Glencoco model versus personalized volume targeting
On this page
- The formula
- The funnel, as it actually is in 2026
- Three ways to buy a meeting
- Model 1: Glencoco. Kill the risk and friction, eat the ROI hit.
- The SDR is the thing that blows up the cost
- Model 2: Amplemarket. Buy the ROI back with better software.
- Model 3: Instantly. Sliding down the friction line.
- The lesson everyone learned the hard way: don’t sell “No SDR”
- Where the opening is
- Notes & sources
I started with a dumb question and it turned into a framework.
The question: Glencoco, a marketplace where you post a campaign and human reps cold-call on your behalf, charges companies something like $1,000 for a single qualified meeting. And you pay only when the meeting is booked and shows up. Meanwhile, a thousand dollars of infrastructure (domains, data, a sending tool, a dialer) will let you touch fifty thousand people. So how is one conversation worth the same as fifty thousand attempts?
The short answer is that volume is not pipeline, and the gap between the two is where every outbound business in 2026 lives or dies. The longer answer is a way of thinking about offers that I keep coming back to.
The formula
Here is the lens I’ve been using to evaluate any go-to-market offer:
Offer Strength = Perceived ROI ÷ (Risk × Friction to get started)
Three levers. You win by pushing the numerator up or either factor in the denominator down. Most companies obsess over the numerator (better data, better copy, better conversion) and ignore that a buyer’s decision is gated far more by risk and friction than by upside. The interesting move is usually in the denominator.
To use the formula you first have to be honest about the numerator, and being honest about the numerator means looking at the funnel as it actually behaves this year, not as the pitch decks describe it.
The funnel, as it actually is in 2026
Cold outreach has gotten meaningfully worse as a raw-volume game, and the numbers are not close.
The platform-wide cold email reply rate is now about 3.43%, down from roughly 5% in 2024 and around 8.5% back in 2019.1 Open rates get quoted around 27.7%, but Apple’s Mail Privacy Protection preloads tracking pixels and inflates that number so badly it’s close to useless as a real signal.1 Roughly nineteen out of twenty cold emails get ignored.
The stage that actually matters, meeting-booked rate, is the one most benchmark reports quietly skip. When you run the funnel all the way down, a cold, untriggered list converts somewhere around 0.3% to 0.6% of contacts into a booked meeting.2 So fifty thousand cold touches, executed competently, land you in the low hundreds of booked meetings, and then a chunk of those no-show, and then a chunk of the rest fail to qualify. You end up with dozens of genuine sales-qualified conversations, not thousands. The rest of your money paid for the privilege of being ignored at scale.
This is why the naive arithmetic (“$1,000 buys 50,000 touches”) is a trap. You aren’t buying touches. You’re buying the handful of qualified conversations that survive the funnel, and on a cold blast that number is brutally small.
Two things move it, and they’re the whole game:
Signal beats volume. Outreach triggered by a buying signal (a funding round, a job change, a hiring surge, a relevant website visit) converts to meetings at roughly 1.2% to 3.2% of contacts, which is four to five times the rate of an untriggered list.2 Same copy, same list quality, different timing. The prospect is already in motion.
Tight beats big. Campaigns under fifty recipients average a 5.8% reply rate against 2.1% for sends of five hundred or more, and verified lists roughly double the reply rate of unverified ones.3 Five hundred well-chosen contacts beat five thousand vaguely relevant ones, every time.
So the modern outbound funnel isn’t “send more.” It’s “send to the right person at the right moment, and prove you did the work.” That’s a reasoning-heavy motion, not a volume one. Hold that thought.
Three ways to buy a meeting
Now the fun part. There are three live models for acquiring qualified meetings, and the formula sorts them cleanly.
Model 1: Glencoco. Kill the risk and friction, eat the ROI hit.
Glencoco is a marketplace. You post a campaign, name your price per qualified meeting, and a network of vetted human reps calls on your behalf. They take a flat 30% of every bounty; the rep keeps 70%.4 Bounties typically run a few hundred to a couple thousand dollars per meeting depending on how senior the target is.4 It’s overwhelmingly a phone motion, roughly 98% cold calling and 2% email follow-up.5 Founded in 2022, it raised a $3M seed in 2023.6 (Some third-party aggregators float revenue and valuation numbers in the tens of millions, but those are estimates, so I’d treat them as directional at best.)
Look at what Glencoco does to the denominator. Friction is near zero: sign up, post a campaign, no burner domains, no tooling to configure, no team to hire. Risk is near zero: you pay per booked, qualified, attended meeting. If their reps dial for a month and book nothing, you owe nothing. They’ve turned “fixed cost with variable outcome” into “variable cost with fixed outcome,” which is genuinely the right trade for most buyers.
The catch is the numerator. At around $1,000 per qualified meeting, the ROI is mediocre unless your deals are large. Run it forward: if your AEs close 20% of qualified meetings, a $1,000 bounty implies a $5,000 customer acquisition cost. That only makes sense if your customer lifetime value is comfortably north of $5,000. Glencoco is beautifully designed for enterprise sellers moving $30K-$150K contracts. It structurally locks out startups, small teams, and anyone with sub-$5K LTV, which is a very large slice of the market, including a lot of the exact tools whose logos you’d recognize.
There’s an inconvenient fact that makes the $1,000 look less crazy than it first sounds, though, and it belongs to the next model.
The SDR is the thing that blows up the cost
Here’s the number that reframes everything. A fully-loaded in-house SDR does not cost the $75-90K OTE that founders anchor on. Once you add benefits, tooling, management overhead, recruiting, and ramp, the real all-in cost is $98,000 to $173,000 per year.7 And only about 43-57% of SDRs hit quota in a given period, with software-selling SDRs worst at around 41%.8 Median tenure is roughly 14 months, ramp is about three months, and the productivity plateau hits right around the time they leave.8
Divide the fully-loaded cost by the meetings a cold-working SDR actually produces and you get a real-world cost per meeting of $900 to $1,400.7
Read that again. The internal, “cheaper,” do-it-yourself option lands on top of Glencoco’s bounty. When a Chief Revenue Officer pays Glencoco $1,000 per guaranteed, attended, qualified meeting, they are paying roughly what an internal SDR costs per meeting anyway, minus the hiring, the ramp, the churn, the domain risk, and the months of paying full freight for zero output. Framed that way, Glencoco isn’t expensive. It’s an insurance policy against human inefficiency, priced fairly for people who can afford the premium.
So the SDR, specifically the salary-plus-turnover-plus-underperformance math, is the single variable that inflates a meeting from a software-baseline cost into the $500-$1,700 range. That’s the crux. Which raises the obvious question: what if you keep the software and remove the SDR?
Model 2: Amplemarket. Buy the ROI back with better software.
Amplemarket is the cleanest example of the opposite strategy: optimize the numerator by making a human absurdly efficient instead of replacing them.
Their pitch is a copilot that crawls the web, tracks your signal triggers, finds the targets, and pre-drafts personalized multichannel messages, then drops them into a unified task queue.9 The human’s job collapses from “manage spreadsheets and stitch tools together” into “review the AI’s work, edit, approve.” They replaced the external spreadsheet entirely with a native list workspace wired to their contact database, and they bundle deliverability safeguards so a careless operator can’t torch your domain reputation.9 Critically, they did not position as “fire your SDR.” They just made the software good enough that one lower-cost human runs a signal-first machine that used to need a team.
The result is a far better numerator. When a competent human is only touching the highest-leverage points and the software handles everything else, your cost per qualified meeting drops back toward the tens of dollars in the theoretical best case. The honest caveat: that floor assumes flawless execution, and real campaigns rarely execute flawlessly. But even the realistic version beats Glencoco’s economics by a wide margin, which is exactly why it’s viable for companies Glencoco can’t serve.
The cost is in the denominator. Amplemarket carries real friction (annual seats, connect your stack, train someone) and real risk (a human can still make a mistake), and it’s priced for teams, not solo operators, at roughly three hundred dollars per seat per month billed annually.9 Great ROI, higher barrier to entry.
Model 3: Instantly. Sliding down the friction line.
Instantly sits in the volume-software lane, but the interesting move is that they’ve noticed pure software leaves too much room for human error, so they’ve been layering an opt-in managed ecosystem: they provide the infrastructure and steer you toward vetted human-in-the-loop operators to actually run it.1 It’s the same recognition Amplemarket had, approached from the other direction: software alone is too risky in unskilled hands, so wrap it in a service that lowers the friction and the risk.
The lesson everyone learned the hard way: don’t sell “No SDR”
The tempting fourth model is full autonomy: an AI agent that is the SDR, no human in the loop. A wave of companies bet on this, positioning explicitly as human replacement.
The category hit a wall, and the reasons are structural, not incidental. Fully autonomous senders scale volume too fast and trip the exact spam patterns that Gmail and Microsoft now hunt for, so deliverability craters. Quality degrades at scale with no human to catch it. And even Amplemarket, no enemy of AI, notes flatly that full autonomy without oversight introduces compounding risk.9 But the deepest problem was positioning. When you tell a founder “this replaces your SDR,” their perceived risk spikes to a hundred percent: they imagine a rogue bot spamming a million-dollar target, and there’s no one to fire when it does. You maximized the denominator with your own marketing.
There’s also a moat hiding in plain sight. If autonomous AI is erasing the emailing SDR, the calling SDR gets more valuable, not less. A live human on the phone can handle a real-time objection, read tone, and earn trust in a way text still can’t fake convincingly. That’s precisely why Glencoco’s phone-first, human-first model survives the AI wave: it doubled down on the one channel automation can’t yet own.
So the positioning that works is not “fire your human.” It’s:
We bring the pre-warmed infrastructure, the signal-first data stack, and the vetted human-in-the-loop operators. You don’t manage a tech stack and you don’t manage a team. You approve the pre-verified signals, and you pay for pipeline.
That sentence lowers perceived risk (a human still guards the gate), lowers friction (the infrastructure is absorbed for you), and preserves the strong ROI of the signal-first, software-leveraged model. It takes Amplemarket’s numerator and wraps it in Glencoco’s denominator.
Where the opening is
Put the three models on the formula and the map is clear:
- Glencoco wins risk and friction, loses on ROI, and only works above ~$5K LTV.
- Amplemarket wins ROI, loses on friction and risk, and is priced for teams.
- Nobody yet fully owns the box that has Amplemarket’s ROI at Glencoco’s friction and risk, delivered to the founders and small teams that Glencoco’s price locks out.
That box is the whole thesis. The unit of outbound work has shifted from ten thousand sends to fifty reasoning-heavy, signal-triggered workflows, and the funnel data above is what makes that true: tight and triggered beats big and blind by four or five to one. Whoever productizes that motion, absorbs the infrastructure risk in code, and reduces the human’s job to high-leverage approvals, gets the unbeatable unit economics of Model 2 with the hands-off, zero-setup feel of Model 1.
The catch, and I want to be honest about it, is the “if.” Every failed autonomous-AI company also believed its execution would be flawless. The floor economics of the human-in-the-loop model are a gem if you can pull them off repeatably, and that word is carrying enormous weight. The difference between a category killer and a graveyard entry is entirely in whether the software actually holds the domain safe, actually keeps the human out of the bottleneck, and actually delivers when a real campaign goes sideways at 2am.
But the shape of the opening is not in doubt. The money is sitting in the gap between “too expensive for most companies” and “too risky to trust.” Close that gap and the arithmetic works.
Notes & sources
A note on method: the earlier version of this analysis leaned on some spuriously precise funnel numbers. I’ve replaced those with grounded ranges from primary 2026 benchmark reports where I could find them, and flagged the estimates that remain soft. Treat every number here as a directional band, not a promise. The funnel is yours to measure.
Footnotes
-
Cold email reply rate ~3.43% platform-wide in 2026 (down from ~5% in 2024, ~8.5% in 2019); open-rate distortion from Apple Mail Privacy Protection; Instantly’s move toward managed/HITL operators. Instantly Cold Email Benchmark Report 2026; Woodpecker 2026 statistics (20M+ emails); Snov.io 2026 cold email statistics. ↩ ↩2 ↩3
-
Meeting-booked conversion of ~0.3-0.6% for cold, untriggered lists vs. ~1.2-3.2% for signal-triggered sends (4-5x lift per 1,000 contacts). Formanorden, B2B Cold Email Benchmarks 2026 (signal-triggered vs. cold-list, B2B SaaS dataset); corroborated by Reachoutly and Cleverly 2026 meeting-rate ranges. ↩ ↩2
-
Reply rate of 5.8% for sub-50-recipient campaigns vs. 2.1% for 500+; verified lists ~2x reply rate of unverified. Belkins analysis of 16.5M emails (via Woodpecker 2026); Instantly benchmark data. ↩
-
Standardized 30% marketplace take rate (rep keeps 70%); name-your-price per SQL; bounties commonly a few hundred to ~$2,000 depending on target seniority. AlleyWatch (2023 founder interview); Glencoco.com; Glassdoor rep reviews; CanvasBusinessModel. ↩ ↩2
-
Glencoco activity mix ~98% cold calling / ~2% email follow-up. Glencoco.com (individual/rep page). ↩
-
Founded 2022 (Brooklyn/NYC), co-founded by Ingwon Chae (ex-Head of Growth, Wiz) and Jason Bao (ex-Verkada, Meta); $3M seed raised in 2023. AlleyWatch; StartupIntros. Note: third-party revenue/valuation figures (e.g. GetLatka) are aggregator estimates and are not independently verified. ↩
-
Fully-loaded SDR cost of ~$98K-$173K/yr (base + variable + benefits/taxes + tools + management + recruiting/ramp), well above the $75-90K OTE most founders anchor on; resulting real-world cost per meeting of ~$900-$1,400 for SDRs working cold prospects at 8-10 meetings/month. Bridge Group 2025 SDR Metrics Report; DevCommX SDR Program Cost analysis (2026). ↩ ↩2
-
SDR quota attainment ~43-57% (software SDRs ~41%); median tenure ~14 months; ramp ~3 months; annual turnover ~34-45%. Bridge Group 2025; RepVue 2025; 6sense BDR benchmark. ↩ ↩2
-
Amplemarket Duo copilot, native Lists workspace replacing spreadsheets, bundled deliverability safeguards, ~$300/seat/month billed annually; and the industry observation that full AI autonomy without human oversight introduces compounding risk. Amplemarket product materials and 2026 commentary; comparative reviews (2026). ↩ ↩2 ↩3 ↩4