The most expensive moment in a B2B sales cycle does not look like one.

An account executive is thirty minutes into a live call. Discovery is finished, the buyer is engaged, and the conversation has moved past polite curiosity into the part where deals get decided. Then comes the question that matters: how the integration handles a legacy data model, or what the security architecture looks like once procurement gets hold of it.

The account executive does not know. The sales engineer who does is booked three deals out.

So the rep offers the only thing available, which is a promise to follow up. The call ends cordially and the follow-up lands sometime next week. Nothing about it registers as a failure, and that is the problem. By the time the expert reaches the buyer, ground that was already won has to be won a second time, and a sales cycle that was already long gets longer.

The Impossible Paradox

That scene is the visible end of three forces pulling against each other: growth costs more than it used to, buyers will wait less than they used to, and the people who could fix both do not scale. Call it the Impossible Paradox, because the usual answer to any one of the three makes the other two worse.

Start with cost. Alexander Group puts the rise in customer acquisition costs at between 40% and 60% across most segments since 2023 alone, and sales cycles have stretched from an average of 107 days in early 2022 to 134 days today. It now takes around 20 months to earn back what a company spends winning a customer, so new accounts sit at a loss for the better part of two years.

Buyer patience has gone the opposite direction. The same executive who spends four months evaluating a platform resolves a consumer purchase in ninety seconds, and does not consciously recalibrate between the two.

Boards have not responded by lowering growth targets. Revenue leaders are asked to grow faster on leaner budgets, and the instinct is to hire. That is where the third force bites. Headcount adds capacity without adding availability at the moment a particular buyer asks a particular question, and intent does not arrive on a staffing schedule.

The Handoff Tax

Most deals do not die at no. They die in the gaps between people. Every time a buyer is passed from one role to the next, context, momentum and trust are lost in the transition. The cost recurs on every opportunity in the pipeline whether or not it shows up in a dashboard. Call it the Handoff Tax.

The relay runs roughly the same way at most companies. A buyer arrives ready. A booking chatbot captures a calendar slot and disappears, converting live intent into an invitation four days out. A sales development rep qualifies the account and passes it to an account executive, and the buyer explains the situation a second time. The AE runs the deal until it needs technical depth, at which point the buyer explains it a third time to a sales engineer pulled in the day before.

The first two transitions are the warm-up. The third is where the real money goes, because it happens after the company has paid for everything: the demand was purchased, the meeting was earned, and the buyer is asking the question that signals serious evaluation. Coverage runs out at exactly that point.

The scale of that loss has a number attached to it. Matthew Dixon and Ted McKenna, whose study of 2.5 million recorded sales conversations became The JOLT Effect, found that between 40% and 60% of deals are now lost to buyers who state an intention to purchase and then fail to act. Those deals are not lost to a competitor. They stall out between wanting the product and signing for it.

The reason that friction exists is structural. Sales engineers, also called solutions engineers, are the people who answer the hard technical question, and there are not many of them. 1mind puts the median at four account executives for every sales engineer, reaching ten to one in many organizations. Independent benchmarks through late 2025 land nearby, at a median around five to one that widens past seven once a company passes a hundred reps. Applied against a full calendar, any of those figures means the same thing. Most live sales calls happen with no technical depth in the room.

That thinning has a price. Alexander Group, benchmarking 100 sales forces, found companies staffed at one sales engineer per five reps averaged $2 million in revenue per rep, against $3.2 million for those staffed one to one. Richer coverage follows more complex products, so the causation runs both ways, but a gap that size is one most revenue leaders would want explained.

Hiring more of them will not fix the coverage problem. Sales engineers are scarce and expensive, and few companies can justify holding one in reserve against a question nobody knew was coming.

“The most expensive handoff in your pipeline is the one where your AE is alone on the call and the expert is booked three deals out,” says Jonathan Kvarfordt, 1mind’s VP of marketing.

Designing for the buyer

The root cause is a design choice nobody set out to make. Go-to-market was assembled to serve the seller’s process, and the buyer’s experience was left to emerge from whatever that produced. Routing, qualification and calendar capture each became a software category with its own budget line, and every one added a place for the buyer to be handed off.

The arrangement held while buyers were willing to wait. It is failing now because buyers arrive having already researched through search, communities and, increasingly, an AI chatbot. They show up with a specific constraint and want a specific answer. Asking them to wait a week while a human frees up means asking them to absorb a vendor’s staffing model as their own inconvenience.

The direction of travel was visible before the tools existed. Speaking at Gartner’s CSO conference in 2023, analyst Robert Blaisdell argued that generative AI and digital humans could transform the seller’s role, and that the shift was about “giving technology more responsibility” instead of handing sellers more of it. The seller-facing version of that arrived first. The buyer-facing version took another three years.

The destination is a B2B buying experience closer to the better consumer ones: immediate, on the buyer’s terms, with depth available at the point of the question. The companies that win the next decade will shorten the distance between a buyer asking and a qualified answer arriving. The economics favor it, since the demand has already been paid for and converting more of it costs less than buying more.

Putting the expert in the room

1mind, founded by Amanda Kahlow, is building against that gap. Kahlow founded and ran 6sense before leaving in 2020, and 1mind emerged publicly in November 2025 with a $30 million Series A led by Battery Ventures, taking total funding to $40 million.

The company’s framing is a set of go-to-market Superhumans: one continuous system with a face, a voice and account-level memory, deployed across the whole buyer journey. On 19 May 2026 it shipped the Superhuman built for the live sales call, which it bills as the first AI to join such calls as a visible, named participant speaking directly to buyers.

The Ride-Along Superhuman appears on Zoom, Teams or Meet as an attendee alongside the account executive, listed under a human name. On 1mind’s own calls the persona is Nigel. It runs in one of two modes, set per meeting: active, or silent and listening only. In active mode it answers technical questions as they arise, presents slides on request, and can run a live demo with cursor control through the product interface. 1mind says proprietary guardrails hold it to company-approved information on pricing and technical claims. It steps back on cue, and the rep keeps command of the room.

The distinction that matters is who the AI is talking to, because there is already a category of AI on live sales calls and it points the other way. Rep-assist tools help the seller: software listens to the conversation and surfaces a competitive fact or a suggested answer in a side panel, and the rep reads it back. The buyer still waits for a human to relay it, and still hears it from someone who learned it four seconds earlier.

“Rep-assist tools whisper to the seller,” says Kvarfordt. “The Ride-Along Superhuman talks to the buyer, on the call, with sales-engineer depth.”

1mind’s own engagement data shows Ride-Along sessions running far deeper than website chat, with a median near 26 conversational turns, almost entirely spoken. One finding shaped the format. On live calls, buyers prefer voice with no face on screen, because during a working session a face competes for attention.

What has to be true

The obvious objection is that this is headcount reduction in friendlier language. No company staffs a skilled sales engineer on every live call at the moment of need, so the Ride-Along Superhuman operates where a human was never going to be. The repetitive technical work moves to the AI. Judgment and relationship stay with the rep.

The less obvious risk is trust. A live call is the highest-stakes surface an AI vendor could have picked, because a security question answered wrongly in front of a buying committee does more damage than the follow-up email it replaced. Guardrails are what 1mind points to, and the difficult part of that engineering is the refusal. A participant that overreaches once costs more credibility than an empty seat ever did.

The commercial numbers deserve the same scrutiny. 1mind reports conversion lift of two to five times, sales cycles 62% shorter, and deals worth $110,000 closed with no human involved. Those are first-party figures from an early-access cohort, none independently audited. They describe what happened at some early customers, which is not the same as what a new buyer should expect.

One system, fewer gaps

The live call is where the Handoff Tax bites hardest, though it is not the only place it gets levied.

The same leak appears when a website visitor gets a calendar invitation and nothing else, and when a new customer explains their business from scratch to a support team that has none of the context sales collected.

Single-purpose tools address one of those moments and leave every handoff around them intact. 1mind’s argument for a single system is that the same brain carries the account from first touch through renewal, leaving no seam for context to fall through.

At least one customer has tested that rather than taken it on faith. Experity, a healthcare software company, replaced its legacy chatbot and ran 1mind against a rival tool across the same funnel. Win rates went from 26% to 50%. “We were running a 2015 motion in a 2026 market,” says Jonathan Moss, Experity’s EVP of patient engagement. The numbers still come from 1mind and it is one company, but the comparison was run rather than asserted, which is more than most vendor evidence offers.

The belief underneath all of it is simpler than the architecture. “The way people buy has changed,” Kahlow said at launch. “The way companies sell has not.”

Whether 1mind is the company that closes the gap between the two will take longer than a funding round to establish. The seat next to the rep is empty on most calls, and the cost of that is now legible enough that somebody is going to sell the fix.

Get the TNW newsletter

Get the most important tech news in your inbox each week.

TNW newsroom and editorial staff were not involved in the creation of this content.