The Whale Curve: Actionable Information for Better Decision Making

We have all been in those meetings where we need to fix a sales problem. Everyone in the room has an opinion based on their experience and training on how to fix the problem. The trouble is far too often our gut and intuition are biased and they are not enough today. When we use bias to solve complex problems, we rarely solve them completely and this results in another meeting or two. It is the purpose of this post to establish that cognitive bias exists in sales decision-making. However, with clean actionable data and thorough analytics the risks of making a decision that negatively impacts the bottom line are greatly abated.    

When I was the Managing Director of Pragmatic Marketing, we gave customers coffee mugs that read:

“Your opinion although interesting is irrelevant”

We provided product management and marketing training for leading companies throughout the world. What we wanted to reinforce was everyone has opinions, but we must use current market data and requirements to shape your new products and growth strategies.

Albert Einstein provides a better quote on human behavior…

“If the facts don’t fit the theory, throw out the facts.”

One form of bias is Cognitive Bias.

Cognitive bias is a mistake in reasoning, evaluating, or remembering. They often occur as a result of holding onto one's preferences and beliefs regardless of contrary information. Biases are detrimental enough when they influence individuals. However, a cognitive bias in business is incredibly dangerous because it severely narrows the scope of perception of the decision-maker. This can lead to negative impacts, even in the most well-intentioned manager.

The good news is that there are a few steps to prevent bias affecting the decisions:

●      Awareness that cognitive biases exist and can distort thinking. Be on the lookout for bias in yourself and your colleagues.

●      Establish a consistent framework for decision-making.

●      Ask yourself if you have the right information to make a good decision. Be armed with data and reports rather than antidotes and narratives.

Case Study: Company A Attempts to Mitigate a Disappointing Q2 in Sales and Profit per Sale

It’s the middle of the second quarter and Company A’s sales are flat. Normally second-quarter revenues are the highest year-over-year. This is worrying to senior staff and they are concerned that is this is a bellwether for the rest of the year.  Is this a blip? Poor execution? Or is it a sign a recession is looming just around the corner?

The Senior Vice President (SVP) of Sales sees this as an opportunity to meet with his team and comb through their current and prospective accounts one by one to bring in additional revenue. To aid their conversations every rep is to come to the meeting armed with 4 years' worth of sales-by-customer, profitability-by-customer, and sales-by-vertical and have a growth plan ready to present. 

During the meeting with his number one salesperson, a 20% discount is requested to tempt big accounts to buy and move away from the competitor by end of the quarter and the VP learns that a small legacy account is having issues that require a good deal of rep time to mitigate. Another rep thinks that prospective-marquee-name-account will pull the trigger if they throw in application services for free and that another account wants last year's pricing, or they won’t buy more products or renew their service agreement. The third rep insists a new product is priced way too high and it’s really hurting the major vertical in his region. Ultimately, the SVP ends up fielding similar requests from his other reps also. After careful consideration, the SVP grants all of the pricing overrides and added value services and reports to the leadership team that he is confident the sales team will hit their number.  

The second-quarter ends and the team eagerly await the final numbers—but there is bad news. The team has come in 17% under target and profit per sale dropped 2%.  Why? Why did this happen we had a plan?

The answer—is that like many of us when we make decisions—the SVP has allowed his and his team’s biases to affect his judgment—which leads to an undesirable outcome.

This sales problem has elevated throughout the organization and now resulting in more meetings.

The (SVP) of Sales and the CFO have called a joint meeting with various departments to discuss why sales are not growing to plan and even more concerning why net profits are falling below plan.

Sales operations chimes in…I think we should help salespeople by creating lists for organic growth targets in their regions and have sales spend more time prospecting. I think they should do region-specific reports that give them insights into where the growth plans are not working.

Marketing shares: we have seen this before and back then we held sales more accountable to follow up on all the leads we produce. We also want sales to stop creating their own presentations and start using the tools we created.

HR says it sounds like we have a number of skills gaps in both salespeople and sales managers and it does not sound like sales managers are spending time coaching as we expected. We need to have the training to close the skills gaps in prospecting, key account management, and sales manager coaching.

The Pricing Manager says sales needs to use the prices we provided and stop overriding, rebating, and discounting and we will hit our targeted net profit numbers.

The CFO shares two slides. 

In slide one, it shows a graph of what our sales plan was for this year and where we are to date. Sales are up but not growing at the rate we planned and more concerning it shows the profit per sale is actually declining. The second slide forecasts what year-end will look like if corrective action is not taken and sales and profits do not improve. We need our salespeople to stop selling on price and start selling based on the value we provide.

The Sales SVP shares how the sales team is meeting their key activity objectives in the CRM like number of ends customer visits, number of net new customer meetings, numbers of calls on current customers, and number of training at distributors. I think we need better sales tools. My team is working hard and there are not enough hours in the day. I don’t want them going into BI and the CRM to create reports I want them out selling, customer face time with better sales tools! I am beginning to question if the market prices marketing and product management provided are accurate given how often my team needs to provide our distributor's rebates to win the business with end customers.

Meetings like this are and have taken place in many organizations over the years.

The good news is you have a smart and experienced team that that is committed to achieving the sales and profit goals you shared with the board and our investors. They are collaborating and working as a team to solve problems and meet cross-functional goals. It is not they don’t want to win; on the contrary, they are obsessed with winning and find themselves frustrated when the focused growth objectives are not being achieved. 

The trouble with the above conversations is each person is giving their opinions based on their own experiences and biases and not actionable data.

Humans are Hardwired for Bias

 Company A’s SVP has successfully led sales (for the last 15 years) to achieving year-over-year increases of 8%-12%. He uses the CRM reporting tool but overall trusts his gut for much of his judgment. It has guided him well in the past. However, unless he occasionally goes against his intuition, it hasn’t been put to the test. There is no way for the SVP to know if his gut is helping him make good choices unless he occasionally ignores it to see what happens.

“The value of analytics projects often has much to do with the psychology of de-biasing decisions and the sociology of corporate culture change.”- Jim Guszcza, Bryan Richardson, Deloitte Review

According to Noble Prize-Winning Behavioral Psychologist, Daniel Kahneman, humans have two “systems” of thinking. System 1—where judgments are fast and automatic, stemming from associations stored in memory—and System 2, which is an effortful, controlled way of thinking that—once engaged, has the ability to filter System 1. It can be quite dangerous to rely solely on System 1 thinking, because our intuitions, especially under stressful situations, often lead us to the wrong conclusions.

Everyone is susceptible to bias, especially if pressured or stressed. Reflect back to the SVP of Sales who is under pressure to “bring in the numbers.” He may be far from decision-ready in this situation, so he copes by relying even more on his intuition—which in this case—means he doesn’t deliver on his number.

 “Our comforting conviction that the world makes sense rests on a secure foundation: our almost unlimited ability to ignore our ignorance.”- Daniel Kahneman, Thinking Fast and Slow

 Kahneman theorizes that System 1 constructs a representation of a typical member of a population and then uses it to make judgments of other members. If actual facts contradict this interpretation, the brain finds it easier to ignore them. In other words, to our brain, the narrative we created about the typical member, beats the actual statistics about the member almost every time.

In an experiment that demonstrates how responsibility is diluted in a crowd, only 27% of participants rushed to save a choking victim. When students were shown the results of the study along with brief interviews with the participants—the students judged most of the participants as likely to help. The reason? The narrative their brains created around the participants as kind and helpful overwhelmed their ability to digest and process the disturbing statistics.

Additionally, the human mind is biased toward stories with positive outcomes, regardless of their veracity. If information is limited, the mind fills in missing pieces. The tendency of humans is to overestimate their prediction of events and to discard the anxiety and uncertainty of not knowing. Hence, nearly everyone thinks too narrowly about outcomes. Some make one “best guess” and stop there. Others hedge their bets.

Unfortunately, our ability to predict the future is terrible at best.

When researchers at the Harvard Business Review asked hundreds of CFOs to forecast yearly returns for the S&P 500 over nine years, their 80% ranges were correct only 33% of the time. That is a very low rate of accuracy for a group with vast knowledge of the economy. Interestingly, projections are even further off when individuals access their own plans because most of us tend to be overconfident—our desire to succeed skews our interpretation of the data.

 Company A’s Overconfident decision making

 Reflecting back to the SVPs dilemma—sales are not at plan and profits per sale are down—and his various actions are not resolving the problems, it’s clear that he and his team are biased in their decision making and actions. Discounts are approved because, in the past, when they are offered, the deal often closes. This, however, probably isn’t the case. The sales team is practicing System 1 thinking because they each have at least one positive narrative of offering a discount and closing a sale. Kahneman would theorize that this narrative seduces the sales team with the “illusion of understanding”, meaning that that the sales team's recall of the-time-they-closed-a-big –deal-when- they-discounted is inaccurate statistically. 

Hindsight bias makes objective assessment almost impossible. Other factors not remembered or known such as luck, timing, industry initiatives, and other external conditions probably were greater factors than the human mind is able to credit them for.

 Using Data to Outsmart the Human Brain’s Bias

 Despite the proliferation of Big Data news over the past decade. It was back in 1954 that psychologist Paul Meehl documented 20 studies comparing the predictions of human experts with those of simple statistical models. The studies ranged from how well prisoners respond to parole to schizophrenic patients' responses to electroshock therapy. The conclusion was the human experts failed to outperform the models in 20 out of 20 cases. Meehl refers to this as his “practical conclusion.”

In this age of almost limitless computing power and big data, it is hard to overstate the importance of Meehl’s conclusion. In virtually every and any circumstance statistical analysis can be used to augment better “expert” decisions. Additionally, if we recollect Kahneman’s thinking systems, statistical analysis, and the resulting data can kick System 2 into gear—allowing slower, critical thinking—resulting in better decision making.

However, recalling our SVP’s conundrum, as he had both CRM data and BI reporting—how did he fail in his decision making? In the context of his decision-making, the SVP was biased because of his own thinking, and the influence of the reps on his team and other managers in the organization. Moreover, while he had a plethora of reports, they only further confirmed his bias because he could only track sales by customer, profitability by customer, and sales by vertical. So, in his own mind, he was able to confirm price overrides and free services actually increase profits. What he didn’t have a view into was what each account was costing him and the net profit of each account. Hence, SVP had the “illusion of understanding” because he was missing this valuable data.

The Whale Curve: A Tool to Support Decision Making

 Traditional accounting methods and reporting are often insufficient to drive prescriptive actions. What is often not recognized or reported is which clients cost the most to serve and the least to serve. Distressingly, even managers who understand the issue are not able to easily distinguish between customers belonging to these two groups because they lack pricing analytic tools to make a sure determination. The total sales size of a customer does not always show the customer is automatically the most profitable; sometimes the largest clients are the most unprofitable.

Luckily there is a type of reporting called Whale Curve. 

The Whale Curve is a snapshot-in-time of cumulative client profitability allowing organizations to capture the cost-of-sales and net profit by the client.

The Whale Curve depicts accounts where the sales team is making healthy profit margins, accounts that are breaking even, and accounts that lose money. On the Y-axis, we plot accumulated net profit, and on the X-axis, we plot clients from most to least profitable. The resulting curve is said to be said to look like a whale coming out of the water.

In most cases, the most profitable customers create the most significant part of the organization’s profit. The Whale Curve creates this as an obvious visual graphic. The Curve generally rises initially (these are the most profitable clients), then stabilizes (break-even clients), and finally declines (clients where profit is lost.)  

As a rule of thumb, the Whale Curve for cumulative profitability reveals that the most profitable 20% of customers generate between 150 - 300% of total profits.  

The middle 60 - 70 % of customers break even, and the least profitable 10 - 20% of customers cause a decrease of 50 - 200% of total profits, leaving the company with its 100% of total profits.  

On the profitability Whale Curve, the difference between the highest point of the chart and current company profitability (100% profitability) represents unrealized profit potential for the company.

Thinking back to Company A’s decision making, if the SVP had a Whale Curve report, which provided him with sufficient data, presented in a useful manner—how might his decisions have been different?

●      He and his team would have an immediate (possibly alarming) read where each of their current accounts fell profitability-wise after considering the cost of sale, making the price override decision significantly less antidotal (System 1). The data rendered would force the brain to kick over to System 2, engaging critical thinking.

●      They may learn that the big client demanding a discount has actually been unprofitable for the last year.

●      Additionally, the report would provide clarity regarding no-charge services, as this would increase the accounts cost-to-serve and lower profitability.

●      Finally, if he ran a Whale Curve on products and verticals, he could determine that the new product isn’t priced too high, it’s already profitable, and that the vertical pricing in question is actually at-market.

Next Steps After Determining Profitability with the Whale Curve

The results of the Whale Curve often prove to be surprising, even more so when biased inclinations are proven incorrect. Using the Whale Curve, the SVP should have:

●      Given the sales reps the Whale Curve report. Discussed the profit-losing accounts and strategized solutions. It is crucial that this data is given to the individual who deals with clients.

●      Find services that cost the company money and remove them. Possibly move targeted accounts into a self-service model.

●      Sold profit-leaking accounts for additional products or services. The blended margin could make the client profitable.

●      Held a brainstorming meeting with the entire sales teams to discuss territory Whale Curves and brainstorm possible profit-leak solutions.

●      Engaged with a third party to deliver additional data and analytics.

●      Done a mid-year gut check

Actionable Information and Self Awareness Arm the User Against Bias

Ultimately there is no cure-all data set or reporting that prevents decision-makers from bias. Humans are naturally wired to seek out data and information that preserves their beliefs and decisions. Kahneman suggests we challenge our own decision-making bias by using these three questions:

1. Is there any reason to suspect the people making the recommendation of bias based on self-interest, overconfidence, or attachment to past experiences?
2. Have those making the recommendation fallen in love with it?
3. Was there group think or were there dissenting opinions within the team?

In addition to posing these questions organizations should constantly be seeking more data and consider using reporting like the Whale Curve that upends the typical, status quo sales analytics and depicts a new, actionable perspective.

For more information about bias mitigation, analytics, and the Whale Curve, please contact me at markrobertsnosmoke@gmail.com .