The proof of concept is the most expensive stage of the enterprise sales cycle and the least well-run.

Between 40% and 60% of POCs end in no decision. Not lost to a competitor. Not lost on product fit. Lost because the vendor never built the conditions for a decision to happen.

I hosted a session on this yesterday with Noam Harel (Head of Growth, Demostack), John Care (author of Mastering Technical Sales and Chief Content Officer at Up 2 Speed), and Gilad Komorov (three-time CRO and owner of SaaStimize). Demostack sponsored the conversation.

What came out of it was less a playbook and more a diagnosis. Sales teams are running POCs the same way they ran them ten years ago, and the math has stopped working.

Here's what stood out.

The number every revenue leader should have memorized: a single enterprise POC costs somewhere between $35,000 and $70,000 to run.

That number stacks the marketing spend to generate the lead, the SDR time to qualify it, the AE hours to nurture it, the SE time to build the demo environment, the customer success prep, the internal infrastructure, and the opportunity cost of the whole team's attention.

Gilad's reframe: view every POC through that price tag. Not as a step in the funnel, but as a $50,000 investment your company is making in a specific outcome.

That single mental shift changes how teams qualify who's worth running one for. It changes how many objections a champion has to overcome internally before the SE spins up an environment. And it changes what "no decision" actually costs the business, because 40% to 60% of $50,000 investments dying is a P&L problem, not a sales problem.

The median enterprise SaaS acquisition cost is nearly $2 for every $1 of revenue. That number stays broken as long as the POC stage keeps running the way it does today.

John's stat is the one every sales team should tape to a wall.

42% of enterprise deals in the pipeline end up won by what he calls Do Nothing Incorporated. DNI. The customer chose to not decide, to stay on their current system, to reallocate the capital, to punt.

Competitive analysis in enterprise sales usually stops at the one or two named vendors in the deal. DNI almost never makes the matrix.

John's argument: DNI should be a column in every competitive matrix. It should have its own battlecard. It should be treated as the most common competitor in the enterprise SaaS market, because it is.

The tactical implication is that consensus-building inside the customer's org isn't a nice-to-have. It's the counter-strategy to indecision. Every additional stakeholder you bring into the buying group is one more person who has to say no to preserve DNI's win. Land-and-expand during the POC, not after.

John's line on this: "It's easier to win a deal from no decision than from one of your competitors."

Gilad and John kept circling the same reframe from different angles.

The default frame for a POC across enterprise SaaS is technical validation. The product either does the thing or it doesn't. If it does, the deal moves. If it doesn't, the deal dies.

That model is wrong.

Gilad's version: "The POC is not just about proving your tech works. It's about building alignment around value, around impact, and around what the customer needs to move forward."

John's version, from his years as a CIO evaluating enterprise software: "The technology and the business are intertwined like DNA. Both strands need to be complete to win the business."

The vendors who lose after achieving the technical win did the technical work and stopped there. The vendors who win did the technical work and then linked it, feature by feature, to a specific business outcome the customer had already told them they cared about.

John shared the story that captured the reframe better than any framework could.

He was a CIO evaluating programmer productivity software. Three vendors ran POCs, each about seven or eight days. The company was planning a $4.5M purchase. The stakes were higher than the ticket: IT was on the critical path for a new product launch, and a delay would cost the business $8M and push the launch back eight weeks.

At the end of the POCs, the three vendors scored 74, 70, and 58 on the technical evaluation. Normal purchasing logic says the 74-point winner takes the deal.

The 74-point vendor didn't win.

The 70-point vendor's SE, in the final presentation, walked through the demo and said: "John, by doing this, you save two and a half weeks off your development schedule. By doing this, you save another twelve days." Feature by feature. By the end, they had accounted for eight weeks of savings.

Which was, not coincidentally, the exact delay John was trying to avoid.

The 70-point vendor won the $4.5M contract. They didn't have the better product. They had the better linkage from product to business outcome.

John's read on it, twenty years later: "It's not always the best product that wins. It's the people who link the tech back to solving the business problem."

The research backs him up. The Challenger Sale found that 53% of the purchase decision came down not to the product itself but to the customer's experience of the evaluation process.

Buyers are evaluating the vendor as a partner. The product is one of several signals feeding into that evaluation.

Gilad laid out six specific patterns that kill POCs. Two of them account for the majority of the damage.

The champion inside the customer's org is not the buyer. If the seller hasn't validated that an executive sponsor or economic buyer is engaged and prepared to approve budget on success, the POC is a technical exercise, not a buying process.

Gilad's rule as CRO: full MEDDPICC analysis before any POC begins, with explicit executive validation of the question "If we succeed, will you allocate resources and funds toward this solution?"

If the champion can't get that confirmation, the POC doesn't start.

The most common failure mode. The kickoff call opens, someone from the vendor side says "let's dive in," and the team starts configuring the environment before agreeing on what success looks like.

Gilad's framework: before any technical work begins, align on five things. Objectives. Success criteria. Stakeholders on both sides. Timeline. Process.

The silent killer inside this killer is scope creep. Customers who keep asking to test additional features are usually signaling that they don't know what success means to them yet. The right response is to hold the scope, explain that anything outside it becomes post-POC deployment work, and force the definition-of-success conversation back to the surface.

John's addition to this: "If the customer can't define success, that's a red flag. If they don't know where they're going, how do they know when they get there?"

The remaining four killers Gilad walked through are generic demo environments, no POC management, zero visibility into engagement, and a weak or missing wrap-up. Each one deserves a section of its own, but the two above cover most of the damage.

Gilad's alignment framework is worth stealing straight.

Every POC should have three stakeholders identified on the vendor side and three on the customer side.

Vendor side: account executive, solution engineer, executive sponsor.

Customer side: business champion, technical champion, executive sponsor.

The more of those six roles are actively engaged during the POC, the less indecision risk the deal carries. The fewer of them are engaged, the higher the DNI probability.

The three-by-three doesn't just build consensus. It builds redundancy. If the technical champion leaves the company mid-POC (which happens), the business champion carries the deal. If the vendor's SE gets pulled onto another account, the executive sponsor holds the relationship.

Losing POCs are usually running with two-by-two or one-by-one alignment and don't know it.

John's team at Up 2 Speed has run the numbers on this across hundreds of clients. His finding: roughly 60% of demos and 50% of POCs are happening too early in the sales cycle.

The reason is structural. When a sales team doesn't know what to do next, the default move is to escalate. The rep says "let's do a demo." The customer says yes. Then the customer says "let's do a POC." The rep says yes. Everyone is doing something.

The problem is that the escalation replaces qualification. A demo before proper discovery is a demo without a story. A POC before executive sponsorship is a technical exercise the customer is running to sharpen their own thinking, not to buy.

John's mental model: think of the seven levels of effort available to close an enterprise deal. Level one is references. Level two is reference visits. Levels three through five are demos of increasing customization. Level six is the POC. Level seven is a full production pilot.

Sales teams skip levels one through five and dash to level six by default. The teams that win use levels one through five to earn the right to run a POC that will actually convert.

This is where Gilad and John disagreed on the panel.

John's read across his 500-client base at Up 2 Speed: only a single-digit percentage of enterprise POCs today are paid. Customers put time, people, and infrastructure into a POC, but rarely money.

Gilad pushed back. He's seeing paid POCs rise, particularly in the $10K to $15K range, which he suspects is the standard budget procurement teams allocate for product evaluations.

His preferred structure, from his CRO experience: a framework agreement. Sign a twelve-month contract with a three-month opt-out. The first three months are effectively the POC. If it succeeds, the customer rolls into the full agreement seamlessly. If it fails, they opt out.

The advantage is on the back end. Post-POC negotiations often stall for months on legal, procurement, and contract terms. Framework agreements move all of that work upfront, which means a successful POC converts to revenue without the six-week paperwork purgatory that kills momentum.

Gilad's caveat: this requires more upfront legal work and only makes sense with customers who have the appetite for it. Not universal. Real when it fits.

John's closing tactical recommendation is the one most teams will implement first.

If you don't have an AI bot sitting on top of every active POC monitoring what the customer is doing (with permission), you're losing information you're paying for.

The bot logs usage patterns. It flags when a customer goes silent for more than 48 hours. It surfaces which features are being tested versus which were agreed to. It gives the AE a signal to reach out before the POC drifts into scope creep or silence.

Noam's version of the same recommendation from the Demostack side: analytics on product simulation environments give the same visibility, structured around agreed success criteria. Shared Slack channels. Structured check-ins. Anything that turns customer engagement into a real-time signal instead of a post-mortem.

Silence during a POC is not neutral. Silence is DNI winning quietly.

If the panel's playbook compresses into something a head of revenue could run on Monday:

Put a $50,000 price tag on every POC. Then decide whether the qualification you have justifies the spend.

Treat Do Nothing Incorporated as your primary competitor. Build a battlecard for it. Track win rate against it.

Run three-by-three alignment on every deal. Six named stakeholders before technical work starts.

Refuse to enter a POC without executive validation of the outcome. If the champion can't answer "will budget be allocated on success," walk.

Link every technical outcome to a specific business result. Feature to time saved, feature to cost avoided, feature to revenue enabled.

Wire real-time visibility into every active POC. AI bots, analytics platforms, structured Slack channels. Silence is not neutral.

Prepare the wrap-up like a board presentation, not a project close. Dry run with the champion. Testimonials from users. A firm action plan for post-POC advancement.

Enterprise win rates have dropped from around 25% to 17-18% in the last few years. The POC is the stage where most of that erosion is happening, and it's also the stage where the most leverage exists to reverse it.

The teams that win the next 18 months of enterprise SaaS won't be the ones with the best product. They'll be the ones running the tightest POC process, competing hardest against no decision, and treating the technical win as a starting line instead of a finish line.

Want to get your brand in front of 60,000+ revenue leaders?

Partner with us and put your name at the center of the conversation.