Consulting | Shop | Website | Newsletter | Speaking

Hey {{first_name}} 👋, I hope you’re having a great week! In the last issue, we discussed about LTV to CAC, and now we are moving with the next topic from Reporting content.

Let’s talk about ⬇️

A company came to me with a 112% NRR they'd been putting in front of their board as proof of product-market fit.

Fair enough. 112% is a good number.

Then we pulled it apart. Gross revenue retention was 84%.

The entire 112% was coming from the top 15% of accounts — mid-market and up — expanding hard enough to paper over everything being lost underneath. Their SMB cohort was churning at roughly three times the mid-market rate. Nobody had looked at the two numbers side by side, so nobody knew.

My recommendation was uncomfortable: pause the expansion playbook for a quarter. Take the CS capacity you're pointing at upsells and point it at SMB triage instead — find the accounts most likely to churn in the next 90 days and get in front of them.

GRR went from 84% to 91% over two quarters. NRR dipped to 104% in the meantime, which nobody enjoyed. But the base underneath was a completely different business by the end of it.

That's the thing about expansion MRR. Tracked on its own, it can make a leaking bucket look like a growth engine.

Here's how to read it properly.

1️⃣ Where expansion sits in the MRR stack

2️⃣ Why it matters more than your dashboard suggests

3️⃣ Calculating it without lying to yourself

4️⃣ NRR, and the benchmark trap

5️⃣ GRR is the floor. Check it first.

6️⃣ Where you should be, by stage

Quick reset, because the components get muddled constantly.

MRR has four moving parts:

New MRR — customers who signed this month
Expansion MRR — existing customers who upgraded, added seats, or used more
Contraction MRR — existing customers who downgraded or scaled back
Churned MRR — customers who left entirely

Net new MRR = New + Expansion − Contraction − Churned.

When expansion beats contraction and churn combined, your existing base is growing revenue without a single new signature. That's net negative churn, and it's about the clearest product-value signal a SaaS model produces.

Expansion itself comes from three places. A customer moves from the $800 plan to the $1,400 plan — that $600 delta is expansion. A customer adds a second module. A customer adds seats, API calls, data volume, whatever your pricing scales on.

What it isn't: a renewal at the same price, one-time services, or setup fees. The test is one question. Did this existing customer's recurring monthly payment go up from last month?

Run the math on a $10M ARR business with 30% annual churn — not unusual for SMB-heavy products.

You need roughly $3M in new ARR just to stand still.

If expansion covers $1M of that, your new-logo team's burden drops by a third. Same growth number, third less pressure on acquisition.

Then there's the cost side, which is where it gets interesting. A dollar of expansion ARR costs about $0.69 in sales and marketing. A dollar of new-logo ARR costs $1.50 to $3.00.

Two to three times more capital efficient. That matters enormously when you're weighing runway against a hiring plan.

But expansion isn't free, and it isn't automatic. It has its own acquisition cost and it requires a pricing structure that actually allows customers to pay you more without renegotiating their whole contract. More on that in a minute, because it's the single most common blocker I run into.

Expansion MRR = the sum of all MRR increases from existing customers in the month.

Pull every customer active at the start of the month with their MRR. Pull the same customers at month end. Where current MRR beats prior MRR, that difference is an expansion event. Sum the positives. Exclude anyone who signed this month — their full MRR is new MRR.

Rate version: expansion MRR divided by total MRR at start of month. $500K starting, $25K expansion, that's 5% for the month.

Four ways this goes wrong:

New customers leaking into expansion. If they signed this month, all of it is new MRR. They count as expansion starting next month.

Annual upgrades booked as lump sums. Mid-year upgrade on an annual contract — recognize the MRR delta from the upgrade date forward, not all at once.

Reporting expansion without contraction. This is the big one. Gross expansion in isolation is an optimistic number that tells you nothing about whether you're actually net-additive. Report net expansion — expansion minus contraction — as the headline.

Reactivations counted inconsistently. Pick a policy on whether returning customers are expansion or new. Write it down. Apply it every month.

NRR = (Beginning MRR + Expansion − Contraction − Churn) ÷ Beginning MRR × 100

Expansion is the only variable in there that pushes you above 100%. Everything else pulls down.

Here's where it sits by segment:

Twenty-one points between enterprise and SMB. That spread is the whole story, and it's why a single "SaaS NRR benchmark" is close to useless.

For a second reference point, SaaS Capital's 2026 survey of 1,000+ private B2B SaaS companies puts bootstrapped scale-ups at $3M–$20M ARR at 103% median NRR, with the 90th percentile at 117.9%. Their one published ACV cut — the $25K–$50K band — shows a 102% median, 111% top quartile, 97% bottom quartile.

Notice those two sources don't perfectly agree. Neither do Benchmarkit (101% median) and ChartMogul (~106% for venture-backed). Different samples, different definitions, different years. That's normal — and it's the first thing worth understanding about benchmarks.

Now the trap.

Everyone quotes 120%. Bessemer's scale — 100% good, 110% better, 120% best — is the most-cited framework in SaaS, and it's built on the companies Bessemer sees. Public company NRR gets thrown around the same way.

Both are survivorship bias. Only the best SaaS companies reach public markets, and only a specific slice ends up in a top-tier VC's portfolio. Private B2B SaaS medians sit in the 101–108% range depending on whose survey you read and how they segment.

There's a second, sharper version of this trap that almost nobody talks about.

I went looking for the source of a widely-quoted NRR table recently. Found it repeated across a dozen sites, all attributed to the same research firm. Went to that firm's actual published report — the numbers weren't in the text. They were readings off a chart image, passed hand to hand until the attribution hardened into fact.

The numbers might be perfectly accurate. But most founders are managing to a benchmark they've never traced, applied to a peer group they've never defined.

So before you accept any retention target — including the ones in this email — ask three questions. Where did this number come from? Who was actually in the sample? Do those companies look anything like mine?

If your ACV is $3,000 and you sell to small businesses, targeting 115% NRR isn't ambitious. It's structurally impossible. SMB median is 97%, and 97% is fine if your GRR holds above 85% and new-logo volume does its job.

I've watched teams chase a borrowed number into aggressive initial discounting — deliberately underpricing so they'd have artificial upsell room later. That's a real pricing problem, created on purpose, to hit a target that was never theirs.

Benchmark inside your own segment. Cite where the number came from. Or don't bother.

Before you spend a dollar on expansion motion, look at gross revenue retention — what you kept before adding any expansion at all.

Private B2B SaaS medians run 87% to 93% depending on ACV band. Top quartile clears 95%.

115% NRR with 78% GRR means you're losing a fifth of your revenue base every year and expansion is covering the noise. The moment expansion slows — market pressure, pricing fatigue, accounts hitting their ceiling — the churn underneath surfaces immediately and there's nothing left holding it up.

Below 85% GRR, fix retention before you build expansion programs. Every time. That's the whole lesson from the story at the top of this email.

Expansion as a share of total new ARR scales hard with company size. This is the table most teams don't have:

By 2024, companies in the $15–30M+ range were pulling roughly 40% of growth from expansion — up from 30% at the 2021 peak. Harder new-logo environment, plus installed base investment compounding.

So if you're at $12M ARR with a 15% expansion mix, you're behind the cohort. And it's one of exactly three things:

Your pricing doesn't support expansion. Your CS function isn't running an expansion motion. Or your customers aren't growing enough to need more of your product.

Three completely different problems, three different fixes. Tracking expansion MRR is what tells you which one you have.

Expansion MRR is the clearest read you have on whether existing customers think your product is worth paying more for.

At $5–15M ARR it should be 25–30% of new ARR. At $15–50M, 36–40%. Materially below that and it's your pricing, your customers' growth, or your CS motion. One of the three.

Track it monthly. Break it down by segment and cohort. And always pair it with GRR, so expansion can't hide a churn problem the way it did for the company at the top of this email.

The teams I work with who have the cleanest expansion picture also have the cleanest board narratives — because they know whether growth is coming from product value or from acquisition dependency.

Investors can tell the difference. Worth being able to tell it yourself first.

Reply with "EXPANSION" and I'll send you the expansion MRR tracking template — the monthly review checklist, net expansion calculation, segment and cohort breakdowns, and the trend flags that catch a GRR problem before it surfaces.

Chat soon,

I'm opening a small group for SaaS finance operators. Every month: one finished financial model and one live session building it — ARR bridges, burn multiple, usage-based pricing, board packs.

Not a course. Just the models I build for clients, plus a room where you can ask why line 47 is doing that.

$490 for the first year, capped at 30 people. $990 after. Closes Aug 30th.

{{first_name}} You can get free stuff for referring friends & family to my newsletter 👇

50 referrals - Cash Flow Models Bundle 💰

10 referrals - SaaS Financial Model 📊

{{rp_personalized_text}}

Copy & Paste this link: {{rp_refer_url}}