I’m just gonna be straight with you… there’s a TON of bad business advice flying around right now…
“Companies are only hiring AI engineers”…
“You need VC money to scale”…
“This economy is bad for starting a business”…
If you’ve fallen for any of these oft-repeated assumptions, you need to read Mercury’s data report, The New Economics of Starting Up. Mercury surveyed 1,500 leaders of early-stage companies across topics ranging from funding, AI adoption, hiring, and more, to set the record straight.
What they discovered is equal parts surprising and encouraging:
79% of companies surveyed, who have adopted AI, said they’re hiring more because of it.
Self-funding is the number one avenue for accessing capital — even for tech companies, with half likely to bootstrap.
87% of founders are more optimistic about their financial future than they were last year, despite prevailing uncertainties.
To uncover everything they learned in the report, click the link below.
*Mercury is a financial technology company, not a bank. Banking services provided through Choice Financial Group, Column N.A., and Evolve Bank & Trust; Members FDIC.
Can bad gross margins ever be a good sign?
We all have our sacred cows of the SaaS metrics universe: Rule of 40, revenue per head, burn multiple, CAC payback period. When you’re in this space long enough, they become less math and more a lens through which we see the (business) world through.
But today, with AI inference costs blowing up cost of goods sold, how useful is gross margin as a benchmark?
I means, GM isn’t even a SaaS metric. it’s a straight up rite of passage on the P&L. All comers must pass thru. Doesn’t matter if you’re a donut shop or a cybersecurity point solution. You won’t get very far if your gross margins are trash.
I asked Sarah Wang, partner at a16z, this exact question. She’s invested in some of the top AI companies like Cursor, SSI, Thinking Machines and WorldLabs. Her answer?
"I don't think short-term gross margins matter."
Sarah! From the top rope!
There's been a lot of FUD thrown at AI companies for their margins (including (ok, especially) by me). But Sarah thinks those arguments are short-sighted.
"The best AI companies right now have lower gross margins to start. And I actually think it's an orange flag if your gross margins are 85%, 90%, sky-high."
Wait, what?
"What that tells me is there's probably not much AI usage in your product. Or it's a very light AI overlay that barely gets used."
She explained that when an AI deal goes to investment committee, partners are skeptical AF if the company’s gross margins look like Asana or Adobe.
There can’t be much ‘there’ there.
Now, getting users to actually use your AI product? And accelerate their usage at scale? That’s the hard part.
But if low margins worry you (like they worry me), here's why Sarah says don't panic…
Remember when everyone picked on subscription companies with heavy usage? Especially those that either mailed you salted meats in a box, or allowed you to attend a $38 Barry’s Bootcamp for a buck?
"They're the next ClassPass!"
Outdated argument.
Pricing aside, the DTC (direct to consumer) subscription companies, and even marketplace models in prior cycles, really suffered from low retention and paid customer acquisition problems.
“Those cohorts look dramatically different from today’s AI applications, which deliver deep customer value, strong usage and retention, and explosive net expansion via spreading to teams and the enterprise.”
While great AI companies are melting NVIDIA chips to support free and paid users, their problem isn’t one of acquiring or retaining customers. The money is spent on a different part of the P&L.
Moreover, the good companies have pricing control. While I’ve argued in past pieces that there are a lot of AI companies who are loose pass throughs for OpenAI and Anthropic, generating a topline figure that’s more akin to Gross Merchandise Value (GMV) than Revenue, Sarah says the best companies can protect their margins through the pricing model itself. Assuming people are capped at $20/month for AI products undervalues what customers are actually saying:
“We're using it and we're retaining."
Plus, depending on your benchmark, inference costs have dropped 10-100X.
My obvious counter:
"But everyone only wants the frontier model. The cheaper stuff is last-gen pricing."
No one wants the iPhone 6 when there’s a 16 available (or whatever the hell version we’ll be on this holiday season).
Sarah's response: No AI product runs on just one model. The best apps have +5-10 models under the hood. For tasks where performance isn't differentiated, they're switching to lower-priced models immediately. That's actually part of their value prop; helping enterprise customers optimize model usage.
Notion CFO Rama Katkar confirmed this on the RTN pod. Their AI costs fell nearly 3X within two years. They actively toggle between commoditized queries and specialized tasks.
Furthermore, if you can become a platform, you’ll meaningful improve customer economics at the overall customer level.
Once users are hooked on your product, you can build the platform. Come for the tool, stay for the platform. You add higher-margin products over time. Tale as old as time.
"Saying 'higher margin equals more sustainable' just misses the mark in AI today."
But we do have an undeniable benchmarking problem.
This has me all in my feelings as a total benchmarking snob. Traditional SaaS companies had big arms (high gross margins) and skinny legs (heavy OpEx).
Don’t skip leg day
AI companies are inverted: their gross margins aren't great, but OPEX is super low, with wicked lean teams of engineers pumping out companies that do $100M in revenue in less than 12 months.
So how do you benchmark them?
Well, you kinda gotta put ‘em in a blender. Because AI-native companies are messing with the scoreboard.
"We still look at metrics like burn ratios, putting it all in a blender. What's the net new ARR relative to burn?
ARR and NRR still matter, but they’re the stat line after the game. They don’t show you how fast the team’s actually moving, iterating, and learning in real time.
We need a new way to measure momentum.
Momentum is the new currency of enterprise value. It’s the speed at which a company learns, ships, and repositions capital based on what it’s seeing in the wild. That’s what accelerates compounding.
The hard part? Everyone knows what momentum feels like. Very few know how to track it.
What would it even look like? Making it up on the fly here…
Feedback Velocity – How fast are insights from the field making it back into product?
Use Case Penetration – How often are we getting pulled into new customer problems?
Iteration Speed – How quickly are we changing direction when something isn’t working?
Capital Allocation Agility – How easily can we re-route budget and people to what’s hitting?
This is squishy stuff, especially if you’re the kind of finance person who used to think of brand as ‘the Department of Shapes and Colors.’
Now we’re trying to wrap our heads, and our abacuses, around momentum.
So, Do We Buy It?
Sarah's making a long term argument: gross margin in AI companies is a dial you can move, not a fixed reality.
And most importantly, if you're delivering value, you have control.
But CFOs aren't wrong to care about margins. The question is: are we in a temporary investment phase where low margins are the price of building real AI usage? Or are we rationalizing unsustainable unit economics?
The early data from companies like Notion suggests margins can improve dramatically as the tech matures. But not every AI company will make that journey.
Some will keep burning cash to deliver a watermelon on a bicycle - at a loss - and call it innovation. The market will decide how long they get to keep pedaling.
THE GROSS MARGIN EPISODE IS HERE!!!!
Gross margins, GPUs, and the future of finance… this one’s for all my metrics nerds.
I sat down with Sarah Wang, General Partner at Andreessen Horowitz, to talk about what happens when the traditional SaaS playbook collides with AI. Sarah shares:
How legacy benchmarks like payback period and burn multiple start to break down in a world where compute, not headcount, drives costs.
Why sky-high gross margins can actually be an orange flag
How finance leaders can think about resource allocation between engineers and GPUs
Why the most valuable finance teams today are deeply operational.
We also unpack what it’s like partnering with AI-native founders, the evolution of pricing models as LLM costs drop, and whether we’ll see a private trillion-dollar company anytime soon.
TL;DR: Multiples are DOWN week over week.
Top 10 Medians:
EV / NTM Revenue = 17.5x (DOWN 2.5x w/w)
CAC Payback = 28 months
Rule of 40 = 45%
Revenue per Employee = $413k
Figures for each index are measured at the Median
Median and Top 10 Median are measured across the entire data set, where n = 147
Recent changes
Added: Navan, Bullish, Figure, Gemini, Stubhub, Klarna
Removed: Olo, Couchbase
Population Sizes:
Security & Identity = 17
Data Infrastructure & Dev Tools = 13
Cloud Platforms & Infra = 15
Horizontal SaaS & Back office = 20
GTM (MarTech & SalesTech) = 19
Marketplaces & Consumer Platforms = 18
FinTech & Payments = 28
Vertical SaaS = 17
Revenue Multiples
Revenue multiples are a shortcut to compare valuations across the technology landscape, where companies may not yet be profitable. The most standard timeframe for revenue multiple comparison is on a “Next Twelve Months” (NTM Revenue) basis.
NTM is a generous cut, as it gives a company “credit” for a full “rolling” future year. It also puts all companies on equal footing, regardless of their fiscal year end and quarterly seasonality.
However, not all technology sectors or monetization strategies receive the same “credit” on their forward revenue, which operators should be aware of when they create comp sets for their own companies. That is why I break them out as separate “indexes”.
Reasons may include:
Recurring mix of revenue
Stickiness of revenue
Average contract size
Cost of revenue delivery
Criticality of solution
Total Addressable Market potential
From a macro perspective, multiples trend higher in low interest environments, and vice versa.
Multiples shown are calculated by taking the Enterprise Value / NTM revenue.
Enterprise Value is calculated as: Market Capitalization + Total Debt - Cash
Market Cap fluctuates with share price day to day, while Total Debt and Cash are taken from the most recent quarterly financial statements available. That’s why we share this report each week - to keep up with changes in the stock market, and to update for quarterly earnings reports when they drop.
Historically, a 10x NTM Revenue multiple has been viewed as a “premium” valuation reserved for the best of the best companies.
Efficiency
Companies that can do more with less tend to earn higher valuations.
Three of the most common and consistently publicly available metrics to measure efficiency include:
CAC Payback Period: How many months does it take to recoup the cost of acquiring a customer?
CAC Payback Period is measured as Sales and Marketing costs divided by Revenue Additions, and adjusted by Gross Margin.
Here’s how I do it:
Sales and Marketing costs are measured on a TTM basis, but lagged by one quarter (so you skip a quarter, then sum the trailing four quarters of costs). This timeframe smooths for seasonality and recognizes the lead time required to generate pipeline.
Revenue is measured as the year-on-year change in the most recent quarter’s sales (so for Q2 of 2024 you’d subtract out Q2 of 2023’s revenue to get the increase), and then multiplied by four to arrive at an annualized revenue increase (e.g., ARR Additions).
Gross margin is taken as a % from the most recent quarter (e.g., 82%) to represent the current cost to serve a customer
Revenue per Employee: On a per head basis, how much in sales does the company generate each year? The rule of thumb is public companies should be doing north of $450k per employee at scale. This is simple division. And I believe it cuts through all the noise - there’s nowhere to hide.
Revenue per Employee is calculated as: (TTM Revenue / Total Current Employees)
Rule of 40: How does a company balance topline growth with bottom line efficiency? It’s the sum of the company’s revenue growth rate and EBITDA Margin. Netting the two should get you above 40 to pass the test.
Rule of 40 is calculated as: TTM Revenue Growth % + TTM Adjusted EBITDA Margin %
A few other notes on efficiency metrics:
Net Dollar Retention is another great measure of efficiency, but many companies have stopped quoting it as an exact number, choosing instead to disclose if it’s above or below a threshold once a year. It’s also uncommon for some types of companies, like marketplaces, to report it at all.
Most public companies don’t report net new ARR, and not all revenue is “recurring”, so I’m doing my best to approximate using changes in reported GAAP revenue. I admit this is a “stricter” view, as it is measuring change in net revenue.
OPEX
Decreasing your OPEX relative to revenue demonstrates Operating Leverage, and leaves more dollars to drop to the bottom line, as companies strive to achieve +25% profitability at scale.
The most common buckets companies put their operating costs into are:
Cost of Goods Sold: Customer Support employees, infrastructure to host your business in the cloud, API tolls, and banking fees if you are a FinTech.
Sales & Marketing: Sales and Marketing employees, advertising spend, demand gen spend, events, conferences, tools.
Research & Development: Product and Engineering employees, development expenses, tools.
General & Administrative: Finance, HR, and IT employees… and everything else. Or as I like to call myself “Strategic Backoffice Overhead.”
All of these are taken on a Gaap basis and therefore INCLUDE stock based comp, a non cash expense.
Please check out our data partner, Koyfin. It’s dope.
Wishing you a gross margin that makes sense for your business model,
CJ
