Boost sales in real life (IRL) with holistic AI

Sales is messy. No matter the latest SalesOps initiatives, sales folks will often do their own thing. They won’t enter competitor info in the CRM. The won’t read the latest product defense bulletin. They won’t submit a request to the win-loss team, and so on. Welcome to sales In Real Life (IRL). This article describes how to boost revenue using AI by recognizing, rather than ignoring, sales IRL.

Dealing with sales IRL requires a holistic approach. Naively applying Generative AI to sales IRL could make things worse, like putting rocket fuel into a car that needs an oil change. Results might seem impressive at first, but after 100 yards – phut!

Enterprise sales is a complex system. With complex systems, it is important to target the right leverage points for enacting lasting useful change.

This is hard for sales organizations that lack a design- or systems-thinking approach. Without these value-hunting tools, sales initiatives are often flying blind. This causes focus on the wrong interventions via reliance upon deck-of-cards assumptions.

To deal with sales IRL, you have to get real – i.e. get to the root of what’s really happening and why. Otherwise, AI will have either mediocre impact or even make things worse.

Design-first, not AI-first

The Chief Data Officer of Jaguar Group, as in luxury cars, had his own ideas about data. He focussed on data quality, believing it to be the cause of all ills. But, after user research (aka “Design Thinking”) he found out that data IRL told a different story. It was the user experience of internal tools that sucked, not the data.

Sometimes, the blockers to sales are deceptively simple. For example, a rule-of-thumb in software is that a UI should react within 100ms. Yet Tableau dashboards can routinely take 30 seconds to load. Worse still, no one seems to think it’s a problem. But, it is! Dashboards IRL affects sales! These seemingly “little things” add up and tank revenue.

Such missteps are also often the result of poor systems thinking. For example, many dashboards suck because the data warehouse is poorly configured. This, in turn, is due to some arbitrary IT cost cutting, optimizing a KPI that, in turn, is causing poor sales.

“They never fill out the win-loss report,” laments the SalesOps analyst. Well, 9 times out of 10, it’s because the experience is painful. Avoiding user pain by paying attention to users is the heart of design thinking. Design thinking means sales-first, not AI-first! AI-first is putting the proverbial cart before the horse.

Boosting enterprise sales with AI requires two preliminary ingredients:

1. Instrumentation – i.e. measure in what way sales tools are working or failing.
2. Design – give users what they actually want, when they want it and how they want it.

Let’s consider each in turn.

Instrument everything aka Observability

In the world of digital products, the buzzword is observability. Folks running complex sales systems are often flying blind, making it hard to measure sales IRL. There’s what you think sales folks are doing versus what they say they’re doing versus what they’re actually doing – IRL – with digital tools. Of course, you need to know the latter.

Let’s return to our car analogy. There’s a weird noise coming from somewhere and fuel consumption is up. You’d want a way to figure out what’s happening, right?

That’s what observability solves for sales IRL.

When sales numbers head South, if you can’t observe causes, you can’t fix them. To be clear, observability means insights into how sales are actually using tools IRL. This is different from measuring the usual metrics, like pipeline coverage etc.

For example, how many folks opened the latest CI bulletin? Which links did they click? Which of those led to a sales action? Which accounts? What revenue? And so on.

Design Thinking

The easiest way to fix sales is often to fix design!

We’ve all heard of user-centric methods, like design thinking, yet it seldom turns up in SalesOps daily practices. Many processes are not aligned to user needs, to sales IRL. Delivery teams go through the motions of shipping data products, yet there’s no clear user-centric outcome. Much of the time, it isn’t even measured.

Dollar for dollar, fixing design can yield more revenue lift than deploying yet another fancy sales tool. Boosting enterprise sales with AI comes after design thinking.

AI Acceleration

There is some unknown set of ideal sales actions that represents winning all winnable deals at maximize possible revenue given various parameters. The problem is discovery: which solutions at which price for which customers? We call the theoretical limit the Efficient Frontier of sales – i.e. where the information is perfect.

Approaching the frontier is always a trade-off. You cannot maximize all parameters because the future is unknowable and there are information gaps, including competitor insights. Selling solution A today might be better than selling solution B tomorrow, and vice versa. This applies across all deals, accounts and regions at once. Optimization is hard, but the only system that can attempt it is AI, or Deep Learning to be precise. (See article about AI Magic.)

Before discussing how to reach the frontier, let’s explore some key sales-boosting scenarios.

Boosting Sales with AI

Let’s consider some example challenges and how AI can meet them:

1. Sales Ramp
2. Partnerships Enablement

Sales Ramp Co-pilot

Sales people come and go, as do acceleration initiatives. With complex products and many solution sets, ramping new sales folks is a huge challenge. Overwhelmed with messy information, they struggle to discover and sell the right solutions. Opportunities get missed. Moreover, folks soon learn to hack the system. For example, they find ways to exploit (inappropriate) discounts to win deals. In doing so, they harm revenue.

Generative AI can help with progressive solution discovery by extracting the best contextualized data to win deals. Critically, it can do so within the context of portfolio optimization. For example, it could dynamically assign SPIFs to drive the right incentives that maximize total revenues. This is data-driven sales.

Partnerships

Here again, complexity is a killer. Any partner facing a complicated portfolio of products will favor the easiest ones to sell, potentially harming revenue. Worse still, if solutions are too hard to fathom, partners will sell other products.

Again, Generative AI can assist by extracting the best solutions from the available data. Moreover, if correctly configured, as part of a larger AI network, it can do so whilst optimizing for the Efficient Frontier. For example, it might recommend different solutions (A and B) to different partners for the same problem. Recommendations will take into account partner capabilities and market intelligence, as understood by the AI.

Allocation of resources can balance partner needs with optimizing the entire partnership portfolio. For example, the AI could decide where to assign deal-support consultants to best effect.

In a complex sales environment, sales enablement is key. However, it’s a limited resource: solution consultants are often spread too thinly.

Generative AI can help by extracting appropriate materials for those who don’t yet need a consultant. Meanwhile, it can route and prioritize allocation of consultants to partners most likely to win deals via enablement. It can also provide the right materials for those consultants, improving their efficiency.

Holistic Approach IRL

Sales IRL requires design and systems thinking to help deliver sustainable value where it matters. It is a prerequisite to the efficient deployment of AI. Design and systems thinking focus subsequent AI efforts on outcomes and value, not tools deployment. Otherwise, AI can easily become YAT – yet another tool.

The unique promise of AI is that it can deal with complex sales systems by handling information at multiple resolutions at once. It can balance individual deal performance against portfolio performance and strategic goals. But this is only if deployed in a systematic way using systems thinking approaches. This calls for a commensurate AI and data strategy that is sales focussed at the systems level.

If you want to know more about Holistic AI strategies for enterprise sales, please reach out.