Is a self-discovering AI taxonomy a reality for revenue enablement tools? I don’t think so, but I believe the industry is heading in that direction. We definitely need the functionality, because too many B2B enablement tech deployments still resemble buying a Maserati only to drive it on Sunday mornings. (Think of the 45% in the best-guess, not-research visual below.)

Utilization of Revenue Enablement Platforms

The core mainstream value of revenue enablement platforms (REPs) lies in surfacing the “what to show” and “what to know” assets for sellers at each inflection point in the buyer’s journey. Every time the CRM-housed sales opportunity evolves — whether through today’s manual updates or tomorrow’s automated capture — enablement teams should push recommended selling and learning artifacts to sellers. For example, when a rep updates a CRM record for a financial services prospect from 10% to 40% likelihood, the REP should immediately present them with selling and learning assets that are proven (or believed) to be effective for opportunities with those new attributes, for that particular flavor of buyer. Currently, we see only half of REP owners doing this (the 50% pie slice).

This approach is essential and provides two crucial benefits: (A) Reps spend less time searching for resources because the content they need appears directly in the workflow and (B) by providing core support in the context of sellers’ deals rather than at their own convenience, enablement should enjoy better adoption, reputation, and linkage to hardcore revenue outcomes.

REPs have offered this capability for years. Without sufficient investment in fine-tuning and scaling accuracy, however, these platforms often overwhelm sellers with more information than they can reasonably process. The issue lies in the fact that standing up a properly CRM-integrated revenue enablement platform remains a heavy lift. Most B2B sales organizations manage hundreds or thousands of buyer-facing and seller-supporting assets. Manually tagging each one with buyer persona, industry, journey stage, and product offering is at odds with the original idea of saving resource costs by automating the provision of selling and learning artifacts to reps at precisely their moment of need. Many REP buyers mistakenly believe they’ve purchased a plug-and-play solution and fail to account for the resources necessary to effectively position and maintain the tool. And let’s be honest, the temptation for vendors to sell that concept is a powerful siren among vendors.

The real gap lies primarily in the taxonomy work — all the content tagging — that enablement teams can perform but often lack the resources to execute at scale. This is where the promise of generative AI comes into play. Imagine directing a large language model across the entire enablement ecosystem — buyer-facing assets, seller-supporting learning and development pathways and courseware, win/loss analytics, rep performance metrics, and more — and instructing it to discover and tag thousands of items based on the variables mentioned above. Once the results pass human oversight and stress-testing, each artifact could be linked to dozens, or even thousands, of predictable outcomes that are likely to occur in the CRM workflow, per the example provided.

To our knowledge, no major REP solution today, nor the dozens of AI-native startups we’ve reviewed, yet provides this true, self-discovering AI taxonomy capability. It’s likely to happen before long and will be a game-changer with ramifications far beyond the initial impact. When, not if, a provider brings this capability to market, the remaining 5% of the graphic — a tiny minority of REP users whose platform self-educates regarding which assets link to positive and negative deal outcomes, truly leveraging machine learning — may eventually become the norm.