In New York, I sat down with the CEO of an investment management platform that tracks performance for hedge funds and PE firms across the spectrum. The conversation turned to portfolio returns. What he said has stayed with me: based on their research, most deals invested over the past three to four years will never achieve the multiples the original thesis assumed.
His explanation wasn’t about market headwinds or product gaps. He called it the end of financial acrobatics, the era when financial engineering alone drove PE multiples. That playbook isn’t working anymore. And when I look at what’s actually happening inside portfolio companies, I know why.
Deal partners are asking the same question on every call: What’s your AI moat? It’s a fair question. AI is reshaping competitive dynamics faster than most organizations can adapt. But it’s the wrong question to lead with. The companies that are missing their multiples aren’t losing because a competitor out-innovated them. They’re losing because their go-to-market engine can’t execute at the scale the thesis assumed.
The product isn’t the problem. The distribution model is.
The Numbers Confirm What I’ve Been Seeing on the Ground
Fullcast’s 2026 GTM Benchmarks analyzed 361,000 opportunities across 316 companies. The picture is grim:
- Sales efficiency dropped 28% year over year
- Win rates fell 13.5%
- Deal values declined 11%
- Sales cycles stretched 7%
Those aren’t isolated data points. They compound into a system that can’t generate the returns the thesis assumed. And the mechanism behind the decline is predictable once you’ve seen it enough times.
When KPIs slide, reps feel pressure to skip qualification steps. Pipeline looks full on the dashboard, but the conversion math tells a different story. Management sees healthy-looking numbers and draws the wrong conclusion. The instinct is: get more leads, fill the funnel. But you can’t outrun a qualification problem with volume. The system gets worse, not better.
I’ve watched this pattern play out in diligence work across companies from $30M to $250M in revenue. The presentation is consistent. A handful of reps carrying the number, acts of heroism that look like sales success until the top performer leaves, or burns out. Founder-led businesses that earned early traction but never built a system that scales. Comp and incentive plans designed for the wrong operating model, what I call role corruption, where the same people responsible for cross-sell and upsell are also hunting new logos, and neither job gets done well. Processes that exist in people’s heads but were never documented. CRM systems that were never built around how customers actually buy, which means the data leadership relies on is fundamentally uninformative.
Put all of that together, and you have a company where the strategy changed, but the execution didn’t.
The Diagnostic Three
Three questions separate predictable GTM performers from organized chaos:
- Targeting precision: Do win rates vary by less than 20% across segments?
- Capacity design: Is pipeline per rep balanced within 2x across the team?
- Stage discipline: Do fewer than 15% of deals skip qualification?
If you answer no to any of these, you have a system problem, not a talent problem. Swapping the CRO won’t fix a design issue. ICP misalignment alone costs companies up to 75% in win rates. Overloaded pipelines reduce conversion by 57% because reps are spread too thin to build the relationships modern buying committees require.
What the Fix Actually Looks Like
The top performers in the Fullcast study didn’t outperform because they found a better market or hired better talent. They rebuilt the operating model. Tight ICP definition so sales, marketing, and account management were chasing the same accounts. Buyer personas grounded in actual decision behavior, not firmographics. A stage-gated sales process with conversion math at every step. Pricing governance so discounting didn’t erode margin. Deal routing based on rep expertise, not territory or round-robin assignment.
Here’s what that looks like in practice.
A portflio company, now exiting, we inherited a 150- to 200-person sales team fielding 10,000 incoming invoices a day, with no model for which ones to pursue. We built AI routing to surface the highest-win-probability opportunities and matched them to the right reps. Then we rebuilt the sales process from the ground up: people, process, and tools, in that order. The multiple went from 5X to 9X. What the acquirer paid for wasn’t a pipeline. It was conversion predictability.
I’ve done this kind of exit work multiple times. The outcome always comes back to the same three variables. AI accelerates all of it, but only when the foundation is solid. AI-enabled teams ramp 32% faster because onboarding is structured and repeatable. Large deals deliver 6.4x the efficiency of small ones, but only when the pipeline is designed to support them. Forecast accuracy improves from 48% to 94% only when stage progression is based on buyer evidence rather than rep optimism.
When the fundamentals are broken, AI doesn’t create an advantage. It scales the inefficiency.
Pull the CRM. Find the Problem.
For the PortCos you’re most concerned about, run these three pulls right now.
- Win rates by segment: if they vary by more than 30%, you have an ICP problem.
- Pipeline per rep: if the top quartile carries 3x the volume of the bottom, you have a routing problem.
- Stage duration in proposal: if deals sit longer than 21 days, you have a qualification problem.
These aren’t sales issues. They’re system design issues. And they show up in the forecast before they show up in the P&L.
What Good Looks Like
| Metric | Target | Warning Sign |
|---|---|---|
| Win rate variance by segment | <20% | >30% |
| Pipeline spread (top to bottom) | 2x | >3x |
| Stage duration in proposal | <14 days | >21 days |
| Forecast accuracy at week 2 | >85% | <50% |
The Clock Is Running
Here’s the consequence that doesn’t get said in board meetings: the question isn’t whether you’ll eventually fix the GTM operating model. You will, or the banks will do it for you after they take ownership and your LPs absorb the loss. I’m working with one of those companies right now. We’re turning it around. But that’s not the success story private equity is looking for.
The back end of a hold period is the worst time to start. When you’re two or three years out from a target exit, and the foundational pieces aren’t in place, the window to turn this into a winning story versus a losing story starts to close. The foundational work doesn’t get easier. It gets more expensive.
Do it now. Don’t regret it later.
The Question Every Operating Partner Should Be Asking
If we had to exit this PortCo in 12 months, would our GTM metrics support the multiple we need, or would they expose the multiple we deserve?
The difference between a 13% win rate and a 25% win rate isn’t just revenue. It’s multiple expansion. Fix the system, and the valuation follows.
