Revenue Insights from Brandon Briggs - It's Just Revenue

The Stalled POC Acceleration: How to Read the Stall and Force a Decision

The POC went well. That part nobody disputes. The product did what you said it would, the users who logged in were impressed, and the technical evaluation came back clean. Then the deal stopped moving. The champion goes quiet. The login count drifts toward zero. The evaluation window quietly expires with no decision attached to it.

Most reps respond to a stalled POC the only way they know how: with more motion. Another check-in. Another training session. Another two-week extension. It feels like progress because it looks like activity. It is not progress. A stalled POC is not a problem you fix by doing more of the same. It is a diagnosis, and it is telling you something specific about the deal, if you are willing to read it instead of paper over it.

What is the Stalled POC Acceleration play?

The Stalled POC Acceleration is a framework for diagnosing why a proof of concept has stalled and driving it to a decision. Instead of adding activity, you read the stall as a signal, re-engage the economic buyer around business value, and set a binary decision point, converting more evaluations into a clear yes or no rather than a slow fade.

At a Glance

Best For SDRs, Account Executives, CSMs, and Sales Managers running evaluations
Deal Size Mid-Market (scales up to Enterprise)
Difficulty Medium
Funnel Stage Opportunity to Close
Impact High
Time to Execute Extended (7+ days)
AI Ready Yes: stall-risk scoring, stakeholder re-engagement, evaluation-plan generation

When to Run This Play

Run this play when:

  • A POC or trial hit its technical goals but the deal has not moved toward a decision
  • Usage has dropped off: logins are trending to zero and nobody is measuring the success criteria
  • The champion has gone quiet or is suddenly “busy with other priorities”
  • The evaluation window is expiring and there is no decision meeting on the calendar
  • A trial got extended, but nothing about the deal actually progressed
  • You have never had the economic buyer in a single conversation
  • The success criteria were never written down as pass or fail

Don’t run this when:

  • The POC genuinely failed on the merits; that is a product conversation, not this play
  • The evaluation is legitimately mid-flight and more data is honestly needed before a responsible call
  • You do not yet know who signs; find the economic buyer first, then run this
  • The stall is a known, dated external event, like a budget freeze with a real reopen date

Most of these triggers have one thing in common: the stall is not new information about the product. It is new information about the deal. The product question got answered weeks ago; the silence now is telling you the value question was never really asked.

The Framework

Start with the distinction that explains almost every stalled POC you will ever see: the difference between a proof of concept and a proof of value.

A POC answers one question: can it work? It is a feasibility test. A POV answers a completely different question: is it worth it? That is a value test. Here is the trap: almost nobody stalls because the product did not work. They stall because you proved feasibility to a room full of engineers and never attached a business-value number an economic buyer could sign against. The engineers are satisfied. Finance is unmoved. A POC with no POV behind it has nowhere to go, so it sits.

“A POC on an unqualified deal is not validation. It is a procurement delay mechanism you volunteered for.”

This is a Framework play, so here is the framework: five moves. The first four build a POC that does not stall. The fifth is what you do when it stalls anyway, because sometimes it will.

1. Write binary exit criteria before the POC starts

Before a single login is provisioned, you and the buyer agree on three to five criteria that are pass or fail. Not “evaluate the reporting.” Something like: “the ops team completes the weekly close in under two hours using the tool, measured in week three.” Binary. Measurable. Tied to a business outcome, not a feature.

If you cannot write those criteria, you do not have an evaluation. You have a science project with a login page. And if the buyer will not agree to them, that is diagnostic information delivered for free, before you have spent a dollar of engineering time.

“If everything in this POC goes exactly the way you hope, what specifically will be true at the end that makes this an obvious yes?”

What good looks like: Three to five pass/fail criteria, written down and agreed by the buyer, each tied to a number the business already cares about, with the decision meeting on the calendar before the POC starts.

2. Put the economic buyer, and a number, in the room on day one

The single most reliable predictor of a stalled POC is an economic buyer who has never been in the room. Engineers can validate that it works. Only the budget owner can validate that it is worth buying. If they are not engaged at kickoff, you are building evidence for a jury that is not in the courtroom.

“Day one” is not a slogan. It means the economic buyer helps define what a successful outcome is worth before the evaluation starts, so a win in the POC is automatically a win they have already agreed to fund. This is the Economic Buyer and Metrics muscle from MEDDIC deal qualification, applied before the evaluation instead of autopsied after.

“Who signs the contract if this works, and what number would make this an easy yes for them specifically?”

What good looks like: The economic buyer named, engaged, and on record about what would justify the spend, with a POV number attached to the POC, not just a feature checklist.

3. Time-box it and pre-book the decision

Every POC needs a hard stop and a decision date, set before kickoff. Two to four weeks for a mid-market evaluation, four to eight for a complex enterprise one. Beyond that window you are not evaluating anymore, you are in pilot purgatory: not killed, not shipped, just running quietly while attention moves on. Clear the bar and the win still has to convert, which is the pilot-to-production conversion play.

The forcing function is a decision meeting that already exists on the calendar. This is where a Mutual Action Plan earns its keep, and where most of them fail. A MAP the buyer did not help build is just your forecast dressed up as collaboration. Build it together, or it is theater.

“What date will we make the go or no-go call, and who has to be in that meeting for it to count?”

What good looks like: A time-boxed evaluation with a go/no-go meeting already scheduled, co-owned by the buyer, with the right decision-makers committed to attend.

4. Scope to one use case, and multi-thread the account

Two failure modes hide here. The first is scope: a POC that tries to prove five things proves none of them convincingly, so pick the one painful, quantifiable use case that maps to the buyer’s real problem and win that decisively. The second is single-threading. If your evaluation lives and dies with one champion, one name on the deal is a dead deal walking, because champions change jobs, get reorganized, or simply go quiet, and your POC goes quiet with them.

“Besides our champion, who else in this organization feels this pain, and who would notice if the problem never got solved?”

What good looks like: One tightly scoped use case tied to a quantified pain, and at least three stakeholders engaged across the account, including someone above the champion.

5. When it stalls anyway, read the diagnosis

Do all of that and some POCs will still stall. Good: now the stall is clean information, not noise, because you can see exactly which element went soft.

A champion who suddenly cannot get you a meeting was probably a coach, not a champion: someone who liked you but could never actually buy. A decision process nobody can describe was never mapped. Pain that everyone agreed was real but no one will quantify was never economic pain in the first place. The stall tells you which specific thing you skipped.

Then comes the move most reps cannot bring themselves to make: take the deal off the table. Offer to pause the POC. “It seems like this isn’t a priority right now, so let’s stop and pick it back up when it is.” You are signaling that you do not need this deal badly enough to nurse a zombie. One of two things happens: either the economic buyer materializes to keep it alive, which is exactly the access you could not get any other way, or nothing happens, which tells you the deal was never real. Both are wins, because both replace hope with truth.

“If we paused this evaluation today, would anyone on your side push to restart it, and who would that be?”

What good looks like: A stall that gets read, not smoothed over. A re-qualification to the economic buyer, or a clean, deliberate walk, inside days, not quarters.

A company I worked with brought me in to look at a deal the whole team called a lock: a little over $218K in ARR, a POC that had gone beautifully six months earlier, parked at “verbal yes” ever since. When I pulled the thread, it fell apart in about an hour. The daily-active usage the whole business case rested on had drifted to zero within weeks of the “successful” pilot. The champion who ran it was an enthusiastic manager with no budget authority and no internal air cover. The person who actually signed for software there had never joined a call. The POC had proved the product worked; nobody had proved it was worth buying, and the one human who could decide that had never been asked. That is not a deal that needs another follow-up. It needs a diagnosis, and the honest one was that it had been stalled from day one. We stopped forecasting it as real, took it off the table, and used the pause to force the question to the person who could finally answer it.

What Success Looks Like

Here is disciplined POC management next to what most teams actually live with. The gap is not a product gap; it is a qualification-and-nerve gap.

Metric Target What Most Teams Actually See
Binary exit criteria set before kickoff 100% of POCs Criteria written after the stall, if they are written at all
Economic buyer engaged By day one of the POC First met at the “why did this stall?” call, if ever
POC-to-decision conversion 50–70% reach a real yes or no Under 40%, and most of the “no”s are silent
Time in evaluation Inside the time-box (2–8 weeks) Open-ended; “let’s revisit next quarter”
Zombie evaluation rate Near zero Roughly a third of “active” POCs are past their window with no decision
Deals walked on purpose A deliberate, non-zero number Approximately zero; stalls slow-fade instead of getting called

The last two rows are the tell. A near-zero zombie evaluation rate and a deliberate walk rate are the numbers almost no team tracks, and the ones that separate a pipeline you can trust from a spreadsheet full of hope.

Handling Resistance

“The champion just needs more time.”

“Happy to give it, if the time comes with a decision date and the economic buyer in the room. What specifically will be true in three weeks that is not true today?”

More time almost never un-sticks a real stall, because time was never the missing ingredient; an extension with no new decision date and no new decision-maker is just a longer runway for the same non-decision. Been there: the “just needs more time” deal is the single most over-forecasted line item in B2B sales.

“We can’t push the economic buyer, we’ll annoy them.”

“Then let’s make it easy: a 20-minute session where they tell us what outcome would justify the investment. If they won’t give us that, we should both know it now.”

A champion who blocks all access to the person who signs is not protecting the relationship. They are revealing the ceiling of their own influence, and that is a warning sign, not a boundary to respect blindly. Sometimes the champion is genuinely powerful and just cautious; more often, the block is the diagnosis.

“Let’s just extend the trial to keep it alive.”

“We can extend, but only against new criteria and a firm decision date. Otherwise we’re not evaluating anymore, we’re subsidizing indecision.”

An extension is the path of least resistance for everyone, which is exactly why it so rarely produces a decision. The question is whether it answers something new or just buys three more weeks of the same silence. In a product-led motion, the same logic drives the trial conversion play: convert on a decision, not a calendar.

“Usage is low because they’re slammed right now.”

“Understood. If this solved a problem that was actually costing them money, would it be the first thing that gets dropped when the week gets busy?”

Busy is real, but also the most socially acceptable way to say “this is not a priority.” When the pain is genuine and quantified, people find fifteen minutes. When the tool sits untouched for three weeks, the usage is not the problem; it is the readout.

“Walking away means we lose the deal.”

“You can’t lose a deal that was never real. The walk is how we find out which kind we have.”

This is where forecast rigor separates the pros from the optimists. I have walked into enough pipelines to know the number a team reports as “active” and the number that is genuinely alive are often not the same universe. Naming a dead deal as dead does not cost you revenue; it gives you back the hours you were spending nursing a corpse, and it makes the rest of your forecast mean something.

Adapting to Your Buyer

By Persona

Economic Buyer (VP, CRO, CFO): They assume the product works, or they would not have let their team spend a month on it. Lead with the value decision: what solving this problem frees up or protects in their numbers, and what another quarter of not deciding costs. Give them a recommendation, not a recap.

Manager (Champion): Usually who ran the POC. Arm them to sell up: the business case in the language their boss uses, the two-slide version, and a clear ask. Do not make them invent the economic story alone; that gap is exactly where the deal stalls.

Individual Contributor (End User): Their enthusiasm is real data, but enthusiasm does not sign. Capture what they loved as evidence, and treat their adoption, or lack of it, as the earliest honest signal of whether the value is landing.

By Industry

SaaS and Technology: Usage telemetry is your friend; the drop-off is measurable and undeniable. Use it to force the conversation early, not as a post-mortem.

Financial Services: The economic buyer is often two levels up and governance-driven. Map the decision and paper process before the POC; the stall here is usually procedural, not emotional.

Healthcare: Evaluations move slowly for legitimate reasons, from clinical workflow to compliance. Time-box generously but still time-box, and name the budget owner early, because “the clinicians love it” rarely reaches the person who signs.

Manufacturing: Tie the POC to a single line or site and one hard operational number, like downtime or throughput. A stall usually means the number was not painful enough, or the plant manager who feels it was never in the room.

How AI Changes This Play

AI cuts both ways on stalled POCs, and it is worth being honest about both.

First, the uncomfortable part. AI made pilots almost free to start, so organizations now run more evaluations with less discipline per evaluation: a recipe for more stalls, not fewer. Deloitte found 66% of organizations experimenting with AI agents and only 11% getting them into production. Gartner expects more than 40% of agentic AI projects to be canceled by the end of 2027, most of them hype-driven POCs launched without governance or a clear return on investment. If you are running a POC right now, “we ran a pilot” is no longer evidence of anything.

Used with judgment, AI is genuinely good at the diagnosis this play depends on:

  • Stall-risk scoring: Feed usage data, stakeholder engagement, and days-in-stage into a model that flags a POC drifting toward zombie status before the window closes, not after.
  • Stakeholder-specific re-engagement: Draft different re-entry messages for the champion, the economic buyer, and the skeptic, each speaking to what that person cares about instead of a generic “just checking in.”
  • Evaluation-plan generation: Turn a messy kickoff into a clean, binary exit-criteria document and a mutual action plan in minutes, so step one stops depending on the rep’s memory.
  • Meeting prep that reads the whole history: Summarize every touch, usage dip, and unanswered question so you walk into the re-engagement knowing which element went soft.

That last one is where persistent context earns its place. A learning system that actually remembers the account, like Tempreon, lets you read a stall against the full qualification history: what the buyer said in week one, which MEDDIC element was always soft, when usage really turned. That beats re-diagnosing a stalled deal from a blank CRM record and a rep’s fading memory.

But notice what AI does not do. It can score the stall; it cannot decide to walk. The model tells you the POC is dying; only a human decides whether to force the economic buyer into the room or take the deal off the table. The judgment is still yours.

Ready-to-use AI prompt for diagnosing a stalled POC:

You are a deal-stage diagnostician. I will describe a stalled POC.
Read the stall and tell me what it most likely means. Do NOT suggest
adding more activity.

Inputs:
- What we were proving (the use case): [use case]
- Success criteria we agreed on (or did not): [criteria]
- Usage during the POC (trend over time): [logins/activity]
- Stakeholders engaged and their roles: [names/titles]
- Has the economic buyer been in a conversation? [yes/no + detail]
- What the champion has said in the last two weeks: [quotes]

Output:
1) The single most likely reason this has stalled, framed as a
   missing MEDDIC element (Economic Buyer, Metrics, Decision
   Process, Champion, etc.)
2) The one question I should ask that would confirm or kill that
   hypothesis
3) A recommended next move: re-qualify to the economic buyer,
   restructure the POC around a value number, or walk
4) A two-sentence "take the deal off the table" message I could
   send if walking is the right move

Sidebar: the most-cited AI-failure stat is itself a stalled pilot

You have probably seen the headline that 95% of enterprise AI pilots fail. It is the most-quoted number in the category, and a near-perfect example of what this post warns against: trusting a headline outcome you did not define yourself.

Inherit someone else’s definition of success, and you inherit their blind spots.

Dig into the MIT report behind it and it gets shaky fast. The authors call their own findings “directionally accurate” and note they rest on interviews, not company financials. Wharton’s Kevin Werbach called the central claim “deeply problematic.” Honest reconstructions land closer to “about three-quarters of launched pilots missed a deliberately high six-month bar,” a very different and far less viral sentence than “95% fail.” The lesson is not that AI pilots are fine. It is that if you accept someone else’s yardstick, you are running their POC, not yours. Define your own binary criteria up front, and you never have to wonder whether the headline applies to you.

Related Plays

  • MEDDIC Deal Qualification – A stall usually names the missing MEDDIC element. Use this to pressure-test Economic Buyer, Decision Process, and Metrics before the POC, not after.
  • Pilot-to-Production Conversion – The companion play for when the POC actually succeeds: how to turn a strong result into a signed contract before it drifts.
  • Trial Conversion Play – For product-led and self-serve motions, how to convert an active trial without defaulting to the endless extension.
  • No-Decision Prevention – Since 40 to 60% of B2B deals die to no decision, this is the upstream discipline that keeps a POC from becoming one.
  • Mutual Action Plan – The buyer-built decision timeline that turns a time-box from a threat into a shared commitment.
  • Competitive Displacement – When the stall is not indecision but a quiet incumbent preference, this is how you surface it and unseat it.

The Close

A stalled POC is not a problem to fix. It is a diagnosis to read.

If you remember nothing else: motion feels like progress, which is exactly why it is so dangerous here. Another check-in, another extension, another training session, they all let you feel busy while the deal quietly dies. The only thing that counts as progress is an outcome: a value decision, yes or no, made by someone who can actually sign. Everything else is theater with a login page.

If you have a stalled POC on your board right now, do not schedule another touchpoint. Read the stall, then act on it. If you have found a cleaner way to force the decision, I would genuinely like to hear it.

Sources & Further Reading

Frequently Asked Questions

Why do proof-of-concept deals stall even when the POC succeeds?

Because a POC only proves the product can work, not that it is worth buying. Deals stall when you prove feasibility to technical evaluators but never attach a business-value number an economic buyer can sign against. The engineers are satisfied and finance is unmoved, so the proof of concept won’t close no matter how clean the result was.

What is the difference between a POC and a POV?

A proof of concept (POC) answers “can it work?” It is a feasibility test. A proof of value (POV) answers “is it worth it?” It is a value test tied to a business outcome and a dollar figure. Most stalled POCs are missing the POV: nobody quantified what solving the problem is worth to the person who controls the budget.

How long should a POC last before it is a red flag?

Time-box it: roughly two to four weeks for a mid-market evaluation and four to eight for a complex enterprise one, with the decision meeting scheduled before it starts. Past that window, a POC becomes pilot purgatory, an evaluation that is neither killed nor shipped, and the odds of a decision drop the longer it drifts.

Should you walk away from a stalled POC?

Often yes, or at least be willing to. Offering to pause or take the deal off the table forces the question: either the economic buyer steps in to keep it alive, giving you the access you could not get otherwise, or nothing happens, which tells you the deal was never real. Willingness to walk is the leverage.

How do you unstick a stalled pilot?

Do not add activity. Read the stall as a diagnosis, identify the missing element (usually the economic buyer, a quantified value, or a mapped decision process), and force a binary decision by re-engaging the buyer around value and setting a firm go/no-go date. If it still will not move, walk, and trust that a clean no is worth more than an open maybe.

About the Author

Brandon Briggs is a fractional CRO and the founder of It’s Just Revenue. He’s built revenue engines at six companies — including Bold Commerce, Emarsys/SAP, Dotdigital, and Annex Cloud — scaling teams from zero to eight-figure ARR and helping build partner ecosystems north of $250M. He now helps growth-stage companies fix the gap between activity and revenue. Connect on LinkedIn.

Part of the It’s Just Revenue Sales Plays Library — practical frameworks for revenue teams who want to stop the theater and start closing.