Four AI Misconceptions that are Costing Retailers Money and Time


Artificial intelligence (AI) and machine learning (ML) have quickly become buzzwords in the retail industry.

With companies like Nike, McDonald's, Nordstrom, and Walmart purchasing retail AI firms, many retailers are prioritizing AI in their budgets for the next several years. But as these technologies become more embedded in the industry, it's important to understand and clarify some common misconceptions lest they end up costing retailers a lot of money and time.

Given that AI projects can take a year or more to launch, no business wants to learn—when it’s too late—that they won’t receive the desired outcome.

If you're thinking about implementing an AI/ML solution for your retail business, here are some common AI misconceptions to watch out for.

Myth 1: AI has a singular definition

AI is an umbrella term that can refer to many different processes, including machine learning, data analytics, predictive analytics, statistical methods and more.

In order to put AI to work for your business, you must first understand what problem you are trying to solve and then apply the appropriate segment(s) of AI that will help you find the solution.

Myth 2: AI will solve all your challenges

AI is often viewed as a magical solution that will address all of a retailer's challenges. But AI alone is not enough.

Instead, AI needs to be applied as part of a common analytics platform that accounts for multiple processes in your business and the factors that affect them. This platform also needs to allow you to build business rules and exceptions to automate this process.

An AI solution also needs to be business-specific to actually be useful and practical. Every retailer is unique and has specific constraints, challenges, opportunities and attributes, so there is no such thing as a one-size-fits-all AI solution.

Myth 3: All AI service providers are created equal

When looking for a vendor to help with AI initiatives, retailers should be wary of any company that claims to "do AI." In reality, it may be some fancy data mining with statistical analytics and a nice dashboard.

Moreover, general "AI companies" that work across industries may not be focused enough to give you the accuracy required for a large and complex retail organization. It's wise to seek out a vendor with specific experience in the retail industry addressing similar challenges.

Myth 4: Anyone can start using AI tomorrow

One of the biggest barriers to adopting AI is your "AI readiness." In order to get the benefits that AI has to offer, retailers need to make sure they are prepared.

First, you'll need to decide which specific problem in your business you’d like to solve by leveraging AI. From there, ensure that you have answers to the following questions:

How to prepare your data to use AI for a selected task?

• How will the selected AI solution affect your business processes?

• What business roles will need to be changed and how?

• How will it be integrated with the rest of your technology ecosystem?

• What KPIs will you use to measure the success of this initiative?

Truth: You can implement AI, but it needs to be done the right way

AI can be very valuable to your business, but don’t rush into it without a clear plan. Once you have your data ready and your goals defined, you can decide which specific AI tools are applicable and how they fit into your current infrastructure.

--Mark Krupnik, PhD, is the founder and CEO of Retalon, an award-winning provider of retail AI and predictive analytics solutions for planning, inventory optimization, merchandising, pricing and promotions. Mark is a leading expert on building and delivering state-of-the-art solutions for retailers.

This ad will auto-close in 10 seconds