Recently, we conducted a series of research surveys and in-depth, one-on-one interviews with inside sales leaders. The purpose of the research was to understand how teams are managed today and identify challenges/opportunities for increasing productivity using advanced analytics, namely Artificial Intelligence (AI) and Machine Learning (ML).
The chart above shows how 200 inside sales leaders responded to various next generation sales management system features. Of note… nearly 80% of respondents use Salesforce.com for their CRM system, and 50% use various dialing and emailing automation tools in conjunction with CRM.
Most In-Demand: AI/ML-Guided Selling
Sales leaders strongly believe that “Guided Selling Actions Using Advanced Analytics” would increase sales team productivity. We clarified in Q&A interviews what people meant by this, and specifically, teams are looking for a way to help their reps decide what to do next, leveraging data-driven insights. The type of guidance teams seek is a combination of algorithmic, to rank leads and activities by sales stage conversion probability (generally MQL to SAL for high-in-funnel efforts), and alignment to the managers defined strategy and the rep’s own work capacity.
Part A: Lead Scoring
Lead scoring has been around for nearly 10 years. One of the earliest lead scoring companies, Lattice Engines, recently sold itself to Dun & Bradstreet. Salesforce offers lead scoring through Einstein. And there are others. But most teams don’t use them, as we discovered that of the 65% of teams that say they use lead scoring, 75% of those rely on marketing automation scores, not AI/ML modeling, which only accounts for less than of 10% of today’s lead scoring. Internal data science teams account for the remainder.
There are two major drawbacks to today’s version of lead scoring. The first is that scores are passive and not tied to the selling process. What should the sales rep do, based on a score? If something scores a 100, it’s easy. Dive in! But most leads are in the murky grey zone. What does a 72 or a 62 mean? Should it be worked on now? Is this new one better than other leads already being worked? Hard to say… and scores change during the engagement lifecycle. Now what? Reps are on their own to figure out what to do when, which means that scores are not applied systematically.
The second challenge with lead scoring is that many reps and managers don’t believe the scores. Their experience shows that scores are not always accurate or helpful, and they come from Marketing which is another planet for most sales teams. Sad but true. Marketing automation scores also don’t use modern analytical methods such as AI/ML. They use bucketed attributes on a 5- or 10-point scale, with point totals added up. When a threshold of points is reached, bingo, that defines an Marketing-Qualified Lead (MQL). This approach only looks at a handful of attributes without statistical evidence of their correlation to sales success, when there can be 20-50+ behavioral and factual elements tied to a lead through time and much more powerful/accurate analytical methods.
There’s a lot of room for improvement in lead scoring, to increase the usefulness and accuracy of scores.
And… lead scoring is just the beginning of the sales process. It accounts for zero of the sales effort, and has very little impact on the the “how” of selling — what sales reps do, when and why. In a manufacturing metaphor, procuring ingredients is important, but not as important as the actual making of the finished good. In the cooking metaphor, what the chef does matters. Sales teams make things, and using AI/ML there is what defines NextGen Smart Process Automation.
Part B: Smart Process Automation
We found that is it a struggle for teams to follow their own plans and processes, mainly due to the sheer speed and quantity activities in inside sales. Having 10’s, 100’s and even 1,000’s of open leads in your name, each with a 5 – 15 touch process… means scale and complexity. It’s easy to lose track of what should happen next.
Existing CRM features and add-on sales acceleration technologies do not use AI/ML to help reps decide how to allocate their time — to manage the overall process, which is different from scoring an individual lead. How should the rep invest his/her time across multiple campaigns? How to align to the manager’s strategy, such as 70% outbound prospecting? What is the rep’s capacity, so that enough new items are started, but not too many, which can lead to mid-process abandonment. Today, there are many places in the sales process where reps are on their own to figure things out, and manager coaching is offline, when there are modern analytical methods and process automation technologies that can help tremendously.
What Defines “NextGen” for Inside Sales Management?
A next generation, smart sales process management solution guides the sales process. Like Waze does for drivers. A NextGen sales solution helps reps determine what to do when, such as when to engage new leads vs. focus on existing, and how to balance effort across multiple processes, so that the manager’s plan is implemented. It uses insights from data to do these things on the rep and manager’s behalf to simplify complexity and adapt to a ongoing change, so reps can do their thing: sell.
And just as importantly a NextGen system learns, continuously. We observed that today’s sales technologies are largely in “set-it-and-forget-it” mode. The initial process is what it is… forever. The goal of these tools is implementation, not improvement. Some tout A/B testing as a feature, but this is not a scientific approach. It’s manual, ad hoc, and both the A and the B processes are hunches. In contrast, a modern process, using advanced analytics gets smarter about itself using AI/ML, and systematically explores new paths, continuously, to finds new ways to improve productivity.
There is overwhelmingly strong demand for NextGen AI/ML sales management technology.
Agree responses above averaged 65% and Disagree 8%.
More to Come…
This article is the first installment of a series of articles and survey results that we’re compiling into an e-book. Stay tuned.
If you like this article, please like and share.
And please comment to let me know what you think!