The Prospect Who Loved Your Product and Still Didn't Buy

Reactivation covers two groups that get lumped together and shouldn't be.

The first is prospects who trialled or ran a proof of concept and didn't close. The second is customers who were paying and didn't renew.

Both are opportunities to re-close a sale that has already had effort invested in it. But they need different conversations, and typically different owners — lost trials generally sit with sales because the contact is still a prospect, while churned customers are usually worked by CS, unless the account is large enough to warrant sales attention.

This piece is deliberately about that after-the-fact triage, not about keeping trials from failing in the first place. I've written separately about converting trials while they're still live, and about building a formal reactivation programme once you've decided a segment is worth the investment. What follows is the step in between: deciding which lost deals earn that investment at all.

The economics are the same for both, and they're the reason to care:

"Being able to convert a lost customer back to a customer is way way cheaper than it is to try and source a brand new customer in the first place."

Even offering a favourable deal to bring someone back usually costs less than acquiring an equivalent stranger. The relationship exists, the evaluation happened, and the product knowledge is already in their head.

By Jon's own estimate from years of watching B2B sales:

"75% of trials don't convert. That's a pretty large pool of opportunities."

Take that seriously for a moment. If three-quarters of trials fail, the pool of people who have used your product and not bought it is roughly three times the size of your customer base — and it grows every month.

Most companies treat that pool as waste. It isn't waste; it's unqualified pipeline. The problem is that nobody has told the sales team which of those thousands of names is worth a phone call, so — reasonably — they call none of them.

Some of that data already exists, just aimed at the wrong moment. Most teams that track trial usage do it to catch decline signals while the trial is still live — see Cracking the Black Box to Light the Way to SaaS Trial Conversions — and then stop watching the instant the trial expires. The same feature-level usage record that could have triggered a save-the-deal conversation in week two is still sitting there in week twelve, just filed under "lost" instead of "watch".

This is the argument I actually want to make. The way to triage that pool is engagement data, and one pattern matters more than all the others.

Start with the basic split. Did they engage at all?

A large share of failed trials, especially anything consumer-shaped, involve no meaningful engagement whatsoever. Someone downloaded it on a Friday, got busy, and forgot. That's not a rejection of your product — it's an unfinished intention. The play is a fresh start, not a follow-up on a conversation that never occurred.

Then the interesting case:

"Let's say there was super engage but then they didn't convert. Okay, that's really interesting, right? That's data you can use in sales and say, 'Wow, is it a timing issue? Is it a budget issue?'"

Someone used your product extensively and still didn't buy. Product fit is demonstrated — they found value, repeatedly, of their own volition. So the obstacle is almost certainly something else: budget timing, a procurement cycle, a reorganisation, a competing priority, a champion who left.

Every one of those is temporary. Every one of those changes.

That's the whole insight. Heavy usage plus no purchase isn't a rejection — it's an unresolved blocker, and blockers expire. A salesperson with that context isn't cold-calling; they're resuming a conversation with evidence:

"That's a hot prospect. They really were engaged. There's something wrong here. I bet I could fix it. I bet I could fix the budget. I bet I could figure out the timing."

Preventing this kind of churn in the first place is its own subject — see SaaS Churn Rate Solutions: Keep Customers Coming Back and the broader case for retention as a growth strategy. For customers who've already left, the equivalent triage question is about trajectory: not how do you stop it, but what does the shape of their usage before they left tell you about whether it's worth reopening.

A customer who was deeply engaged and churned anyway is genuinely strange, and strange is worth investigating. Something happened that had little to do with your product's value — often budget pressure that nobody surfaced in time. Frequently the honest answer is that you'd have negotiated, and were never asked.

Then there's the pattern I find most actionable: a customer who stayed for a while but never activated the features that deliver your core value.

"They've been a customer for a while, but they're not using our key features, right? We haven't activated them in the maximum value we can deliver."

That's a failure of onboarding rather than of product, and it hands you an unusually honest re-entry: you churned, and looking at it, you never got the part that would have made this worth keeping. Can I show you? It's specific, it's true, and it isn't a discount.

Any triage framework that never says no is just a list. So:

"If basically you can just see over time their usage going down, then you could be like, not such a hot prospect. Maybe they just don't need this product anymore."

Steadily declining usage over a long period, ending in churn, usually means what it appears to mean. The need changed, or was never strong. Pursuing those accounts is how reactivation programmes acquire a reputation for wasting sales time — and how they get shut down.

Say no clearly, so the yes carries weight.

Four buckets:

| Pattern | Interpretation | Action | · |---|---|---| · | High engagement, no purchase | Blocker was budget or timing, not fit | Highest priority — lead with what they used | · | Zero engagement, no purchase | Never actually started | Re-offer the trial with a guided start | · | High engagement, then churned | Something external changed | Investigate; often a negotiation nobody had | · | Key features never activated | Onboarding failed, not product | Re-approach with the specific unrealised value | · | Long declining usage, then churned | Need genuinely went away | Deprioritise |

Put a number on it and the case gets harder to dismiss. Say a company runs 1,000 trials a year. At roughly a 75% non-conversion rate, that's about 750 people who evaluated the product seriously enough to sign up, and then didn't buy — against maybe 250 who did. The lost pool is three times the size of the win pool, every year, compounding as it goes unaddressed.

Split those 750 by engagement pattern and the picture changes. In my experience watching this kind of triage run, something like 300-350 of them never really engaged at all — an account created, a login or two, then nothing. Call that bucket dead on arrival; it needs a fresh start, not a follow-up. Another 150-200 will show the declining-usage pattern — real use early, tailing off for months before the trial expired — which is the closest thing to a genuine no the data can offer. That leaves the two buckets worth working hardest: maybe 150-200 who used the product heavily right up until it lapsed without ever converting, and a smaller group, perhaps 50-100 depending on how complex your onboarding is, who are customers rather than trials — they stayed, hit a wall, and were never shown the feature that would have made the difference.

None of those numbers are measured Nalpeiron data — they're the rough shape I'd expect from watching this problem across enough B2B trial funnels, and yours will differ. What doesn't change is the ratio that matters: out of 750 supposedly dead prospects, somewhere around 150-250 of them are still hot, sitting untouched, because nobody split the pool. That's not a rounding error in a pipeline. That's close to a quarter's worth of new-business quota, waiting in a spreadsheet nobody opened.

None of this is possible without granular usage visibility as a baseline capability. The same infrastructure that underpins usage-based billing turns out to be exactly what you need to run this kind of triage, even if you never charge by usage at all.

A framework without language is still just a spreadsheet, so here's roughly how each opening conversation should sound.

For high engagement, no purchase, lead with specificity, not a general check-in. Reference the exact features they used and ask directly whether the blocker was timing or budget, since you already know it wasn't fit: something close to "I saw you were deep into [feature], right up until the trial ended — did that stall on budget, or did the timing just not work?" That's not a cold reopen; it's picking a conversation back up.

For zero engagement, no purchase, don't apologise for the silence or ask why they went quiet. Offer a structured second attempt instead, ideally with a guided onboarding session already on the calendar, since an unstructured retry usually just repeats the same result.

For high engagement, then churned, the honest question beats the discount: ask what changed, and be ready to hear that nobody offered them a renegotiation before they left. That's a process failure worth naming, not papering over with a coupon.

For key features never activated, be specific about the gap rather than general about value. Show them exactly what they paid for and never used, and offer to close that gap directly rather than re-selling the whole product from scratch.

For long declining usage, then churned, mostly don't. If you do reach out, keep it low-effort and low-pressure — a single check on whether anything's changed, not a campaign. Spending real sales time here is how the whole programme loses credibility.

None of this works if a rep has to remember to go and look for it. Every one of these patterns is exactly the kind of thing worth wiring into an automatic alert that reaches a rep the moment it appears, rather than relying on someone remembering to check a dashboard — see Maximize Sales Alerts: Boost Revenue and Enhance Customer Engagement.

Timing matters almost as much as the message. Re-approach too soon and you look like you didn't listen the first time; wait too long and the blocker that stalled them has been replaced by a new one, or a competitor has closed the gap in the meantime.

For the high-engagement, no-purchase bucket, a reasonable default is somewhere in the 60-90 day range after the trial lapsed — long enough that a budget cycle or reorg has plausibly resolved, short enough that the product knowledge in their head hasn't gone stale. If you know their fiscal calendar, align to that instead of a fixed count of days; a budget that resets quarterly behaves differently from one that resets once a year.

For customers who churned without ever activating the key features, don't wait at all. That's a live account until the day it lapses, and the best moment to intervene is the first quarter of visible non-use, not after they've already left — it's cheaper to prevent this kind of churn than to reactivate it, even though the conversation you'd have is nearly identical either way.

For zero-engagement trials, timing barely matters, because there's no clock counting down from a resolved blocker — there's just an unfinished intention. Re-offer whenever it's convenient to run a properly guided second trial, ideally tied to a new capability or a different pricing structure that gives them an actual reason to try again, rather than a repeat invitation to do the thing that didn't work the first time.

For declining-usage churn, there's no clock worth setting at all. If the need genuinely resurfaces, they'll usually come back through a normal inbound channel rather than because you scheduled a follow-up eleven months out.

The obvious objection: this requires knowing what individual prospects did inside your product, at feature level, months after the fact.

Most companies can't answer that. They know a trial started and ended. They may know login counts. They rarely know whether a specific prospect used the three features that actually predict purchase.

That's an instrumentation gap, not a sales-skill gap. And it's why reactivation so often becomes an undifferentiated email blast to everyone who ever signed up — which performs poorly, confirms everyone's suspicion that reactivation doesn't work, and buries the pool for another year.

Closing that gap is separate, and honestly more foundational, work. I've gone into what actionable product usage data actually requires elsewhere — the short version is that most tools stop at session counts when the question that matters is which specific feature someone touched, and how often. Once that data exists, this stops being a sales-intuition exercise and starts looking like revenue intelligence: a rep opens the account and sees the same usage shape a data analyst would, without having to ask anyone for it.

The alternative isn't more outreach. It's less, aimed better.

You spent real money acquiring every one of those lost prospects. Marketing found them, sales worked them, and they used your product. The only thing missing was a reason to come back, at a moment when the original obstacle no longer applied.

The prospect who loved your product and didn't buy is not a failure. They're a deal with an unresolved blocker — and blockers expire.

Somebody just has to notice when.