GTM公司应对估值折价的3个动作：按量计费、拆分AI收入

Salesforce agreed the other day to buy Braze for $5.66 billion, 6.8x trailing revenue. In most of B2B that price would be unremarkable. In go-to-market software it’s a 58% premium, because Braze traded at 3.7x on Monday.

I pulled the multiples for the public GTM and front-office apps and compared them with infrastructure and security. GTM software trades at a fraction of what infra and security trade at, and in several cases the growth rates are the same.

Multiples below are enterprise value over trailing twelve months revenue, at mid-October closing prices. Growth is year-over-year revenue growth in each company’s most recent reported quarter.

Braze, Klaviyo, and CrowdStrike All Grew 26%. CrowdStrike Trades at 42.6x, Braze at 3.7x, Klaviyo at 3.1x.

The GTM and front-office group has a median of about 2.7x. Cloudflare, Datadog, and Snowflake have a median of 22.6x, roughly 8x higher.

Part of the gap is growth. Those three infra companies are growing in the mid-30s, about 10 points faster than Braze and Klaviyo. CrowdStrike, though, grew exactly 26%, the same as Braze and Klaviyo, and trades at more than 11x Klaviyo’s multiple.

With growth that close, the difference comes down to what investors expect AI to do to each company’s revenue over the next five years.

ZoomInfo Trades Under the Ticker “GTM,” at a $1.14 Billion Market Cap

ZoomInfo’s ticker is GTM. Its market cap is $1.14 billion. Its enterprise value is about $2.50 billion, more than double the market cap, because of its debt.

ZoomInfo has $1.26 billion in trailing revenue and trades at 2.0x. The equity is worth less than one year of revenue.

Every AI Agent Call Runs on Infra. Fewer of Them Need a Seat in a GTM App.

Every AI agent makes API calls, stores data, and generates logs. Cloudflare, Snowflake, and Datadog charge for all three, and they bill on consumption, so more agents means more revenue for them. Security grows the same way: each new agent is another identity and another endpoint to protect.

GTM apps mostly charge per seat or per contact, and agents cut into both. If an AI SDR does the work of five human SDRs, the sales engagement tool loses four seats. If an agent writes and personalizes the campaigns, the marketing team needs fewer people logged into the platform.

We’ve lived this at SaaStr. We run the company with 3 humans and more than 20 AI agents, on track for about $20 million in revenue this year. A few years ago we had far more people, and every one of them had a seat in a GTM tool. Our agents do much of that work now, and most of them don’t need a seat. We’re also evaluating whether to replace Marketing Cloud entirely.

That risk hasn’t shown up in reported revenue yet. HubSpot grew 20% last quarter and trades at 3.1x. Investors are discounting revenue they expect it to lose later.

Twilio Trades at 6.8x Because It Charges on Usage

Twilio is the one company near GTM trading above Salesforce. It sells messaging and voice, the same channels Braze and Klaviyo run campaigns through, but it charges per message and per minute.

An AI agent that sends a million messages generates a million billable events at Twilio. At Braze or Klaviyo, the same agent might replace a seat or two.

Clay Is Valued at 35x Revenue. HubSpot Trades at 3.1x.

Private investors are paying far more for GTM companies than public investors are.

Clay raised at a $7.1 billion valuation in September. CEO Kareem Amin told Reuters Clay’s annualized revenue would reach about $200 million this quarter. That’s roughly 35x revenue, more than 10x HubSpot’s multiple.

Sierra raised at $15.8 billion in May, on about $150 million of ARR as reported by TechCrunch and CNBC, which is over 100x. Rox raised at $1.2 billion on roughly $8 million of ARR, about 150x. Profound raised at $1.8 billion in September, and Benchmark led Monaco’s $50 million Series B in May. Neither has disclosed revenue.

None of these companies sells seats to human reps. Clay sells data and workflows that agents and ops teams run. Sierra charges for each customer conversation it resolves. Rox and Monaco are built around agents doing the selling. Profound helps brands show up in AI answers, a budget line that didn’t exist two years ago.

Investors aren’t avoiding GTM as a category. They’re avoiding GTM that’s priced per seat. Public investors expect agents to shrink seat counts, and private investors are funding the companies that sell those agents.

Strategic Buyers Are Paying 2x What Public Investors Will

Salesforce paid 6.8x for Braze, almost double Monday’s market price. It has also announced deals for Contentful, m3ter, Fin, Listen Labs, and Loops this year.

Salesforce is valuing Braze on the 2,770 enterprise brands it can sell Data 360 and Agentforce into. That customer base is worth more to Salesforce than the public market will pay for the same revenue.

There will be more deals like this. When a category trades at 2-3x revenue, an IPO stops being an attractive exit, and strategic acquirers and private equity become the main buyers. A company growing 26% with positive free cash flow at 3x revenue is an easy deal for either one.

What GTM Founders Should Do About a 3x Category

Charge on usage or outcomes. Twilio charges per message and trades at 6.8x. Klaviyo sells into some of the same channels and trades at 3.1x. Sierra charges per resolved conversation and raised at over 100x. If agents reduce your customers’ headcount, revenue tied to headcount goes down with it.

Report AI revenue separately. Roughly a third of Braze’s large customers pay for its AI products, and analysts still called its AI monetization early. The AI revenue was bundled into overall growth, and overall growth didn’t move. A separate SKU lets you report attach rate, deal size uplift, and contribution to growth.

Plan for the multiple to change if your revenue model changes. CrowdStrike and Klaviyo grew at the same rate last quarter. GTM companies that show AI adding to their revenue can close some of that 10x gap. So far, most of them haven’t shown it.