[Originally posted on LinkedIn]

What business value will your Data Project deliver?

The uncomfortable truth is that “AI & Machine Learning ” hype has not consistently delivered value for many non-tech organisations. They struggle to build dynamic teams capable of transformation with legacy systems, navigating political management and processes dating back over a decade.

Bottom line up front:
Begin by solving a business problem worth solving by being very clear on how your problem will improve revenue, reduce cost, or mitigate risk.

—————————————————————–

More detail:
Data projects have a high failure rate and Data teams often do not directly address the connection between business value and their activities.

– 85% of big data projects fail (Gartner, 2017)
– 87% of data science projects never make it to production (VentureBeat, 2019)
“- Through 2022, only 20% of analytic insights will deliver business outcomes” (Gartner, 2019)

So how can we mitigate this risk?

There are many approaches here, but I recommend starting with the bigger picture – starting with defining valuable problems to solve.

How does solving this problem affect the key pillars of value creation?

REVENUE: How will this directly or indirectly increase revenue?

COST: How will this reduce our costs? Cost of Ownership? Cost of Effort?

RISK: How will this reduce our likely or unlikely business risks?

Complexity (additional dimension) : How will this reduce our complexity? (Simplifying solutions often indirectly impact 2 or 3 of the other pillars) 

Don’t waste effort developing a hammer in the hope of finding enough nails to justify your time and resources invested. Begin by solving a problem worth solving.

Great follow on read 📚

For my brief introduction, Revenue, Cost & Risk- these are functional forms of value but they’re many more which provide clarity on different benefits possible from not only data science projects/products but for general endeavours. https://hbr.org/2016/09/the-elements-of-value