Which retention metric best reflects a software company's health: gross retention or net revenue retention?

Chad Peets · Aug 23, 2024

Neither gross revenue retention nor net revenue retention is more important; GRR measures whether the product is good while NDR measures expansion and reflects go-to-market and customer success as well

They measure different things: 90% GRR means a good product, 70% means a product problem, whereas high GRR with low NDR points to go-to-market or CS issues rather than product

50:57 20Sales: How Snowflake Built a Sales Machine | Why You Have to Hire a CRO Pre-Product | Why Most Sales Reps Do Not Perform | Why Hiring Panels are BS in Interviews | Why Remote Sales Reps Do Not Care About Their Development with Chad Peets

Gokul Rajaram · Mar 16, 2026

Gross customer retention and net revenue retention are the two fundamental indicators of business quality — he would take a triple-triple-double-double company with excellent retention over one growing 10x a year with bad retention and sub-100% NRR

Retention is what reveals whether the revenue under the headline growth numbers is real and durable

35:19 20VC: The 8 Moats of Enduring Software Companies: How to Analyse for Durability and Defensibility in a World of AI | Why Dropouts are "AI Maxing" the World & Remote Early-Stage Companies are Dying with Gokul Rajaram

Winston Weinberg · Jan 19, 2026

Investors in AI are making a huge mistake by ignoring GRR and looking only at net new ARR

Companies that sign customers fast without the infrastructure to support their promises will start losing them quickly, producing a reckoning once they pass $100M ARR

Scope: applies especially to vertical AI companies past $100M ARR

33:46 20VC: How Model Performance is Plateauing | Two Key Rules for Effective Deal-Making | Company Building Lessons from Keith Rabois, Brian Halligan and Pat Grady | Why Enterprise AI Adoption is Years Off with Harvey CEO Winston Weinberg

Mitchell Green · Mar 7, 2026

Gross dollar retention is the single most important number in a software company: ~90% is good, 95% great, 98% amazing, and below ~88% is uninvestable

A company with 95% GDR can grow fast without burning cash because it isn't spending sales and marketing to refill a leaking bucket

Scope: gross, not net retention — excludes upsells

41:48 20VC: Why the SaaS Apocalypse is BS | Why China Will Win the AI War | Why 50% of VCs Should Not Exist and are Tourists | Why Stock-Based Comp is the Hidden Sin of the Valley with Mitchell Green, Lead Edge Capital

Arvind Jain · Jul 11, 2026

The absence of US open models is not due to a weak open source movement but because training models requires huge upfront investment, which is incompatible with the unfunded skunkworks model of open source

Open source software historically got built with no funding attached; models can't be built that way, so it needs techniques that don't require enormous investment

44:14 Open model training needs capital volunteer open source cannot supply

Luke Harries · May 23, 2025

People measure retention wrongly for consumer and prosumer products: individual seat retention below 100% is fine, and the number that should exceed 100% is account-level NRR including the organic growth virality creates

Enterprise NRR already works this way — individual users churn inside an account while the account expands via seats; in consumer, a user who builds on the product shares it with friends who sign up, so virality is the equivalent expansion mechanism

56:33 Account level ndr not individual seat retention is the real signal especially with organic virality