Written by GoodData Author |
Historically, analytic offerings—including dashboards and exploration tools —have been one-size-fits-all. But different end-users have different needs, and they expect the tools they use to be personalized. Catering to each of those needs can feel like a lot of work, so product owners may choose to give customers more than what they need and hope they figure out how to search for and find the right information on their own. Unfortunately, when they do this, users often become overwhelmed, which results in confusion, frustration and ultimately, low analytics adoption.
Analytics offerings can start to deliver more value if product owners prioritize the individual end-user needs. You can read more about this in a recent blog authored by my colleague, Petr Tarant. My suggestion, like his, is to approach this simply: define your highest priority end-users, their daily workflows, and the associated decisions they need to make. Then, document each persona and each workflow, mapping decision points in those workflows. Next, look at the particular context of just one of these decision points and think through how you can embed actionable insight into the workflow to drive improved decision-making.
Let me give you some examples:
Consider the loan approval process as an example of an operational workflow. At different decision points within the workflow the loan approver can be offered actionable information based on relevant data right at the point of decision. For example, a recommendation about whether or not a secondary review is needed leveraging historical approval rates and other relevant data, could be provided right at the point of decision. With appropriate information embedded right inside the loan application workflow at the designated point of loan approval, the analytics are abstracted away from the user yet the next step is abundantly clear. The loan approver knows exactly what decision to make and what action to take, all without even realizing that they are leveraging analytics. Compare this to the alternative of having to go outside the context of the workflow into a dashboard and filter through different views to make that decision.
Let’s look at another example, this time of a tactical decision. Imagine a sales manager monitoring sales performance. Perhaps a trend is observed, like a dip in sales in a particular region. In this case, the manager needs to be notified and then will immediately want to understand why the dip occurred and what to do about it. In this case, the analytics offering must provide more than just a notification, extending to include the flexibility to dig in and take action to resolve the issue. The manager needs an insights-driven notification on their phone and also the ability dig into the data quickly, all while on the go. They need a responsive, curated data discovery experience that allows them to dig in but ensures accurate and consistent results. Compare that experience to a static dashboard without an alerts that may or may not be regularly reviewed.
Let's look at one more example, this time of a campaign manager and a decision that requires scenario analysis. Campaign managers need to set-up audiences for the campaigns that they run. They want to optimize the performance of the campaign in advance to ensure that highest yield possible. The campaign manager needs to do this regularly as part of their campaign creation workflow. There are many inputs to performance including, for example, audience age, gender, preferences and past behavior. For this scenario, the user needs to be able to test scenarios and see predicted results in real time based on various inputs that can be changed dynamically. Compare this to a historical campaign performance dashboard without clear context and relevance.
Leveraging a decision by decision approach, product teams can narrow in on what information the user needs to see to make better decisions and what format would make that process most intuitive. When properly presented, analytics can have the business impact for which it was designed: improving the decision-making process for the full range of decisions — from operational to strategic and everywhere in between.
As a product owner, it’s important to build products that meet the needs of end-users and the nuanced decisions they need to make on a daily basis. Overwhelming them with data is just going to result in unhappy users who aren’t sure what to make of the data in front of them and as a result, will most likely ignore it.
Written by GoodData Author |