Building Internal AI Capabilities: How company culture impacts adoption

Building internal artificial intelligence capabilities requires a strong cultural foundation. Businesses must go through a progressive paradigm shift where a culture of innovation first exists, and then there’s a progression to a culture of fast, probabilistic, and data-driven decision-making. 

In April 2019, Deloitte published an executive survey that concluded most companies lack analytics-driven cultures. The survey found that insight-driven organizations represent a minority of businesses. It also uncovered that 67% of executives were uneasy about applying insights delivered by data tools and other resources and that 62% still used traditional tools such as spreadsheets. 

We’ve come to expect disruption from tech companies like Microsoft, Amazon, and Google. Mostly because being disruptive requires a healthy investment in research and development. But more businesses must be open to taking risks and investing in new technology, along with being willing to embrace the fundamental operational changes necessary for making decisions based on probabilities. These changes include:

  • Moving from silos to interdisciplinary collaboration. Cross-functional teams with different skill sets and perspectives should work side-by-side to map problems to data and then create solutions.
  • Transitioning from experience-based or leadership-driven decision-making to data-driven decision-making. Avoid the top-down approach where decisions require input from a manager. This process limits the possibilities of AI.
  • Explaining to the business why you’re considering AI. Implement an open-door policy to talk through the thought process behind your AI aspirations.
  • Anticipating and talking with employees about barriers and fears. Prepare for common concerns, such as job loss from automation, or the outright rejection of machine-generated insights. With proper coaching and transparency, many of these concerns can be mitigated.
  • Abandoning top-down decision-making. As AI gets better at making decisions and predictions than seasoned leaders, it makes sense to shift from leader-led or experience-based decision-making to data-based decisions (made by those on the front lines). This requires trust of the algorithms and empowerment at the line level.
  • Replacing a rigid and risk-averse mindset with agile, experimental, and adaptive ones. Make “move fast, test, and learn” your new mantra. Ideas don’t need to be fully baked; solutions don’t always need full functionality to deploy. Getting user input as early as possible is less costly and faster than fixing major problems. 
  • Setting AI budgets that rival IT budgets. Allocate 50% to AI tech development and 50% to things like integration, adoption and change management. Activities should include workflow redesign, communication, and training. 90% of companies with successful scaling practices spent more than half their analytics budget on adoption.

Want more insights into how you can scale up your internal AI efforts? Download our free whitepaper here