Product vs business outcomes – an independence story

Recently, I was speaking to a customer in the home healthcare Industry who was looking to purchase an electronic health record (EHR) system. They told me about all the features and capabilities advertised as part of the solution. Dashboards, analytics, integrations, and interoperability - and no sales pitch would be complete without mentioning AI! I could sense their frustration. They told me that they were hearing the same thing from many vendors and had little or no confidence in any of the systems actually improving their business. 

The timing of this conversation coincided with a clip from a Teresa Torres podcast where she spoke about product outcomes and business outcomes. This clip resonated with my conversation. My customer was telling me that they didn’t believe these vendors could help to enable their business outcomes, while they were repeatedly being told about product outcomes. Sometimes not even outcomes, just features and capabilities.

By nature, a data or machine learning analytics product is unlikely to deliver business value on its own. The insights these products provide require a user to act on the insights and value they create - without this action there may never be any value realisation. How then, can technical product managers ensure their machine learning analytics products deliver meaningful business outcomes by defining, tracking and linking their product, operational and business outcomes?  

Product outcomes can come in many ways, shapes and forms, depending on the type of product. But all measure the behaviour of customers and end users. It could be the number of transactions completed within the application. It could be any number of adoption metrics (including but not limited to DAU, MAU, and stickiness) or the length of time a user spends in the application. They are often leading indicators that the Product is driving an outcome for the business.

Business outcomes are lagging indicators often measured to show tangible value that has been realised. Examples include productivity, cost reduction, increased revenue, or market expansion.

But for data and machine learning analytic products, positive leading indicators (product outcomes) do not always translate to business outcomes. 

There is a critical dependency often overlooked: action and/or operation. 

Empathetic machine learning and analytic product managers with deep domain knowledge understand these operations and actions in detail, and are able to align operational outcomes into their product and business outcomes. This is formed from a foundational understanding of the value that your product creates and ensuring that this value is realised by your customers.

In reality…

I'm currently managing a machine learning analytic product that supports an operational process that directly impacts regulatory quality measures that play a significant role in a patient's healthcare provider selection. 

The product outcomes that were identified for this product were:

The business outcome was an improved quality measure for this specific operation.

Mapping these product outcomes to the business outcome was relatively simple: If the insight we provide an end user is accurate, and we know that an end user is reviewing/observing that insight, then they can act to improve the quality outcome of the operation. However, the business outcome is dependent on many other factors - none more so than the correct actions being taken on the insights provided by the dashboard.

Our algorithm accuracy was a lot higher than the quality score. Similarly, the DAU were right at the number we were expecting to see. We were confident in the value created by the product. While there had been a marginal improvement in the operational quality score, both of these product outcomes indicated that the product was not being adopted and leveraged optimally. Rather than repeatedly pointing out that the product did what we said it was going to do, and that this was not our fault, we embraced the challenge of helping the customer to achieve their business outcomes. 

There was a disconnect between the insights provided and the business outcome. When we analyzed the operation, it was clear that actions were not being taken on the insight. We then defined measures for these operational outcomes so we could easily track and monitor these operations to ensure that the value was realized. 

Furthermore, we worked with operations leaders to define best practices when reviewing the dashboard. We learned the standard operating procedures (SOPs) for actions to be taken based on certain data points and built solutions to track adherence to these SOPs. We improved transparency on the algorithm, building further trust with the End Users taking action on the insights it created.We went above and beyond the original statement of work. 

In doing so, we increased our understanding and empathy with the Customer, and used this increased understanding to improve the Product and build complimentary offerings. The customer interactions changed noticeably - it was clear that they viewed us as a "trusted partner", rather than simply a "technology provider". In subsequent customer success meetings, they shared that we had “supported their value realisation above and beyond their expectations”. 

Over the course of 12 months, we improved the quality outcomes for this Customer by 42% compared to their previous year performance, with continued improvements being realised in the 2nd year of the engagement. 

Particularly in B2B markets and outside of the tech Industry, business outcomes are a priority for customers. If you’re focused only on product outcomes, you're missing the most important half of the big picture.

For analytics products (dashboards, data science or machine learning), do not overlook the value of training, education and ongoing consultation. Depending on the maturity of your product and technical organisation, this can be achieved by traditional means, or from a more product-led approach.

Lean into the challenges a customer is facing in realising value in your product - this does not mean you have built a bad product, or not achieved product market fit. This builds trust that significantly improves your chances of retaining that customer, leading to long-term brand loyalty that increases your customers lifetime value for this product and provides further opportunities with additional products that you can cross or up sell in the future.

Educate and collaborate marketing and sales teams to speak to the mapping of product outcomes to business outcomes

Do not be lazy or cut corners when engaging with prospective customers. Data related to product outcomes is a lot easier to source and far easier to achieve “superficially” impressive numbers. Business outcomes may require more effort, but are a far greater demonstration of a proven value proposition which carries significantly more weight in your market.

Discover more great content on Mind the Product