Salespeople operate in constantly shifting environments. Every day they contend with new situations and changing information. No two customers are the same.

To remain in control, companies deploy a CRM as their system of record, or “single source of truth”. They ask the sellers to keep the system up to date – if it’s not in Salesforce it doesn’t exist – and then, armed with bucket loads of data, they generate reports and dashboards to try to understand the business. While easy to generate, reports and charts are often hard to interpret. Often the glut of data makes it hard to assess what is relevant and what is just noise.

What options are available to sales organizations to address this challenge?

Artificial Intelligence: Over Promised, Under Delivered

An abundance of product offerings using artificial intelligence or predictive analytics have recently emerged as potential solutions to solve the problem for sales teams. However, in the context of enterprise sales, these approaches fall short in two ways.

  1. First, enterprise sales is a small data problem, not a big data problem — there is not enough consistent data for an individual rep to draw accurate conclusions from patterns — and the individual seller is a key variable.
  2. Secondly, any functioning artificial intelligence system must first be taught the intelligence. It can’t learn it on its own. The system needs a base of knowledge from which it can make assertions based on data, and it needs to understand the context of the problem it is solving.

Relying on machine learning, or pattern based predictive analytics, alone to deliver sales insights [in enterprise B2B sales] is dangerous.  The problem with Artificial Intelligence (without a foundation of knowledge) is that it frequently mistakes correlation for causation. Just because the first 43 US presidents were white males (clearly) doesn’t mean that the pattern will continue.

If Google Maps did not have maps with knowledge of all the places in the world on which to base its routing algorithms, it could not give you directions.  It is the same in enterprise sales AI systems.  You need a map to start with.

Google’s driverless car had its first accident in February 2016, “because it did not know that a bus was less likely to give way in traffic than a car.” No one had taught it that. It is the same in enterprise sales AI systems.  The system first needs to be infused with knowledge before it can make decisions or recommendations. Then through practice it can get smarter.

Here’s another example.

Did you know that there is direct correlation between a person’s shoe size and their reading ability? It is true, at least in the United States. If you gather together everyone’s literacy score and plot it against their shoe size, you will find that those with greater reading ability tend to be those who have larger shoes. The correlation holds true in fact everywhere in the world. The reason, of course, is that one’s shoe size correlates more significantly with age, and most babies or young children (with small shoes) are not great readers.

Context matters and understanding the nature of a correlation as it pertains to a specific domain is essential to deriving meaningful insights. Before you look to the data to provide answers, you must be really sure that you know the right questions to ask, and the circumstances or frame of reference in which are operating.

Augmented Intelligence: The Practical Answer

While automation has recently been a key trend, we know that decision-making in complex domains is still safeguarded by humans. The idea of autonomous technology making decisions and recommendations for us may sound attractive, but unguided it delivers sub-optimal results and imposes business risk as well.

We are better served focusing on Augmented Intelligence, a perfect blend of human insight and knowledge with powerful computing abilities – pattern recognition and machine learning – to leverage both man and machine.

There are five main steps to the approach we have taken with for Augmented Intelligence for Sales:

1. Decide what data matters – the key Insight Signals – that need to be monitored and acted upon. In the case of managing a sales opportunity, these Insight Signals will likely include Deal Size, Pipeline Stage, Competitors, Executive Relationship, Qualification Status, Customer Requirements, Previous Account History and so on.

2. Employ an existing base of knowledge and derived insights – say a codified sales methodology – in the CRM where your opportunity data resides. Examples of such insights might be:

  • If your competitor has a stronger offering for one of the customer’s critical decision criteria, then either qualify out or take actions A, B and C to change the customer’s perspective on what is really important;
  • If the opportunity value is greater than $500,000 and the seller has only identified one buying role in the customer’s organization then point out that for a deal of this size it is likely that there will be a buying committee and take actions D, E, and F to broaden coverage and gain access to all of the key players;
  • If your average winning sales cycle is 100 days and the opportunity has only progressed to the second stage of the funnel in 90 days then there is clearly some risk. Highlight the risk of the slow moving deal and recommend actions G and H to re-invigorate the deal or recognize that the deal should no longer be in the sales funnel.

3. Let the machine do its work. Utilize the consistent and (effectively) limitless computing capabilities that are now easily available to monitor all events and data, assess impact of change by applying the knowledge and insights to the data, and then notify the user of the need to act with prescribed knowledge based actions to progress the sales opportunity.

4. Learn from experience. As more opportunities are processed, the system can learn the effectiveness of each insight, the competency of the individual users – if the same deal vulnerabilities are present for a sales person time and time again, it probably points to a competency issue. Then the insights can be refined to reflect that learning.

5. Enable additional knowledge infusion by the business. One of the challenges with traditional Artificial Intelligence and Predictive Analytics solutions has been that all of the power is in the hands of the data scientists who are typically too far removed from the knowledge of what works in the business. To ensure that the knowledge in the system is optimal for the business and reflective of the peculiarities of a specific set of circumstances that pertain only to that business, the business users need to be able to add addition Insight Signals and Insights or Knowledge that is learned in the everyday running of the business.

-o-

The dynamic nature of sales engagements and the differences between every customer requires solutions that identify and adapt to those differences, solutions that learn and grow by assimilating the experiences and outcomes so that others can benefit from the knowledge gained — based on the ‘muscle memory’ of millions of successful and failed sales engagements.

Augmented Intelligence enhances the performance of the people who use it, increases engagement and improves productivity, while also evolving as the knowledge base of the users grows. It starts with knowledge embedded and codified “in the box” and improves with the learning that comes from using the technology.

Most sales people understand that they need to connect with the right people and take inventory of the business problems those stakeholders and their influencers care about most. They need to identify the pain points that their buyer is ready to spend money solving and then they need to demonstrate that their solution solves the problem equally or better than the competition. Wouldn’t it be helpful to the sellers, and the sales organization as a whole, if the system they use ‘understands’ these factors and monitors all of the opportunities on their behalf?

Augmented Intelligence improves on traditional AI. It acts as an extension of the people using them and fulfills the dream of encapsulating ‘best sales person behavior’, following the salesperson through the sales process, and adapting to each new step of the customer journey in real-time. Unlike artificial intelligence alone, augmented intelligence doesn’t follow just look for patterns in the data, but anticipates actions, and when combined with insights, can dramatically improve the results of the sales team.

Augmented intelligence bridges the gap between salesperson and software to make sales people more proficient at their jobs, more professional in their engagements and more valuable to their customers — which, ultimately, is our primary goal.


If you want to read more of my blogs please subscribe to Think for a Living blog. Follow me on Twitter or connect on LinkedIn. I want you to agree or disagree with me, but most of all: I want you to bring passion to the conversation.

Donal Daly is Executive Chairman of Altify having founded the company in 2005. He is author of numerous books and ebooks including the Amazon #1 Best-sellers Account Planning in Salesforce and Tomorrow | Today: How AI Impacts How We Work, Live, and Think. Altify is Donal’s fifth global business enterprise.