Overcoming the human obstacles to becoming a data-driven selling organization

Most sales executives I talk to are trying to do three things as they seek more effective ways to grow in 2021:

  1. Find ways to grow more for less. Most organizations are being asked to cut back on sales travel, field salespeople, and the resources that support them as they try to balance reductions in demand with the need to fund digital transformation according to the Markets in Motion report.
  2. Increase the precision in which they allocate these increasingly dear sales resources against markets, territories, accounts, and opportunities. In most organizations the return on growth resources – content, people, and digital selling infrastructure – is very low.  Every sales organization can dramatically improve their return on sales assets by better aligning sales resources and efforts with market opportunity, account potential, and buyer intent.
  3. Use big data and analytics to accomplish both.  Budgets for advanced analytics and AI, and algorithmic selling are growing in double and triple digits and now make up over 10% of the growth formula. In sales, these dollars are going to fund models, tools and simulations that attempt to better align sales resources and efforts with profit potential.  Strategically, organizations are using sales analytics and performance management tools to support data-driven account plans, territory definition, coverage models, and incentive design. Tactically, analytics are being used to drive algorithms that prescribe opportunity potential, lead priorities and next best sales actions.

This sounds good on paper, in earnings calls and board meetings. But these steps don’t stand up to the actual behavior of sales and marketing teams and their leaders when it comes to prioritizing leads, accounts, and time in day-to-day practice. 

It’s not because they lack the technology, data, or expertise to do so. Rather, the obstacles to becoming a data driven sales organization are largely human nature, according to Cam Tipping, who has managed hundreds of sales planning sessions and simulations with thousands of sellers over the past decade. “One of the biggest obstacles to leveraging advanced analytics in selling is the fact that sales leaders, managers, and front-line reps cannot overcome human, instinctual, and entrenched behaviors that fly in the face of optimizing and automating sales resource allocation,” reports Tipping. “This dissonance is true at the individual and organizational level.”

Jeff McKittrick, a Managing Director of the Revenue Enablement Institute echoes this sentiment. “One of the most difficult things for a salesperson to do is treat customers and prospects differently and not run down every fly ball, even when the data tells them it’s not in their best interest” reports Jeff McKittrick who has over a decade of experience leading sales enablement programs at Cisco and Hitachi.  ‘It’s very hard to ignore RFPs that are obviously written for the competition, not spend time with unprofitable customers you like, or waste time on prospects that are more than likely “tire kickers”.

According to Tipping, the human factor, gut instinct, and deeply ingrained behaviors and beliefs, can get in the way of optimizing and automating sales resource allocation and priorities. These human factors go a long way towards explaining why, while almost three quarters of sales reps (74%) say they use data driven insights to direct their behavior, the majority still miss their quotas according to research by Salesforce.com.  Until growth leaders understand and embrace these dynamics, they will continue to trust their “gut feel” more than they trust machines, black box algorithms and the data on which they have spent millions to help them better allocate resources.

  1. Don’t let perceived opportunity cost trump objective optimization. “Salespeople all understand the 80-20 rule, but they are reluctant to put it into practice because they are too afraid to miss out on business,” relates Tipping. The 80-20 rule was famously discovered by the Italian engineer, Vilfredo Pareto, when he discerned that 80% of the Italian wealth was controlled by 20% of the population. The same, or similar, rule applies to selling according to Cam Tipping.  Measured in value or profit, the top 20% of customers in almost any market purchase most of all the products or services. On the other end of the spectrum half of your client base are low purchasing customers or, in other words, the biggest group of customers only accounts for 20% of what you’re selling. Putting this rule into day-to-day practice is a bigger issue. “When they are confronted with hard data and facts that show most of their target customers are poor candidates for sales resources, only a fraction of salespeople and their managers have the discipline to not sell to those customers,” reports Tipping. “After running sales planning simulations with over 5,000 professionals, I can count on two hands the number of sales reps who were able to resist the urge to apply at least some sales resource and effort to every single customer in their territory.”

2. Don’t apply the scientific method without first having a hypothesis to test. Most sales leaders intuitively understand their markets.  But collectively very few organizations can agree upon, describe or even document how they grow and quantify the cause-and-effect relationship between commercial activity and sales (the sales response curve).  Businesses will create profitable growth and firm value from their investments in analytics and AI if they ask the right questions of the data. This means defining an organizational growth hypothesis on a company, product, territory, and sales rep level, then testing it, proving it right or wrong, and refining it over many tests. “A big part of the success from any analytics endeavor is how one frames the problem,” relates Professor Raghu Iyengar, who leads the Wharton Analytics Initiative. “Having clarity on what you want to test (the hypothesis so to speak) also helps one understand what type of data to collect.

3. Treat AI like a tool rather than a goal – In their zeal to become data driven, managers tend to get ahead of themselves and make AI and machine learning a goal rather than a tool to help them achieve the objective of profitable growth.  Instead of focusing on AI tools and techniques, managers need to be spending most of their time asking the right questions of those tools.  “At the end of the day, technical sophistication should not be a substitute for better thinking,” continues Professor Iyengar, who teaches executives how to use advanced analytics to grow their business . “What I mean by that is understanding your customer journey, the role that you can play in that journey, what are the critical hurdles etc.…these questions haven’t changed. What has changed is how you can address them using AI and ML. So, before embracing the shiny new object, think about the customer journey, and where AI and ML can have the best returns.”


4. Don’t let perfect get in the way of profitable.  Organizations often fail to leverage analytics because they expect perfection. For example, some executives demand perfect data before they can make a decision. For others, data can be a security blanket that provides them a false sense of precision. Organizations adopting advanced analytics tend to let analysis in a vacuum overwhelm common sense and human nature in their rush to algorithmic and data-driven way of reallocating sales resources to the right accounts, territories, and markets.  This is one of the simpler and more impactful ways data-driven algorithms can create value today with limited analytics acumen and data scientists in short supply. But we live in a world of messy data and fast-moving market dynamics – so it’s important for managers to understand the difference between “good enough to be actionable” and 100% precise. “It’s more profitable to be vaguely right than precisely wrong” warns Leonard Lodish, Professor of Marketing at Wharton in his book the Advertising and Promotion Challenge. In it, he argues it is more important for businesses to balance precision with action by embracing imprecise but actionable measures of sales and marketing effectiveness than over rely on reams of statistical data that hide the real truth.  

5. Remember nothing happens without Implementation. “AI and machine learning will account for nothing if your sales team doesn’t implement the outcomes,” says Cam Tipping. “This takes more than a few PowerPoint presentations asking people to change their call behavior.  Remember behavioral change is a process not an event and it’s more likely to take place when people are allowed to discover the reason why rather than being told.  Leaders of change know that a presentation is a poor substitute for an experience.”       

So, what can sales teams do to balance computer algorithms with common sense? A good way to overcome these human obstacles to leveraging analytics in sales is to get humans more involved in defining, gaming, and vetting those algorithms. 

Using a structured business simulation – or game – as a planning tool to define resource allocation and growth strategies is a structured but collaborative way for sales teams to forge their disparate knowledge into a consensus plan that can be tested and refined.  These games capture the collective knowledge, experience, and judgments about their markets and sales model. But then they quickly and iteratively create a company-wide consensus on the fundamental relationships between resources, response and yield that underlie all sales algorithms. A business simulation combines the energy and competition of a workshop with the scientific discipline of a test, learn, and iteration cycle.  Using simulation tools in a planning context gets teams working together to solve real resource allocation problems. They also have the advantage of “compressing time” so feedback loops are shorter, and teams can learn, evolve, and refine their business assumptions faster.

Professor Iyengar of Wharton agrees that getting a broader cross section of management and sales input into models is important to institutional learning and the development of algorithms that reflect market reality and sales experience.  “All companies need to embrace a culture of test and learn,” relates Iyengar. “It is through continuous iteration and feedback from users (customers) that one can start optimizing how to leverage AI in the best possible way.”

You May Also Like