What PE Deals Just Told Founders About the Next Five Years

I read multiple private equity deals this week. Different industries, different stages, different geographies. But the same handful of words kept showing up — and they painted a pretty clear picture of what buyers will reward in 2028, 2029, 2030.

If you're building a company right now, this is the scoreboard you'll be graded on later. Worth knowing the rules before the game ends.

Here's what jumped out, and what every term actually means.

Retention is the new growth.

Every single deal led with retention. Net Revenue Retention of 120%, 123%, 161%. Gross Retention of 96-98%. These aren't outliers anymore — they're the bar.

Net Revenue Retention is the math that asks: take the customers you had a year ago, what are they paying you today? If it's more, you have negative churn — your existing book grows even if you never sign another logo. Gross Retention strips out the expansion and just asks how many dollars stayed. One tells you about your growth engine. The other tells you whether your bucket leaks.

The old playbook said NRR of 110% was great. The 2026 reality is that 120% is table stakes for a premium exit. If your pricing model can't mathematically produce expansion — flat per-company subscriptions, no seats, no usage tiers, no module attach — you've capped your ceiling before you've even started climbing.

Fix this at Series A. Retrofitting expansion later is brutal.

The Rule of 40 came back from the dead.

For years, founders were told to ignore profitability and chase growth. That advice aged badly.

The Rule of 40 says your growth rate plus your EBITDA margin should add up to at least 40. Grow 60% and lose 20%? You pass. Grow 30% and make 10%? You pass. Grow 40% and burn 30%? You don't.

EBITDA, by the way, is the closest proxy to "cash the business actually generates" — earnings before interest, taxes, depreciation, and amortization. It's what PE firms use to value businesses, and what your future buyer will use to value yours.

Of the deals I looked at, three were radically above 40. One hit 113. Another 101. The era of cheap growth funded by cheap capital is over, and the businesses commanding attention are the ones that figured it out early.

If you're growing 40% and burning to do it, you're now below median. Plan a path to profitability earlier than your 2021-vintage mentors told you to.

Proprietary data is the only moat that's getting deeper.

In an AI world where models commoditize every six months, the asset that isn't commoditizing is the proprietary dataset accumulated through customer workflows.

This is what people mean by a data flywheel: more customers produce more data, which makes the product better, which attracts more customers. The moat compounds. A new entrant with infinite money still can't catch up, because they don't have the years of customer activity.

Of the deals that I looked at, the ones that led with data assets — "250,000 contracts reviewed," "35 million data points," "14,000 implantations" — were the ones with the highest valuations and the most aggressive growth metrics. That's not a coincidence.

Ask yourself one question: if a competitor cloned your product tomorrow with no historical data, would it work just as well? If yes, you don't have a moat. Redesign the product so normal customer usage produces a dataset only you own.

Vertical beats horizontal. Every time.

Not one of the deals was a horizontal "productivity tool for everyone." Every single one was purpose-built for a specific buyer — financial institutions, CPAs, corporate travel managers, orthopedic surgeons, regulated enterprise verticals.

Vertical SaaS has less competition, deeper workflow integration, more pricing power, and a clearer expansion path within the industry. It's easier to dominate and harder to displace. The phrase that kept appearing was "purpose-built for" — and it appeared because buyers reward it.

Resist the urge to chase TAM by going horizontal. "We could sell to everyone" usually means "we sell well to no one." Pick a vertical, win it, then expand.

AI is a wrapper around a system of record. Not the other way around.

Every deal mentioned AI. None of them were AI companies in the way founders use that phrase. They were systems of record with AI inside.

A system of record is the authoritative database for a particular type of information — Salesforce for customers, NetSuite for financials. Once a tool becomes the system of record, ripping it out is so painful customers basically never do. It's the holy grail of software valuations.

Pure AI features are getting commoditized in real time. The defensible position is system of record + AI. If a customer stopped using your product tomorrow and lost nothing important, you're a feature. If they'd lose the only authoritative record of something they need to run their business, you're becoming a system of record. Engineer toward the second answer.

Channel-led GTM is quietly winning.

One of the deals, sold almost entirely through partners. 30+ partners, 4,000+ corporate clients, growing over 70% year over year, with margins that didn't make me wince.

Channel sales have lower customer acquisition cost and stickier customer relationships, because the partner has already done the trust-building. CAC, by the way, is what it costs you fully loaded to land one customer. LTV is what that customer is worth over their lifetime. The ratio should be at least 3:1, and channel deals routinely outperform direct sales here.

Identify who already has trust with your buyer. If a partner sells to your customer 100 times a year, riding their relationship is cheaper than building your own. Look hard at this before you hire your fifth account executive.

Multi-year contracts get rewarded out of proportion.

One deal pointed out that more than 50% of its ARR sat on multi-year subscriptions. Buyers noticed.

Multi-year contracts dramatically reduce churn risk and give buyers visibility into future revenue. A book of business that's 60% multi-year is worth materially more than the same ARR sitting on month-to-month plans. Offer a discount for annual prepay and a bigger one for multi-year. The valuation premium at exit will dwarf the discount.

The summary.

If I had to compress the whole thing into one paragraph for a founder to tape to a monitor:

Pick a regulated or workflow-heavy vertical. Become the system of record, not a feature. Use AI to deepen the workflow, not replace it. Engineer your product so customer usage compounds into a proprietary dataset only you own. Sell multi-year contracts from day one. Track Net Revenue Retention obsessively and aim for 120%+. Watch the Rule of 40, not just growth. Find a channel partner before you build a sales team.

The single biggest shift from the old playbook: profitability and retention are now co-equal with growth, not subordinate to it. Every metric in this article is a different way of measuring the same underlying question — how predictable, defensible, and efficient is this business?

That's the question your future buyer is going to ask. The only thing you control is whether the answer is ready when they do.

If you found this useful and want the longer founder glossary version — every term unpacked, with the questions buyers will actually ask — let us know. Happy to share.