From outbound to outcomes: 2.6× meetings, 45% lift on close rates, no new sales hires.
A 78-person commercial team at a Central European industrial-SaaS company deployed eight production agents on Claude Code and trained 31 SDRs and AEs to operate them — rebuilding the outbound motion as a precision instrument rather than a volume game. Meetings booked rose 2.6×, SQL-to-close climbed from 11% to 16%, and the company added €2.4M in net new ARR in nine months.
The client is a Central European industrial-SaaS company — a profitable, growing business selling operations-management software to mid-market manufacturers across Europe. Anonymized here, but recognizable in the segment: ARR in the €40–60M range, 78 people in the commercial organization, average deal size €60K–€220K ARR, sales cycle 3–7 months. They had grown well through 2022 on inbound and partner-led pipeline. By 2024 they had hit a wall.
The wall had a specific shape. Their outbound motion — the part of the business they would need to scale to triple ARR over the next three years — was not working. They had hired and trained 18 SDRs over 18 months. The team had collectively sent more than 240,000 outbound emails, made more than 60,000 calls, and built a pipeline that was not converting at acceptable rates. The CRO's diagnosis was correct: the volume was high, the signal was low.
Meanwhile, the AEs were drowning in their own version of the same problem. Each AE managed 25–40 active opportunities. A real read of the situation in any one of those opportunities required 4–6 hours of research that no AE actually did. They went into every meeting with a generic deck and a hopeful smile, and the close rates reflected it.
The CRO came to us with a sharp brief: "I do not need more people. I need each of these people to operate at the level of the best account executive in the company. I want them in Claude Code by the end of next month."
The constraint: outbound is broken because it sounds the same
Most "AI for sales" tools are designed to increase volume — more emails, more calls, more sequences, more touches — at a moment in B2B sales history when the marginal value of additional volume has already crashed to zero.
Buyers are drowning in identical outbound. They have learned to filter it ruthlessly. The marginal email and the marginal call have negative expected value: they consume your team's reputation faster than they generate pipeline.
The opportunity is not to send more. The opportunity is to say something a buyer has not heard from anyone else in the category in the last six months. That requires depth: real research, real understanding of the buyer's situation, real reasoning about why the buyer might genuinely benefit from your product right now. Depth is what Claude Code enables, when it is on the SDR's terminal and operated as a craft tool.
The portfolio: eight agents covering the full sales motion
Wave 1 — Built by Orchestrary in months 1–4
ACCOUNT-INTELLIGENCE · Deep account research per target
- Time per dossier: 5–40 min generic → 6–10 min review (dossier generated unattended)
- Depth: reaches a research depth that would have required 4–6 hours of manual work
- Personalization quality: reply rate 2.1% → 5.8% (2.8×)
- AE preparedness on first meetings: improved by 1.4 points on the company's 5-point scale
OUTBOUND-COMPOSER · Personalized outbound drafting
- Reply rate on cold outbound: 2.1% → 5.8%
- Meetings booked per SDR per month: 8–12 → 22–28
- Outbound volume per SDR per month: 1,200 → 580 emails (52% reduction)
- Bounce rate, spam-complaint rate, and unsubscribe rate all dropped meaningfully
DISCOVERY-PREPPER · Pre-call preparation for AEs
- Time per discovery prep: 15–30 min skim → 8–12 min structured review
- Discovery → SQL conversion: 38% → 52%
- Internal call quality scoring: improved across every dimension tracked
PIPELINE-AUDITOR · Weekly pipeline review and signal extraction
- Forecast accuracy at quarter close: ±18% → ±7%
- Time per AE per Monday-morning prep: 2–3 hours → 40 minutes
- "Stalled deal" identification: now flagged 2–4 weeks earlier on average
Wave 2 — Built by the sales team itself using Claude Code (months 5–9)
PROPOSAL-CRAFTER · COMPETITOR-DECODER · CHURN-WATCHER · EXEC-BRIEFER
PROPOSAL-CRAFTER reduced proposal drafting from 8–12 hours to ~90 minutes; proposal-to-close rate improved 7 points. COMPETITOR-DECODER aggregates win/loss notes into competitive playbooks per opportunity; the AE who built it has personally won three deals against the named competitor in four months. CHURN-WATCHER identified five at-risk accounts in the engagement period — four were saved, totaling €480K in retained ARR. EXEC-BRIEFER produces complete briefing packs for the CRO or CEO joining customer calls.
The aggregate impact: a sales motion that compounds
| Metric | 12 months prior | 12 months from start (annualized) | Change |
|---|---|---|---|
| Outbound emails (org-wide) | 240,000 | 115,000 | −52% |
| Reply rate on cold outbound | 2.1% | 5.8% | +176% |
| Meetings booked per month | 220 | 572 | +160% |
| Discovery → SQL conversion | 38% | 52% | +37% |
| SQL → close rate | 11% | 16% | +45% |
| Sales cycle length (median) | 142 days | 104 days | −27% |
| Net new ARR (12-mo rolling) | €8.4M | €10.8M | +€2.4M |
| Headcount (commercial org) | 78 | 78 | Flat |
| Forecast accuracy | ±18% | ±7% | −61% |
SDR time on outbound research and drafting: ~60% → ~24%. AE time on pre-call prep and post-call admin: ~40% → ~19%. AE meaningful customer-facing time: ~30% → ~52%.
Strategic outcome
The CRO's three-year ARR plan — which his board had treated as ambitious — is now ahead of schedule. The company has chosen to grow into the existing headcount rather than adding sales capacity. The discussion in board meetings has shifted from "do we need to hire 20 more SDRs" to "do we want to capture this margin uplift or reinvest it in product."
Why "deploy AND teach with Claude Code" was right for this sales org
1. The leverage is in depth, not volume
Hosted AI-for-sales tools are almost without exception designed to scale volume. The market has moved past the point where this works. The leverage is in depth — saying something specific, grounded in real research, that the buyer has not heard before. Volume tooling produces noise.
2. The reps had to operate the agent themselves
The CRO refused on principle to deploy a system that produced outbound with no rep in the loop. He had watched competitors do this and watched their domains end up on every spam blocklist within 90 days. Claude Code preserves the rep-in-the-loop discipline by design.
3. The data, the playbooks, and the leverage stay with the company
A SaaS deployment becomes a hostage situation within 18 months: the vendor raises prices, and the customer's choices are paying or rebuilding. Building on Claude Code, against the company's own CRM and call records, means the productivity gain is permanently the company's own.
The Academy — adapted for a sales organization
The four Track 3 graduates now form the company's "Sales Engineering for Agents" function — staffed inside RevOps, allocating ~40% of their time to maintaining and extending the agent stack.
The human dimension
"I told my board nine months ago that I would hit our ARR plan with the team I had. They thought I was being political. The plan is ahead of schedule and the team is the same size. The single most important thing we did was put Claude Code on every rep's desk and teach them to operate it like a craft tool."
Chief Revenue Officer
"I was burning out before this. I was sending 1,200 emails a month and getting 25 replies. I knew it was bad. I knew I was making it worse for the next person who emailed any of those prospects. Now I send 580 emails a month and book 24 meetings. The quality went up even as I sent less."
SDR · year 2 in role
"I was already doing depth-led prospecting on my best accounts before the engagement — that is why I was the top performer. The agents let me do it on every account. My pipeline coverage doubled. My close rate went up. The gap between me and the rest of the team has closed in a way I did not expect — and I am not threatened by that, because the company is winning."
Senior AE · top performer in the company
"I built EXEC-BRIEFER over a long weekend in the Academy. The CRO uses it every time he joins an executive call. The CEO uses it every time she joins. Three months in I realized I had built a tool that was being used by the executive team weekly."
RevOps Lead · Wave 2 builder
What this engagement model does NOT do
- It does not replace SDRs with autonomous outbound bots. Every touch ships under the SDR's name with the SDR's review.
- It does not replace AEs with chatbots. Every customer conversation is held by a human AE.
- It does not replace your CRM. The agents work with your CRM. The CRM remains the system of record.
- It does not magically fix bad pricing, weak product positioning, or fundamental ICP mismatch. If your product is wrong, no agent will save you.
The deliverable
The contracted commitment was nine months. We exited on schedule. The company's commercial team operates the agents themselves, has built four more we never touched, and the CRO is hiring his next two RevOps engineers against a job description that explicitly includes "agent operations" as a core responsibility.
"You did not sell me a sales tool. You sold me a sales team that operates at a different level than the team I had nine months ago."
Chief Revenue Officer · Engagement closeout
Six engagements. One method.
Want to be the seventh case study?
A 60-minute discovery call. No software pitch. We map your most painful workflow, scope a first agent, and tell you honestly whether this engagement model fits your organization.