I am fascinated with Private Equity. I organized startup events, helped at others, mentored some startups in Tampa Bay for ten years. Investors were part of the startup environment. I have watched PE buy up MSPs, SaaS, ISPs, houses and more. So I follow the news. This popped up on LI: a little bit about the PE Report, then a post about Gong, a PE backed SaaS company.
Here’s another interesting fact from Bain & Company’s recent Global Private Equity Report 2026 (released around February 2026): The private equity industry is sitting on a massive backlog of 32,000 unsold companies valued at $3.8 trillion in unsold assets. This “exit bottleneck” has contributed to the prolonged liquidity issues, with average holding periods for buyout funds now stretching to around seven years (up from 5–6 years between 2010 and 2021).
We see this in MSP vendors and in TSBs.
Private equity’s “dry spell” refers to a prolonged slump in deal exits and investor payouts, with distributions at just 14% of net asset value in 2025—marking four straight years of decline, per Bain’s report. It’s worse in duration than post-2008 due to high interest rates, tariff uncertainties, and $3.8T in unsold assets. Ramifications: Longer asset holds (avg. 7 years), tougher fundraising (down 16%), pressure for 12% annual EBITDA growth, potential job impacts in PE-backed firms, and broader economic drag from stalled company sales. Still, PE offers unique diversification.
Comments on this LI post were also interesting:
When private equity freezes, it’s not sentiment… it’s funding cost. Higher rates compress exit multiples, lenders tighten terms, and leverage math stops working. Dry powder isn’t dry… It’s expensive. Capital only moves when the spread justifies the risk.
Comment 2:
In 2008 it was credit collapse. Today it’s exit paralysis. PE exit value fell ~30–40% YoY, IPO window shut, and $2T+ in dry powder sits idle. This isn’t a liquidity crisis it’s a valuation mismatch between sellers stuck in 2021 and buyers in 2026. [The Gong case below is a perfect example.]
So PE money is tied up longer with smaller returns, but they still have money to invest – and they are in less risky areas with assets like data centers and fiber in the ground.
One case of valuations versus exit explained on LinkedIn: The Gong case:
- Gong was valued at $7.25B in 2021.
- They’re now worth $4.5B in secondary markets
- That’s a 38% drop while revenue tripled!
So why haven’t they gone public? Because they can’t…
Here’s what nobody’s saying out loud: Gong should have IPO’d 2 years ago.
Category leader, Gartner Magic Quadrant, 4,000+ customers, massive ARR – Every metric screams “go public”.
Except one – their valuation. The 2021 problem: $100M ARR, VCs paid $7.25B. That’s a 72x revenue multiple. Even the most generous public market SaaS multiples in 2021 were 20-30x.
They priced for perfection and got it – for about 6 months.
The 2025 problem: they’re at $300M ARR now and secondary markets value them at $4.5B – That’s a 15x multiple. If they IPO at that valuation, it’s a 38% down-round from their last raise. No founder wants that headline.
But here’s the real issue – their moat disappeared. In 2021, conversation intelligence was Gong’s proprietary tech. In 2025, it’s a feature in every CRM:
→ HubSpot has it
→ Salesforce has it
→ Microsoft has it
→ ZoomInfo has it
Market data shows 40% increase in “Gong alternative” evaluations in 2024-2025. When your core product becomes a commodity feature, your 72x multiple becomes a 15x multiple.
The CEO’s tell: March 2025 – “An IPO is very interesting but not the most important thing. We are focusing on building amazing products”
Translation: We can’t go public at current valuations and we’re hoping the market recovers.
“We’re nearly profitable and still have plenty of cash from our 2021 round”
Translation: We don’t need to raise, which is good because we can’t raise at a higher valuation.
“We almost haven’t touched it”
Translation: We’ve been very careful with cash because we knew this was coming.
This is what happens when you build a point solution that becomes table stakes
Conversation intelligence was revolutionary in 2018. By 2025, it’s expected functionality in any sales tool.
Gong tried to expand – forecasting, deal management, revenue orchestration. But they’re competing with Clari (who just merged with Salesloft), Salesforce (who just acquired Momentum), and HubSpot.
All of whom have deeper integrations and broader platforms
The trap they’re in: too big to sell – who’s buying a $4.5B conversation intelligence company when every CRM already has it built in?
Can’t IPO – market won’t reward a down-round from 2021’s peak
Can’t raise – no VC is writing a check at a higher valuation when secondary markets say you’re worth 38% less.
So they wait – And hope that AI hype brings back 2021 multiples.
Great comment about Gong: The real headline isn’t “why no IPO.” It’s “did they become the system of record for revenue decisions, or just the best UI for call snippets.”
There was a SaaS sell off in the stock market this month.
“Software stocks have undergone the largest nonrecessionary 12-month drawdown in over 30 years, shedding more than $1.3 trillion in market value.” – Prof Galloway
Everyone thinks they are going to vice code their own personalized software. Yeah good luck with that for so many reasons. Unless you are using a localized LLM, your data will be in a public LLM. Tech companies do not value data privacy! When they go broke, like 23andme, someone will buy the assets and have all your data!!!
I wonder why anyone would use SF. A CRM at its heart is a UI to provide you insights into your customer data. All of your data can sit in an SQL database and Lovable can provide you with a nice dashboard and portal to it. Why pay $199 per user? Now your data isn’t sitting in a software company’s silo, you have it.
The biggest hurdle to AI projects is the data problem. Most data is not accessible. It sits in Salesforce, Stripe, a PSA, Quickbooks, any number of SaaS apps, Gmail, Office365 — your business doesn’t have your data, many other companies do. So you will pay for them to have your data and then pay them for AI agents so you can have insights into your data. Does any of that make sense?
For $150K upfront in hardware costs and $3k per month in colocation, you can own your data and the LLM.
While we are discussing AI, I have had to sit through a 100 pitches and demos on AI Agents in the last year. Very few are impressive. Most of my customers are meh on the idea, since they don’t want to spend money on emerging tech without concrete outcomes. They will wait till it is baked and ready to eat, which will be a while yet.
From an AI class I am in: ‘85% of the workforce does not have a valuable AI use case. That’s what [Section School] sees in almost every enterprise we work with. Companies deploy ChatGPT or Copilot, hold a few training sessions, and leave the workforce to figure it out for themselves. Almost no one does.’
We don’t train people anymore (we aren’t hiring either). The technology has outpaced most users’ ability to assimilate it. If you want the best results, you have to train for user adoption.