Loading today’s stories
Invest1 distinct publisher3 min readUpdated
Amjad Masad now expects a majority of his headcount in sales. SaaStr's numbers put go-to-market at 23% to 44% of revenue, and the AI cohort simply pays that bill later.
The Investor · Invest desk
Compiled by The InvestorSomething wrong?How this is made
The arithmetic of deferral is what gets left out of the pitch. Replit went from roughly $2.8M to $150M in annualized revenue in under a year [2], and it ran into the ceiling with four reps in the building [6]. That is about $37.5M of annualized revenue per sales head [1]. No sales organisation in the pre-AI cohort ever had to produce a number like that, and it is not a productivity result. It is the measurement of a queue.
SaaStr puts the new wall at $100M to $250M ARR against $30M to $50M for the 2015-2022 cohort, and calls that roughly a tripling [7][8]. The edges of the two bands say something less tidy: the low ends triple, the high ends quintuple [2]. A founder planning against the low edge who gets hit at the high one has spent an extra year handing enterprise buyers to account managers.
The cost side is already fixed by public comparables. Sales and marketing settles at 23% to 44% of revenue at the public leaders, ahead of R&D at most of them, according to SaaStr [9]. Hold that band against Replit's $150M annualized and the go-to-market line is $34.5M to $66M a year [3]. Replit can carry it: the company raised $400M at a $9B valuation in March 2026 and named international expansion and go-to-market build-out as the use of funds [3]. That raise is roughly 2.7 times current annualized revenue [5], at a valuation near 60 times it [4]. In other words, the deferred sales line is being underwritten by the private market. Companies that do not raise $400M fund the same line out of the revenue they were told they would not have to spend to acquire.
The recurring failure is not locating the wall. It is denying there is one. In February 2015, with Slack at roughly $30M ARR and nobody holding a sales title, Stewart Butterfield described the work as account managers midwifing deals past vendor review and in-house counsel marking up the terms of service, and said he believed Slack could operate without commissions or outbound forever [13]. Slack staffed a serious sales team crossing $50M ARR [12]. Box, which started at 0-1% of revenue through sales as a pure freemium product, now sells roughly 90% of it [14].
Masad's own conversion came from watching the dinners work [4], which is evidence arriving well after the decision had been made for him by companies emailing to ask how they could buy properly, with one person fielding those conversations on top of three other jobs [5]. That is the sequence every time. These companies discover the sales line rather than budget for it, and that is the part of the pattern the AI cohort has not changed [8].
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
Sales and marketing ends up at 23% to 44% of revenue at the public leaders, and is the largest line item in B2B software, larger than R&D at most of them.
Anthropic added 140+ salespeople in 18 months and now posts more sales roles than research roles.
SaaStr says the trigger for building sales is demand arriving faster than a self-serve motion can absorb it, the first team arrives disguised as account management or support, and waiting has a real cost whose bill arrives all at once.
Amjad Masad, founder and CEO of Replit, posted: "I thought I hated sales culture. By the end of this year, more than half my company will be salespeople."
Replit went from roughly $2.8M to $150M in annualized revenue in under a year.
Replit raised $400M at a $9B valuation in March 2026, with the stated plan of international expansion and building out go-to-market.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Single publisher, mostly company-reported figures
One source carries the entire cluster. Its strongest evidence is first-hand: the author moderated the February 2015 SaaStr Annual session and links the recording, and he quotes Masad's public post verbatim. Everything quantitative, though, is secondhand and unverified in the piece: Replit's revenue ramp and round terms, Gamma's $100M ARR, Anthropic's 140+ hires, Box's 90% sales-led revenue, and a 23%-44% S&M band with no named filings. The central generalization about the threshold moving is drawn from a hand-picked company list rather than a cohort dataset.
Concrete but self-reported GTM buildouts at named companies
Adoption of the behavior the story describes is visible in several specific, dated actions rather than intentions: Replit's $400M March 2026 round earmarked partly for go-to-market, a disclosed plan for a majority-sales headcount mix, inbound procurement requests from companies whose employees already use the product, Anthropic's 140+ sales hires with sales postings exceeding research postings, and Box's shift to ~90% sales-led revenue. All of it is relayed by one publisher from company statements, which caps the score.
Framework outruns its sample
Mildly overstated. The concrete anchors are real and specific, and the underlying thesis is conservative rather than promotional: it says sales is deferred, not deleted. Overstatement comes from precision the sample cannot bear. A move from $30M-$50M to $100M-$250M is presented as a measured 'threshold ladder' but rests on two AI-era examples, and 'roughly tripled' understates the top of its own range (the high edge is 5x). The headline number, more than half of headcount in sales, is a founder's forward-looking statement about an unfinished year, not an observed org chart.
GTM-content publisher extending its own thesis around a fundraise narrative
Visible in the material itself rather than inferred. The publisher's argument is self-referential: it restates a piece the same author wrote five years ago, and its central evidence is a session the author ran at his own SaaStr Annual conference, an event whose audience is the sales-and-marketing function the article says everyone must staff. On the subject side, Replit's revenue, valuation and headcount figures are company-sourced and land months after a $400M raise premised on building out go-to-market, so the numbers serve the company's positioning as well as the article's thesis.
Directionally credible, quantitatively unverified
Confidence is moderate-low. The mechanism, bottom-up adoption forcing enterprise procurement and therefore sellers, is supported by a first-hand 2015 account plus a documented 2026 parallel, and is consistent with several named end-states. But there is one publisher, one interested primary voice for the lead example, no primary documentation for any headline number, and the most quotable claims are a forecast and a cohort-level threshold band derived from a small curated sample.
leadership
Every's 30 people, four products and one cloned editor: the self-driving company in practice1 distinct publisher
invest
Canva guided 2026 growth down to 20%. The worse detail is who never got asked.1 distinct publisher
invest
Gamma hit $100M ARR with 50 people and no reps. Its CEO calls the reflex behind it a mistake1 distinct publisher
product
Canva's forecast cut turns model routing into a product line item1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.