The Expansion Signal

Problem: You watch your customer accounts closely for signs that someone is about to cancel. You have alerts, health scores, a whole early-warning system pointed at the exit. The exact same data shows which customers are ready to buy more right now, and almost nobody is looking at it.

Quick Win: An expansion signal is a sign that an existing customer is ready to buy more from you, read from the same account data most teams only use to predict who is leaving. The strongest ones are public or sitting in your own usage logs: a customer hits a plan limit, their usage keeps climbing, a new executive joins on their side, or the company itself is growing. Expansion revenue is also the cheaper revenue. In the 2025 Benchmarkit SaaS report, winning $1.00 of revenue from an existing customer cost a median of about $0.61, versus $2.00 to win the same dollar from a new customer, roughly a third the cost. Watch the same signals for offense, not just defense.

What an Expansion Signal Actually Is

Most companies run some version of a churn alarm. (Churn just means customers leaving or canceling.) When usage drops, a key contact goes quiet, or a login stops happening, someone gets a warning that an account is at risk. Good. That system exists to protect the revenue you have.

Now flip the question. The same account data that shows a customer pulling away also shows a customer leaning in. A team that keeps adding users, keeps hitting the ceiling of its plan, keeps pulling more people into the tool: that is not a risk, that is a buyer raising a hand without saying a word.

An expansion signal is that raised hand. It is an event or a trend that tells you an existing customer is ready to spend more: buy a bigger plan (an upsell, meaning a more expensive version of what they already have), buy a second product (a cross-sell, meaning something new alongside it), or add more seats and usage. The signal answers one question: which of my current customers is about to outgrow what they are paying for?

This is the offensive read of data most teams only use defensively. Same inputs, opposite question.

Same Data, Opposite Question

Here is why almost nobody acts on this. The team watching account health is wired to look for danger. A rising usage line, a new person added to the account, a second department poking around: to a churn-focused system, none of that even registers, because nothing is going wrong.

But those are the exact fingerprints of a customer ready to grow. Consider two accounts sending the same signal, a sharp jump in usage this month:

• The churn reader sees a customer straining against limits and worries they will get frustrated and leave.
• The expansion reader sees a customer who has outgrown their plan and is ready to pay for more room.

Often both are true at once, which is the whole point. A customer bumping against a ceiling is both a churn risk and an expansion opportunity. Ignore them and the friction turns into resentment. Reach out with the right upgrade and it turns into revenue. The signal is neutral. Your response decides the outcome.

If you already run a churn early-warning system, you have most of the raw material built. You are just reading it for one question when it answers two.

The Five Signs an Account Is Ready to Buy More

Not all expansion signals are equal. Some mean buy-more-now; some are early tremors that a bigger conversation is coming. Here are the five that matter, ranked from hottest to earliest.

The top two are events: something specific happened, and the window is short. The bottom three are trends: they tell you where to point your attention before the hard signal fires. A customer that just raised money and is hiring fast is the account most likely to hit a plan limit next quarter. You do not wait for the limit. You watch the account that is about to.

Why Expansion Beats New Customers (The Math)

The reason this is worth building a system around is not sentiment about loving your customers. It is arithmetic.

Winning revenue from someone who already pays you is dramatically cheaper than winning it from a stranger. In the 2025 Benchmarkit SaaS report, the median company spent $2.00 in sales and marketing to win $1.00 of revenue from a new customer, but only about $0.61 to win the same dollar from an existing one. Expansion was more than three times as efficient. The odds are lopsided too: the widely cited figure from the Marketing Metrics textbook puts the probability of selling to an existing customer at 60 to 70%, versus 5 to 20% for a new prospect (via Zuora).

This is why the metric serious software companies watch above almost all others is net revenue retention, the share of revenue you keep and grow from your existing customers over a year, written as a percentage. Keep everyone and sell nobody more and you sit at 100%. Every dollar of expansion pushes you above it. FE International puts healthy territory above 110%, premium above 120%, and best-in-class above 130%, while below 100% buyers mark down what your company is worth. The one thing that gets you past 100% is expansion. You cannot reach best-in-class on retention alone; the math caps at 100% if nobody ever buys more.

And this is not a rounding error on the growth line. Benchmarkit found expansion made up a median of 40% of all new revenue, rising past 50% at companies above $50M in yearly revenue and roughly two-thirds above $100M. The bigger the company, the more of its growth comes from customers it already has. Snowflake, one of the most-watched examples, reported 125% net revenue retention for its fourth quarter of fiscal 2026, with product revenue up 30% year over year, growth driven mostly by existing customers spending more (Snowflake Q4 FY2026 earnings).

The takeaway for a mid-sized business: the most expensive revenue you chase is the stranger. The cheapest is the customer already sending you signals you are not reading.

Building the Expansion Queue

The output of this is not a dashboard someone checks when they remember. It is a queue: a short, ranked list of accounts ready to buy more, refreshed automatically, handed to whoever owns the relationship. Four steps get you there.

1. Watch the same data twice. The usage, seat, and account signals feeding your churn alarm feed this too. You are not building a new pipe, you are adding a second question to the one you have.
2. Score for readiness, not risk. Rank accounts by how many expansion signals stack up and how hot they are. A customer that hit a limit, is growing headcount, and just added a second team outranks one with a single slow-climbing usage line.
3. Sort by how fast the window closes. A plan limit hit today is a same-day conversation. A funding round is a this-quarter setup. Put the perishable signals at the top, same as you would with signals for reopening a dead deal.
4. Pair each account with the right offer. Not "want to upgrade?" but the specific plan, the specific reason, tied to the specific signal. A customer who hit a seat limit gets a seat expansion, not a generic pitch. The signal tells you what they need. Bring exactly that.

The result is a list to approve, not a report to read. Every account on it has a reason to be there and a move attached.

Who Owns the Expansion Play, Sales or CS

This is where good intentions die in most companies, so be blunt about it.

Detection should never belong to a person. If spotting that an account is ready to buy more depends on a customer success manager happening to notice a usage chart during a busy week, the signal will fire and nobody will act. The watching is a system's job. It does not forget, take vacation, or get buried.

The conversation belongs to whoever owns the relationship. Usually customer success sees the climb first, because they live in the usage data. Usually sales closes the bigger contract, because expansion past a certain size is a real negotiation. The clean answer is a shared queue: the system detects, customer success qualifies and flags, sales closes anything above a set threshold, and small upgrades the customer can do on their own get handled without a human at all.

What you are avoiding is the gap. The gap is where an account sends three expansion signals, sales assumes customer success has it, customer success assumes sales has it, and the customer quietly upgrades to a competitor who actually called. A single source of truth for every account, the same backbone that powers follow-up recovery, closes that gap.

When to Leave an Account Alone

An expansion signal tells you the timing might be right. It does not tell you the answer is yes. Push at the wrong moment and you turn a healthy account into a churn risk. Here is where this breaks, honestly.

The account is unhappy. A rising usage line on a frustrated customer is a churn signal wearing an expansion costume. If the account has open complaints or a falling health score, fix the relationship before you sell into it. Pitching more to someone already regretting what they bought is how you accelerate the exit.

They have not gotten value yet. A customer three weeks into onboarding who hits a limit does not need a bigger plan, they need to succeed with the current one. Sell the upgrade before the first win and you confirm their worst fear, that you care more about the next dollar than their result. Wait for value, then expand.

The growth is real but not yours to serve. A customer that raised money and is hiring hard is a great signal, unless the hiring is all in a department you have nothing to offer. Company growth is not automatically your expansion. Check that the growth touches the part of the business you actually serve before you act on it.

You are guessing at the signal. If your "expansion signal" is really just intent data, a vague guess that a company might be interested, you are back to chasing rumors. Expansion works because the signals are real events in your own data. Keep it that way.

Related Reading

• Your best customers are quietly leaving, the defensive read of the same signals
• Reopen dead deals from dormant pipeline, working accounts that already know you
• Buying signals vs. intent data, why real events beat guesses about interest

Frequently Asked Questions

How is an expansion signal different from a churn signal?

They come from the same data, read for opposite questions. A churn signal asks "who is about to leave," an expansion signal asks "who is about to outgrow their plan." A sharp usage jump can be both at once: a customer straining against a limit will either get frustrated and leave or upgrade and stay, and your response decides which. Most teams only ever read the data for the churn question, which is why the expansion opportunity sits unworked.

Which expansion signal is the strongest?

A hit plan limit, because the customer is actively blocked right now and the friction is fresh. A second team inside the same company adopting your tool is next, because it opens a cross-sell. Usage climbing steadily, a new executive joining, and the company growing are all earlier and softer: they tell you where to look before the hard signal fires, but they are not yet a buy-more-now moment on their own.

Is expansion revenue really cheaper than new revenue?

Yes, by a wide margin. The 2025 Benchmarkit SaaS report found it cost a median of about $0.61 to win a dollar of revenue from an existing customer versus $2.00 from a new one, roughly a third the cost. Expansion also made up a median of 40% of all new revenue, and roughly two-thirds at companies above $100M in yearly revenue. The bigger and more mature the company, the more of its growth comes from customers it already has.

---

You already watch your accounts for the ones about to leave. The same data shows the ones about to buy more, and it is the cheaper revenue by every measure that matters. We install that second read as a working queue: usage, hiring, and account signals turned into a ranked list of customers ready to expand, each paired with the right offer, handed to the person who owns the relationship before the window closes. See what we build for companies →