Anyone who has sat in a board meeting in the past year has heard some version of this narrative. And they’re not wrong. Based on ICONIQ’s latest State of Software report, ARR per FTE has been climbing steadily across every revenue band. Companies are getting more efficient. And the market has rewarded this discipline. 

But here’s the problem: Fewer people are inspecting where that efficiency is actually coming from. When you drill into the GTM organization — which drives 45% of OpEx in sales-led GTM motions — the picture gets uncomfortable.

What about CAC payback periods? Stuck at ~16 months for top-quartile companies, and degraded over the past five years.

And net magic number? Stabilized after years of decline, but still well below pre-2020 benchmarks. 

GTM financial metrics haven’t just stagnated — in several segments, they’ve actually atrophied. And, according to a recent report from Kyle Poyar, 53% of GTM leaders are seeing little to no impact from AI.

This isn’t just a reality for private companies. A few weeks back, David Spitz shared similar trends in the public markets:

The truth is, the efficiency gains that everyone’s been celebrating are still nascent in GTM departments. And that disconnect is about to become a serious problem.

Bessemer nailed it: “COGS is the new CAC.” As AI and compute costs eat into gross margins, companies can’t afford to carry both high COGS and high acquisition costs. Something has to give.

GTM inefficiencies could hide behind 80% gross margins in traditional SaaS businesses; there was room to absorb expensive GTM motions. But when companies lose meaningful margin points to AI infrastructure and compute, they simply cannot afford to do more of the same on the OpEx side. As CJ Gustafson has been known to say, you can either have big arms (high gross margins) and skinny legs (high OpEx spend) or chicken arms (low gross margins) and jacked legs (low OpEx spend), but you can’t have both.

In 2026, boards are moving past activity metrics and effort signals. They’re going to start asking very direct questions about how your team’s efforts and investments drive P&L impact. They’ll expect to see the fruits of your labor and your AI experimentation flow through to your GTM financial metrics. 

I’m calling this the PRIME framework:

These aren’t new metrics by any means, but expectations are changing. Advancing PRIME is our non-negotiable filter for GTM investing, and we implore CROs to consider it as their own non-negotiable for prioritizing and determining where to place their AI bets.

CROs have been reliable early adopters of AI tooling. Sales enablement, conversation intelligence, lead scoring, outreach automation — their experimentation has been prolific. 

Yet, the gains to be had from these optimizing tools tend to be incremental: Individual reps get slightly more productive; win rates improve by a few percentage points; you receive better meeting summaries. 

Consider, for example, digital deal rooms: Sure, they may slightly accelerate cycle times or marginally improve win rates, but multiple steps of math are required to translate that back to the P&L. 

While marginal gains are certainly valuable, too few solutions — and teams — are focused on thinking more transformationally. 

If you want to be among the best-in-class commercial leaders, you need to be committed to driving an order of magnitude impact. Take these examples:

Beyond these, though, it’s imperative to understand which PRIME metrics matter most for your specific company at any given point in time. 

For example, Invisible Technologies has enlisted a “solution sprint” model, where FDEs deliver real, production-grade outcomes before long-term contracts are signed. While that may look like a hefty front-loaded investment, the strategy has yielded multi-year, seven-figure opportunities on the other side. As Varsha Udayabhanu (Head of Finance, Invisible) has said, “Enterprise AI GTM is not a traditional funnel: It’s a portfolio of bets with adjustable timelines.” 

The takeaway: In lieu of focusing on ARR in isolation, consider momentum metrics like use case expansion, deployment velocity, and human leverage. 

And so, for those companies that can consistently deliver real outcomes, their gross and net retention metrics compound quickly, which results in drastically different customer acquisition dynamics.

In contrast, for a fast-moving Vertical AI player, sales efficiency might matter less in the early years if the dynamic is a land grab for new logos in a limited TAM (though other PRIME metrics will still very much apply!).

AI has finally made it possible to transform the GTM P&L. Workflow barriers are collapsing. Historical silos between sales, marketing, and customer success are being integrated. Data ingestion is easier. The technology exists to attack inefficiency at the root, not just optimize around the edges.

But technology alone won’t save executives. The question is whether GTM leaders are ready to be held accountable — not for activity, effort, or adoption rates — but for measurable P&L impact. 

As CJ reminded me, it’s critical that GTM leaders are always thinking about resource allocation from a total budget envelope perspective: labor and software/tech spend. 

That said, there’s an important question to consider about whether the existing GTM P&L metrics are actually still the right ones in today’s AI world; as I was writing this article, Primary’s VP Strategic Finance Kurt Chessman and I wound up down a fun rabbit hole about maybe creating an entirely new metric to really get folks thinking about P&L impact: OpEx payback periods! 

Expect more thinking on that soon. But, for now, when it comes time to draft the next board deck, I implore you to tell the PRIME story.

Author’s Note: At Primary, we’re actively investing in founders who share our conviction about transforming the GTM P&L. If you’re building something that fits that bill, please get in touch.

Become a Topline insider by joining our Slack channel.

We want to hear your feedback! Let us know your thoughts.