A good doctor doesn't write a prescription off one number. A patient walks in, their heart rate is up, and the lazy read is "they're stressed, here's something for that." The careful read is to check the rest of the chart first: temperature, history, what changed this week, what the patient actually came in to say. One reading is a prompt to look closer. It is not a diagnosis.
Most account managers expand off a single reading. They see one vaguely encouraging number and act on it.
The vaguely encouraging thing is usually this: usage is up. The account "seems healthy." So the AM decides it is time to go and ask for more money. And then the deal stalls in some buying committee they never met, and nobody can quite explain why.
Here is the problem. Healthy and expandable are different words. An account can be perfectly healthy and completely unready to grow. Reading one as the other is one of the most common, and most expensive, mistakes in the job.
Three things have to be true at once
Before you decide an account is ripe, three conditions need to hold simultaneously.
Value. The customer is genuinely getting what they signed up for. This is the precondition. If they are not getting value, you do not have an expansion conversation, you have a retention problem wearing a disguise.
Demand. There is a specific next job they need done. Not a vague "they might like more." An actual problem with an actual owner.
Capacity. There is a person or a budget able to act. Wanting more and being able to buy more are not the same thing.
Miss any one of the three and the expansion stalls. The signals that tell you whether all three are true sit in four categories. Read all four before you move.
Product usage signals: are they actually using it?
Start with the product, because this is where the lie of "they seem healthy" usually originates.
Depth. Are the existing users using it more, not less, over the last 90 days? Are they completing core workflows end to end, or abandoning them halfway? If the curve is flat or down over 90 days, do not expand. Fix adoption first.
Breadth. Are more people inside the existing team active? Have new logins appeared that were not there at kick-off? Has an admin added seats without you asking? Breadth inside the existing team is the strongest short-term seat-expansion signal there is.
The hit-a-limit signal. This is the gold standard. The account is bumping against a plan limit, a seat cap, an API ceiling, a storage wall, a project cap. Kyle Poyar's data at OpenView shows hit-a-limit signals convert to expansion at multiples of any other signal. Every tiered product should fire an alert the moment an account crosses 80% of a limit. That alert is the highest-converting moment in the entire job. Treat it like one.
One warning. A usage spike followed by a drop is not a signal, it is a project that ended. Read the shape of the curve, not the height of the peak.
Commercial signals: is there money and motion?
Funding rounds. A new round is the most consistent high-converting trigger in B2B. The money is in the bank and there is real pressure to deploy it. But the window is weeks 2 to 12 after the announcement, not months. Move fast.
Hiring. Net headcount up 10% or more inside the team you sell to is a seat-expansion signal. Hiring in an adjacent function is a new-team signal.
Leadership change. A new VP or C-level in the function you serve triggers a vendor review within 90 to 180 days. This cuts both ways. It can be your opening, or the start of a consolidation review. Either way, get a meeting in the first 60 days.
Re-orgs, M&A, and strategic announcements. A reorganisation moves the whitespace. An acquisition creates brand-new business units that are budget holders on day one. And when a customer announces "we are entering market Y" and your product is load-bearing for that, there is now a named problem with executive sponsorship that did not exist a month ago.
Relationship signals: do you have permission?
Your champion got promoted. This is the single best expansion signal of all. It combines a bigger budget with proven advocacy in the same person. Drop what you are doing.
A new stakeholder is engaging. Someone who was not on the original deal is now attending calls and asking questions. If they are senior or from a new function, that is multi-threading happening to you for free. Convert it before it cools.
Advocacy behaviour. They have given a reference, done a case study, spoken on a webinar. These customers expand two to three times more than the silent ones.
And the anti-signal, the one to be honest with yourself about: if your only friend in the account is the original buyer and they have gone quiet, you do not have permission to expand. You have a retention problem.
External signals: is the world moving them?
Press coverage where the CEO names a problem you solve. Job postings that tell you what the company believes it is missing. Earnings call transcripts where the CFO hands you the buying committee's language for free, "we are doubling investment in customer retention" is a direct trigger for any retention-adjacent vendor. Regulatory shifts that make your product more load-bearing overnight.
You do not have to dig for these. They are public. Most AMs simply never look.
The 2x2 that actually matters
Plot every account on two axes. Value realised, low to high. Expansion-ready signals, low to high.
- High value, high signal. Act this quarter. Your top 20%. Drop everything else.
- High value, low signal. Happy but dormant. Harvest the reference, do not push. There is a right way to hold and harvest these.
- Low value, high signal. A trap. They want more but the basics are broken. Fix delivery first or you will break the account selling them more.
- Low value, low signal. Not an expansion conversation. A retention one.
What this looks like coached well
Here is the honest bit. Every AM nods along to this and then, on Monday, looks at an account, sees usage is up, and reaches for the upsell anyway. Knowing the four signal categories and actually checking all four against a live account, every week, is the gap. It is a behaviour gap, not a knowledge gap.
That is the slice we work on at Replicate Labs. Not a slide deck about expansion signals, but a coach in the workflow that, before you reach for that upsell, asks you the inconvenient question: which signals, in which categories, are actually green here? Or are you about to act on a single reading? It's one application of how AI sales coaching works across the whole job.
Read the whole chart, not one number. The reading-the-whole-chart part is the part everyone skips.