Felicia Chen
Felicia Chen
Updated: August 17, 2021

Four Tips for Successfully Implementing AI and Automation at the Enterprise Level

Four Tips for Successfully Implementing AI and Automation at the Enterprise Level

Although AI stirs up grand visions of the future, it’s also gotten a bad name. There’s a lot of hype around AI, but there’s a lot of failure stories too. It’s possible to prevent your own AI initiatives from also failing. Sidney Madison Prescott, Global Head of Intelligent Automation at Spotify, shares her tips on how companies can succeed with AI and automation.

Understand AI and automation are tools for business optimization

Most business leaders understand the value of AI and automation. If you need to make the business case for it, you’ll have to make sure company executives know what the value of these tools is.
AI, machine learning, and automation are all ways to optimize and modernize your business processes. All these tools work together to provide you with “a cohesive way to address your business problems.” They create a compelling value proposition because they can facilitate a digital transformation that moves the business forward and puts you at the forefront of your industry.

Take a phased approach

According to Prescott, AI and automation initiatives have often failed because:

[T]hat’s primarily because [you’re] biting off more than you can chew [or] going for moonshots instead of low-hanging fruit.

There’s a lot of different steps you have to take to actually prepare yourself, prepare the environment, prepare your workforce for this change. And when you skip past those steps, that’s when you start to see the hurdles and the challenges in relation to actually being able to execute successfully on a vision to have an artificial intelligence imprint in your company or build machine learning models in your company.

This is why Prescott emphasizes how important it is to understand that adopting these tools will be a phased approach.

Prepare and start with clear, realistic, and short-term goals

For Prescott, preparation is key to succeeding with AI. “There’s a lot of hype around the AI piece, but we often don’t hear a lot about what it takes to prepare your company and even your employees for that initiative. And I believe that’s what sets you up for success.”

To be successful, Prescott advises companies to start with automation. Automation is the biggest indicator that your company is ready to advance to AI. When you are ready, the best processes to apply AI to are mature processes that offer a high volume of data and where the data is reliable, accurate, and complete. Once you’ve identified which processes are ripe for AI, you need to deeply think about where you want to go with AI, be realistic about what you can achieve, and start with your short-term objectives. Then, you can work toward your long-term key objectives and results.

Take a top-down and bottom-up approach

Prescott says about 30% to 40% of companies who attempt AI fail within the first one to two years. Spotify is a rare success story though. In 2020, Prescott rolled out 100 bots across the company. Her secret to success is taking both a top-down and bottom-up approach.

Going from the bottom up, they trained employees who were on the ground to identify automation opportunities and understand where processes can be optimized. By knowing how to look for use-cases in their daily work and in their work with other teams, employees across the company helped bring automation opportunities to Prescott’s team. Instead of keeping their initiatives at the senior level, Prescott’s team engaged the entire company. As a result, they sourced a wide variety of use-cases and had a better understanding of cross-functional work and systems.

At the same time, their top-down approach helped them understand the reasons why senior leadership would embrace digital transformation. They identified the goals that Prescott’s team could help them achieve. This helped them automate processes that would provide ROI for multiple stakeholders and teams.

Most companies will only take one approach. They’ll train employees to automate some simple tasks, or they’ll go for such large enterprise-wide initiatives that they miss great opportunities that are smaller. By using both a bottom-up and top-down approach, Spotify was able to scale their automation and AI efforts rapidly.

To learn more about implementing AI in your business, see how chatbots and virtual agents are disrupting the customer service industry.

Sign up to be notified of upcoming Replicant events to engage with and learn from AI experts like Sidney Madison Prescott.