Artificial Intelligence is Here to Stay. So, What’s the Impact on Sales and Sales Compensation?

Arun Shastri co-wrote this blog post with Chad Albrecht

While the concepts of artificial intelligence have been around for decades, the recent acceleration has been astronomical. We’ve seen advances in big data processing technology, image and voice recognition, complex computing capabilities and more.

While we may think of personal applications such as Alexa, Siri and Cortana when we think of AI, several applications in sales and marketing organizations have already begun to significantly impact revenue. By some accounts, more than 50% of the impact generated by advanced analytics are in marketing and sales. So how exactly has AI impacted sales and marketing?

Customers now dictate how they interact with your company, not you

This could mean email, social platforms, chatbots, live phone calls, brick and mortar stores, or something else. This customer-centric way of viewing sales and marketing means that the salesperson plays just one of the many parts—albeit an important one—to increase revenue. And they must seamlessly coordinate and integrate with technology to optimize who they call on, as well as when and how.

Score the potential of an incoming opportunity

Machine learning (a part of AI) models are regularly used to score the potential of an incoming opportunity, which in turn is increasingly used to direct salesperson activity. These models continue to improve and become more accurate as more data is captured, allowing better optimization of which accounts should be targeted. And not just which accounts to target, but how: What are the best methods of communication (email, phone)? What are the best times to reach them, and how many times should you call? This eliminates a long period of slow human experimentation by sales leaders, and often less-than-optimal results.

Identify the next best action to engage the customer

Another common method in which sales is being transformed is through a class of machine learning models generically termed “next best action.” For years, digitally native companies such as Amazon and Netflix have run such models to deliver value to customers (“Customers like you also liked…”). For sales organizations, an orchestration engine can optimize for the next best sales action and communicate this to the required channels. Next-best action models are being leveraged with sales organizations to address a range of questions for each customer: what product, what content, which channels and when to contact them. In addition, machine learning models can also identify upsell and cross-sell opportunities using the same logic.

Anticipate customer churn

Advanced data science has also developed models that predict the other side of revenue growth: customers who are most likely to churn—or perhaps downgrade—in the very near future. These models are immensely helpful in triggering preventative measures for the account managers responsible for these customers. While these models have been around for a while, they continue to get increasingly sophisticated and accurate as more data is captured and results are evaluated against predictions.

Optimize pricing at the account level

Another area in which AI is supporting sales is in the area of pricing when it’s feasible to optimize pricing at an account level. Tools that optimize pricing mean higher revenue that drops straight to the bottom line, so getting salespeople to adopt and adhere to the recommended pricing models is critical to growing revenue and profitability. Without expert knowledge about a customer’s willingness to pay based on a complex algorithm, salespeople too often have a habit of “racing to the bottom” of the allowable discount range, leaving money on the table.

With the customer at the center of all sales and marketing activities, the role of the salesperson is evolving. They will increasingly use technology to identify the best leads, learn how and when to best reach them and the best way to approach them, and the product that they’re most likely to be interested in. Given this change in how they operate, we need to determine whether their motivation programs should change and how we should measure their effectiveness. We’ll explore this topic further in our next blog.

A version of this post was originally published on ZS’s blog The Carrot.