According to ChartMogul's Growth Levers report, an analysis of 6,525 software companies with revenue records going back more than a decade, only 3.5% reach $20M ARR within ten years of first monetizing. The ones that cross that line rebuild the business when the old model stops working: the expansion bet, the sales motion, the pricing, and the marketing system.
About ChartMogul
Company: ChartMogul, a B2B SaaS company founded in Berlin in 2014. Used by more than 6,000 subscription businesses worldwide.
Product: Subscription analytics platform. It connects billing, usage, and marketing data into a single revenue view, tracking MRR, ARR, churn, LTV, and expansion revenue for SaaS teams working with messy billing data. In 2023 it launched a CRM purpose-built for B2B SaaS, turning the analytics tool into an integrated revenue platform.
Stage: scale-up.
Marketing Team: Marketing sits under a single Chief Revenue Officer alongside sales and operations, with no separate head of marketing.
The expertise here belongs to Sara Archer, Chief Revenue Officer at ChartMogul. She joined in 2018. First commercial hire, and she has run the commercial side ever since.
Scaling B2B SaaS: adaptability is what separates the 3.5%
The number comes out of ChartMogul's own dataset, analysed by Kyle Poyar in his role as Analyst-in-Residence. Two reports cover it: Growth Levers: The Path from $1M to $20M ARR for the $20M threshold specifically, and Against the Odds: The 2025 SaaS Growth Report for the wider milestone curve starting at $1M. The clock starts at first monetization and runs ten years, so the 96.5% is the share that hadn't crossed $20M inside that window. Plenty of them still will.
Seven years in the commercial seat gave Sara a read on both groups. The plays are the same. Winners and stallers alike run the standard GTM motions, and from the outside you can't sort them by tactics. The split shows up when the old shape of the business stops working: one group rebuilds it, the other keeps pushing the same approach uphill for another year and dodges the harder question of what should change. What got you to $2M won't get you to $20M. The companies that grow act like they believe that.
One caveat: reinvention needs self-awareness behind it. Change for its own sake is a distraction. You have to know when to change, why, and how, instead of chasing whatever tactic is loud that quarter.
The wrong second product splits a small team in two
At one point ChartMogul built a revenue recognition product. The logic looked sound. They needed one internally, and enough customers were asking that it read like latent demand.
It was the wrong call. The sales team dreaded every revenue recognition demo. ChartMogul's strength was selling subscription analytics to ambitious SaaS CEOs, and revenue recognition meant selling a compliance product to heads of finance. Different buyer, different motion. Nobody there knew the domain either. Sales cycles stretched. Support tickets piled up with accounting questions nobody could answer well, and retention on those customers dropped.
The deeper cost sat underneath all of that. A small engineering team was now split across two unrelated domains, and the quality of both suffered. They decommissioned the product.
Years later they built CRM capabilities inside ChartMogul instead. That one was natural to sell, and customers adopted it quickly. The lesson is narrower than "don't build a second product." The expansion lever has to match the expertise the team already has and the sales motion that already works. When it doesn't, the warning signs show up in the same order every time: sales dreads the demo, support tickets take longer, retention drops.
When to leave founder-led sales, and why you hire two reps
Sell to your first hundred or so customers yourself. Then watch for one signal: the moment you can no longer give every prospect a genuinely good evaluation experience. That's when you hire.
Then hire two salespeople:
"You should hire two sales people if you can afford it. Two allows you to see what constitutes good. And so two sales reps will compete against one another, and if you only have one you'll never know if they're performing at 60%, 80%, 120%."
Two reps also give you redundancy if one doesn't work out. Manage them yourself until they start asking questions about negotiation or objection handling that you can't answer well. That's the signal to hire a first-line sales manager.
On who to hire, pick someone you'd want to buy from, someone whose thinking makes you want to spend time with them. Every hire should add energy to the room.
At low price points, reverse-engineer the buying process
Adding a sales team can slow growth when the price point is low. Sara starts from the buyer instead. How do they want to evaluate and purchase? The process gets built backwards from that answer.
Buyers split between wanting a call and wanting pure self-serve. Most want to move between the two as questions come up, so the job is to identify buyer types and route each into the right path.
Sara's example: a ChartMogul customer getting a thousand trials a month with no sales team. Call every signup within five minutes and offer help, she suggested. No pitch. If everyone hangs up, you know inside a week that sales calls aren't the path there, and the test cost you nothing. If they don't hang up, you've found something.
Where AI earns its place on a commercial team
Sara's team uses AI heavily on the commercial side.
Customer case studies were the clearest win. She interviewed customers, ran the raw transcripts through AI to surface the best quotes and draft the narrative, and cut a week-long process down to two and a half hours.
Call analysis came second. They fed 25 recent sales-call transcripts into AI and asked what objections were shifting over time. That grew into a coaching tool. It reviews call transcripts twice a day now and flags what could have gone better.
Both wins share a shape. The raw input already existed, and all AI did was compress the reading of it.
Where AI fails, and the "AI tourist" sitting in your churn
It broke on one-to-one email responses to product questions. The AI handled the technical part fine. Ask it how to find failed transactions by country inside ChartMogul and you get a correct answer back. What it couldn't do was read why the customer was asking. Someone worried about involuntary churn in a specific region has a business problem sitting behind the technical query. Reading that requires what Sara calls "layers of theory of business": the stakes, and the context that makes a decision personal rather than purely rational. They switched the tool off.
The second AI lesson comes straight out of ChartMogul's data, which shows elevated churn among AI-product users. Bad products aren't the cause:
"With AI specifically we see this concept of the AI tourist... A lot of folks that are trying AI tools or products, they don't intend to receive recurring value or impact because it's experimental in nature... And so that's why you see high churn."
Recurring revenue is a result of recurring value, and experimental signups don't produce that. For founders, separate experimental users from users with real intent to adopt. You can still monetize the tourists. Their churn is structural, and mixing the two poisons your metrics.
Dismantling marketing that looked healthy
The hardest commercial decision Sara made recently was pulling apart a marketing system that looked like it was working. Conferences, content, blogs, partnerships, monthly panels. The brand was active and the calendar was full. The number of trials didn't move.
"It was certainly supporting the brand, but it wasn't meaningfully moving the needle on number of trials. And so I've disassembled that way of marketing... You shouldn't do the marketing things because you think you're supposed to — you should have conviction about the growth experiments that you're running."
She replaced it with faster, more technical marketing, closer to growth experimentation than brand building. Part of that was rewriting customer case studies to stop hiding the hard parts. The polished "connect Stripe, get a beautiful dashboard" story gave way to versions that talked about messy billing data, missing invoice periods, and the cleanup tools ChartMogul actually provides. Marketing had to build deep product knowledge to write them. Revenue growth over the following seven to eight months was positive, and Sara ties part of that to a team putting out work it's genuinely proud of.
Treat pricing as a muscle you exercise
Nobody wants to own pricing. It's high stakes and hard to model, and people rarely buy the way your spreadsheet says they will.
Sara's answer is to make it a habit. Someone in the business gets comfortable with the discomfort and builds real knowledge of pricing and packaging, then revisits it every two to three months. Often the decision is to change nothing. The questions still get asked. Should we price on a different value metric? Should the plans and packaging look different?
A full year between pricing reviews means you're already late, and ChartMogul's own data backs that up: companies with a regular habit of experimenting with plans and packaging grow faster across nearly every segment.
When growth stalls, the answer is usually three calls away
For founders with a product and early sales who can't scale, Sara's advice is to go listen to customer calls. Put them on during the treadmill or the dog walk. If you've abstracted yourself away from the sales conversation, that's usually where the problem is hiding, and no quarterly dashboard review is going to surface it for you. ChartMogul's CEO can open a sales call and name the problem inside three minutes. That's the bar.
The habit underneath all of it
One thread runs through all of it. Every decision Sara reversed had looked fine at the time. The second product had real demand behind it, the marketing calendar was full, the pricing had been set once by someone competent, and the sales motion was the one that got them their first hundred customers. None of that was broken. The habit worth copying is the calendar entry: a recurring date to reopen pricing and packaging every two to three months, where "change nothing" counts as a decision only if someone had to argue for it.
What Holds Up
- Only 3.5% of software companies clear $20M ARR within ten years of monetizing. The ones that get there keep rebuilding.
- ChartMogul built a revenue recognition product because customers were asking and the demand looked real, then found itself selling a compliance tool to heads of finance in a domain nobody on the team knew. An expansion bet has to fit the expertise you already have and the motion that already sells. When it doesn't, sales starts dreading the demo long before anyone reads it in the numbers.
- Hire two salespeople. One rep gives you no baseline for what good looks like.
- At low price points, adding sales headcount can slow growth. Work backwards from how buyers want to buy.
- AI compresses research when the raw input already exists. Hand it a one-to-one customer question and it answers the technical part while missing the business problem underneath.
- High churn on AI products is often structural, because experimental signups never intended to recur, so separate the tourists from real adopters before you read retention.
- A full calendar proves activity. At ChartMogul the number that settled it was trials, and trials weren't moving.
- When growth stalls, go listen to customer calls. ChartMogul's CEO can name a deal's problem three minutes into one.
This article is adapted from a podcast episode of Rebuilding SaaS Marketing by Digital Hunch, where Sara Archer, Chief Revenue Officer at ChartMogul, explains what ChartMogul's data on 6,525 software companies reveals about scaling B2B SaaS, and the commercial decisions she made across seven years to get it right.
Watch on YouTube and listen on Apple Podcasts and Spotify.
