The Missing Piece Of The Data Puzzle: People
While 92% of leading companies are increasing their investments in data and analytics (New Vantage, 2019), they are faced with the daunting reality that 85% of data projects fail to move past the preliminary stages (Gartner, 2016). Given these startling statistics, it’s clear that a key component of the data equation is missing. To truly transform your business processes, functions, and customer experiences, you need to evolve your data model design to factor in the most crucial component of the data equation: people.
Recently, we unpacked an innovative approach to human-centered data design and shared the prevailing challenges that leaders must overcome in order to spearhead a human-centered data transformation. In this article, we build on that foundation by sharing 3 steps that will help you successfully achieve a human-centered data design revolution and maximize the ROI of your data investments.
Stage 1: Research & Understand
The crux of human-centered data design is the notion that all data originates from a person. Regardless of whether the data is captured through a machine and analyzed by an algorithm, a person is responsible for defining that data point and creating that algorithm. Humans are deeply and inextricably involved in the data design process, so when you create an insight, it should be with the knowledge and intention that it impacts a person – whether that person is the Customer Success Manager (CSM) interpreting it to take action, or a customer on the receiving end of that data point.
Therefore, the first step in any human-centered solution is conducting thorough stakeholder research and analysis to understand the needs and pain points of the people you’re problem-solving for. For example, if you’re designing a churn prediction model, you need to begin by deeply understanding your customers and the personas that are engaging with your product. Consider what would make them feel like they’re getting value from the use of your product (which is a very human experience) and thus need to continue using it by answering the following questions:
- What does your customer care about and what do they get value out of? What goals and outcomes are they trying to achieve from the use of your product?
- How do they experience value with your product? How does that value manifest for them?
- How can they determine and verify that they’re achieving value versus not?
Once you determine what drives value for your customers, you can theorize about how to verify that that value is being achieved based on data. At this stage, it’s all about understanding where the process is today, while thinking about how you can design empathetic solutions to improve experiences and maximize value for both customers and employees moving forward.
Stage 2: Theorize & Brainstorm
After you’ve conducted in-depth stakeholder research, you can create an initial hypothesis. You can then build upon this hypothesis with data analysis. The combination of the two will help you better understand the domain you’re solving for, such as reducing churn, and guide your initial approach to solution engineering. You can then brainstorm what a solution looks like, what the key requirements are, and how this solution will fit into your employee’s workflow and your customers’ journey.
While the theorizing and brainstorming phase may be clear cut at times, the data you’re using and the kind of value that your customer is trying to achieve, can complicate this process. For example, if your product is designed to increase the value of your customer’s Marketing spend, then both you and your customer can easily and clearly see if that value is increasing or decreasing. While there may be an emotional connection to that value, there is also a specific, measurable and verifiable data point. However, not all insights are as cleanly aligned with a single data point as this. For instance, if your product or service provides value by adding efficiency to your customer’s day-to-day, that may be more difficult to measure and quantify. In this case, you really need to dig into your data and research to theorize how you might be able to observe whether or not a customer is getting value and enjoys using the product and how they are different from a customer who is not getting value or doesn’t enjoy using your product. You will have to define how you’re going to differentiate patterns of data and create a compelling and impactful theory for your model.
Stage 3: Operationalize & Iterate
At this stage, you must work with your stakeholders to introduce and analyze your solution in the real world. For example, if you created a churn prediction model, how does your new model compare to existing benchmarks and previous models? How does the solution impact your employees’ workflows? Did the new model help you predict churn more effectively? Based on this analysis, you may need to go back to the drawing board and repeat the process from step 1, gathering feedback from stakeholders and adding additional data to improve and iterate on your model.
Iteration is a critical piece of creating an effective human-centered data model – your product and customers are constantly changing and evolving and your data needs to be updated just as frequently in order to reflect these changes. This step may also help you uncover obstacles to your human-centered design that exist in your business processes and within your organizational ecosystem. Based on your evaluation, you may need to change how your team works or how they are incentivized, as they may be at odds with how you want your solution to be operationalized. Many leaders are challenged with aligning Sales and Customer Success around a centralized data model. If you’re currently working through this, check out our eBook for an in-depth guide on how to integrate your customer lifecycle to accelerate value realization and best-in-class NDR.
Closing The Data Divide
Human involvement in data is unavoidable – it’s impossible to extricate and separate the two. The reality of the relationship between data and people means that organizations must fundamentally transform their approach to data design in order to meet the needs of the people most impacted by their data projects. By conducting thorough stakeholder research, theorizing possible solutions based on your research and then testing and iterating your data model, you will maximize the success rate of your data projects and guarantee a tenfold increase in ROI from your data investments.