People
Jason McDannold
Partner & Managing Director, Americas Co-Lead, Private Equity
Chicago
Artificial Intelligence (AI) and Machine Learning (ML) have rapidly transformed from niche capabilities to mainstream tools that more companies are using to drive real growth and operational effectiveness across commercial teams. Behind the hype and fascination with generative artificial intelligence (ChatGPT and its cousins) lies tested and powerful technologies that can produce quick, material, and even transformational improvements in a company’s ability to understand, serve, and support its customers. This has become very real: We are seeing more winning companies thoughtfully applying AI and ML to increase sales, cut costs, and improve customer satisfaction—all at once.
But surprisingly few companies understand or have seized the full opportunity. Instead, they opt for piecemeal approaches—improving analytics here, rolling out a new chatbot there. Lacking a strategic view of what ML and AI can do on the customer end, they’re leaving money on the table and giving customers less than they could.
The primary challenge facing business leaders today is not the scarcity of ML talent (which was the case just a few years ago), but rather how to harness the power of ML to create meaningful value impact. To succeed in implementing ML programs, business leaders need to approach the process strategically, combining technical and soft skills to identify the right use cases, select the most relevant attributes that drive the ML models for their business, and drive adoption/change management.
Based on what we have seen working well across a range of industries, here are key steps to get started:
First, select a focused portfolio of use cases that drive tangible business value.
We have seen companies bite off more than they can chew, and quickly get lost in complex, cross-functional implementations. The key to success is to realize value fast by focusing on deriving accelerated insights into the large volume of operational data on hand, making informed decisions on which use cases to chase in the early stages, and then scoping small, tactical applications to go after. In short, pursue progress over perfection.
Selecting the right initial use cases can be tricky, but should be driven by the business, not IT. This helps winning companies stay grounded in the core strategic intent without chasing shiny objects. Machine learning is especially applicable in use cases where prediction is required, such as customer acquisition and retention, which rely on propensity to buy and churn models or predicting customer lifetime value (CLV). Our experience says to make that decision based on a combination of enthusiasm and impact; You’re more likely to get a crucial early win if you concentrate your initial investments where you have the most engaged leaders and teams.
Second, focus on attributes that matter first, and then think big because you can.
Once you’ve identified your initial set of tactical use cases, the next step is to identify the attributes that matter for your business. What are attributes? These are key variables that drive success or key measurements and elements upon which the success of the business depends.
Choosing these key attributes associated with topline growth is easier than it may seem. A common misconception is that hundreds of attributes are required to build an effective model, but we’ve found that in most business applications, five to ten attributes account for 90%+ of model accuracy. This should entail collaborative cross-functional discussions with domain experts in the organization to boil down the few key variables that really matter to increase the likelihood of ML models generating accurate predictions.
Companies win by staying nimble and quickly changing their marketing and sales efforts to meet the dynamic needs of the market: AI/ML models must continually evolve to maintain high predictive capabilities and to continue to drive value. The good news is that changing these attributes in AI/ML models is much easier than with traditional data analysis. For example, in six states, data compilers (like Acxiom) are no longer allowed to share information regarding religious affiliation, race, and ethnicity, among others. We have seen companies with AI/ML solutions in place implement fundamental changes in attributes in a number of days rather than weeks or months.
An example: One of the market’s largest online training providers was struggling with pricing its training programs across segments and territories and struggled with defining dynamic pricing models and monitoring compliance across the reps on whatever prices were set. We quickly identified key attributes across enterprise customers, unique needs, and behaviors, and surveyed key market forces influencing potential pricing. Within 4 weeks, we implemented a new AI-enabled pricing model that tracked compliance and lost value due to unnecessary discounts, offered real-time pricing configuration for each deal as needed, and proactively alerted leadership whenever a salesperson was attempting to sell a deal with pricing that did not conform to optimal models. Immediate result: The client achieved $20M in improved revenue within 90 days.
Third, drive adoption and change management to capture value and sustain momentum.
Once companies achieve momentum with leaders who have been building confidence in their insights and performance, sustained governance and accountability become critically important. We have seen winning companies focus on the following measurements to continually improve:
- Shifting from reactive to proactive decisions:
The key sign of success is when cross-functional commercial teams (including Marketing, Sales, Customer Support, and Finance) are able to make decisions regarding investment, customer segmentation strategy, account focus, and sales effectiveness ahead of time, before accounts at risk attrit, or before customers are lost. Shifting the mindset to proactivity takes time as teams learn to trust the insights, attributes, and ever-evolving market demands. - Overcoming paralysis with accelerated insights:
Commercial organizations are often afraid to rock the boat, often relying on the idea that “whatever you do, don’t disrupt revenue.” We have seen companies learn within one to two quarters a new way of translating large volumes of chaotic data into useful information to make decisions on the fly. This new type of rapid information flow requires AI/ML to parse and make sense of the emerging patterns. The change is remarkable: Sales and Marketing organizations shift to a more rapid, agile approach to campaign spend, sales coverage, account strategies, and retention programs based on rapid insights. With speed comes value and a competitive advantage as competition is left behind with vague notions of actions to take. - Shared definition of success:
With visibility and alignment comes clear ownership to drive tangible growth. With ever-improving visibility into commercial operations and customer needs, we see companies moving away from anecdotal decision-making (“that customer will never accept that price” or “don’t disrupt our product portfolio”) to data-driven, rational, and proactive decision-making based on a shared vision of success. Clear accountabilities across the organization for leaders to meet specific goals also serve to remove roadblocks and friction. In short, the team pulls together as one, with a shared view of reality and an aligned view of what to do to win.
An example: Sustained discipline, empowered by AI/ML works. We worked with a large telecom provider who was struggling with slow enterprise sales and worsening churn across their enterprise customers. Their on-demand CRM solution hindered progress because the enterprise sales shunned the system, and as a result, account strategy languished. After launching a new AI solution that automatically tracked sales rep engagement (rep productivity and activity), account sentiment (based on tracked emails and phone calls), accounts at risk (neglected accounts), pricing compliance (money left on the table), and propensity to churn (accounts requiring immediate attention), we put in place a “Revenue Win Room”, which formalized a cross-functional commercial review of performance. The result: The team shifted to proactively focusing on key areas of the business to drive sales, realigned sales incentives to shape the right behaviors, and pursue high-value customers with a range of solutions and hands-on services. The result: $50M improved topline within 6 months, and over $100M revenue improvement within 18 months.
From vision to reality in weeks—AI & topline growth
As the examples above illustrate, winning companies can move quickly to gain better visibility into their operations and key attributes that drive growth and utilize AI and machine learning to move fast, change behaviors, work proactively, and drive results. AI can’t do it alone—it takes a coordinated program and a committed leadership team, but it can be done. And it’s getting better all the time.
