poor User Adoption
When a user is not competent using a platform examining existing data, it can result in both frustration and miscalculations. Even while the data in the system is obviously correct, the user may pull their numbers from the wrong place, input the wrong dates or otherwise misuse the platform.
This, of course, hurts productivity and increases user frustration. But most importantly, it may lead to inaccurate data at the final layer.
Analytics platforms, however, are where things can get really dirty. With the second group of information systems, the user adoption problem is twofold. It consists of all the same adoption issues as the first group, and, additionally, data input in the system determines the quality of the data going out.
If the user inputs in the wrong information, all subsequent conclusions will be skewed. One mistake not only affects that one data point but countless other data points as well. For example, if a salesperson is not adept in using the CRM or doesn't follow data procedures, they may not know how to accurately input the probability of closing the deal. This affects probability forecasting.
How To Examine Your Data Culture
Software provides competitive advantage only in so much as its users know how to use it, and a preemptive look at the data culture and system usage in your organization may simplify (if not eliminate) large-scale data cleaning projects.
Information systems promise superior business outcomes — better insights, easier reporting, more robust data structure. However, just as with any complex tool, a novice will not wield the same power as a skilled professional. At the end of the day, the platform’s users influence its success or failure. Here’s how to get the most of your data and mitigate poor user adoption:
• Know Who Is Using The Platform. Your users are the ones with their hands on the data. Understanding the data journey within your organization is critical to managing the flow of data and keeping it clean and consistent.
• Establish Skills, Tools And Processes. Put as much emphasis on using the right software as making the software you have work for you. Training is important, but so is continual daily support. Do your employees have the guidance and reinforcement needed to master the information systems they use every day? Are there processes in place to maintain data accuracy?
• Create And Implement A Comprehensive Adoption Plan For New Analytics Systems. Whether onboarding a new system or retraining employees on an existing system where technical skills are lacking, an adoption strategy is necessary. Without alignment of employee actions and organizational goals, the entire data project could be jeopardized.
