← Back to selected work

Nordic spa & sauna manufacturer (US)

Multi-vertical outbound system that grew with the business

Built a multi-vertical outbound pipeline for a US specialty manufacturer — 9,000+ leads processed, 18 campaigns across 15 verticals, evolved from volume outbound into signal-based prospecting and conference re-engagement layered on top. Built-to-handoff — client's internal team now operates the system.

9,000+

leads processed

15

verticals covered

18

live campaigns

12.24%

reply rate on top vertical

Clay Instantly Hunter.io Firecrawl Google Maps API Building Connected

Context

A US specialty manufacturer in the wellness equipment space (saunas, cold-plunge, steam) needed to expand beyond its core hotel-leads vertical without building an in-house outbound team. The constraint: a small commercial team that couldn’t absorb the operational load of multi-vertical prospecting, deliverability management, and per-vertical copy iteration.

Phase 1 — Volume foundation (hotels)

  • 9,033 leads processed → 4,329 unique → 2,268 qualified for outreach.
  • Five segmented Instantly campaigns by contact type (Architect, GC/CM, Developer, Manager, Primary).
  • Email deduplication + work-email waterfall enrichment recovered 48 additional contacts.
  • Four sending accounts across four domains, 100% health score, 25-35 sends/day per account.
  • Full deliverability config — warmup ramp, open tracking off (reply rate is the only meaningful signal), stop-on-reply, A/B/C subject line testing.

Automated weekly performance reporting delivered to the client team every Monday: per-campaign, per-step, and per-subject-line breakdowns. Built with Python + Resend + GitHub Actions, no human in the loop.

Phase 2 — Pipeline expansion (8 new verticals)

The same architecture extended outward. Different sourcing strategy per segment, because no single source covers wellness centers, country clubs, athletic recovery facilities, hotel spas, and active-build GCs equally well.

VerticalSourceQualified
Wellness CentersGoogle Maps API104
Nordic SpasGoogle Maps API89
Building Connected ArchitectsBuilding Connected364
Architecture FirmsFirecrawl (Architizer)271
Athletic / Sports RecoveryGoogle Maps API143
Country ClubsClay Find Companies780
Hotel SpasGoogle Maps API113

Twelve volume campaigns live across all verticals, full 5-touch sequences. LLM-powered “best contact” selection block in Clay (JSON output mapped to columns). Two-tier Hunter waterfall enrichment (executive/management first, general fallback). AI-treated company-name field for email personalization.

Standout signal: the Nordic Spa segment hit a 12.24% reply rate — order-of-magnitude above industry baseline, indicating real product-market resonance worth doubling down on. A copy rewrite test on Wellness Centers held volume constant and lifted reply rate 2.5% → 5.0% — copy framing accounted for the lift, not list quality or sequence length.

Phase 3 — Signal-based outbound layered on top

Volume gets you coverage. Signals get you conversion. Phase 3 layered intent signals on top of the existing volume engine without disrupting it.

  • Signal sources: Clay Find Jobs (hiring signals) + Claygent (news signals).
  • Flow: Find Companies → company enrichment → AI ICP scoring → Find People → contact scoring → signal enrichment → signal_context + subject A/B → email waterfall → push to Instantly.
  • Personalization: opening sentence + signal-explicit subject A + curiosity-led subject B per contact, generated per-row in Clay via AI block.
  • Sequence: 4-step over 18-21 days (vs. 5-step for volume campaigns) — shorter because the signal does the heavy lifting.
  • Dedicated Instantly campaigns kept separate from volume so signal-vs-volume metrics stay clean.

Two pilots running at time of writing: mid-size hotel chains (65 leads → 61 with confirmed email), and fitness/recovery networks (56 → 44).

Phase 4 — Conference re-engagement

CRM-sourced contacts from past industry conferences re-segmented by AI (country club, golf resort, skip). Three- and four-step sequences with AI-generated personalization. Three campaigns activated across PGA 2026 (country club + golf resort) and HFA 2026 (athletic club).

Multi-channel validation — three architectural choices, each proven

Positive replies landed across three distinct outbound motions, each validating a different architectural choice. The motions run in dedicated Instantly campaigns so per-motion baselines stay clean — preventing the common mistake of averaging signal-based performance into volume noise.

  • Signal-based outbound (Phase 3): first deal converted in client CRM — validates the signal-based vs volume metric isolation. A second mid-size hotel proposal followed within weeks.
  • Conference re-engagement (Phase 4): first meeting booked from the HFA 2026 athletic-club campaign within the first week of activation — validates the AI segmentation block + 3-step conference sequence.
  • Volume outbound, architect channel (Phase 1): first positive reply from the architect campaign landed a standing offer to spec the client on future projects — validates the Architect > GC/CM > Developer > Manager priority routing as the core motion that lands the most architecturally valuable outcome a manufacturer can win from an architect.

Why this engagement matters

Three things make this case different from a typical outbound build:

  1. Multi-source by design. Google Maps for local operators, Clay Find Companies for established institutions, Firecrawl for directory scraping, Building Connected for active-build leads. Each source chosen for what it does best — not “one tool to rule them all.”
  2. Contact priority routing. Architect > GC/CM > Developer > Manager > Primary, with hook-based outreach angles per type. AIA CEU course as the architect hook. Functional relevance scoring over seniority for architect selection (Project Architect beats Principal when the work is project-stage).
  3. The system grows. Volume → signal-based → conference re-engagement. Each phase compounds on the previous one. Same underlying infrastructure, new conversion vectors stacked on top.

Built-to-handoff outcome. The engagement matured into client-internal operation — the client’s team now runs the system in-house, with the documented next-phase architecture (CRM integration, semantic signal monitoring, queued ICPs) on their roadmap. The result is a pipeline the client owns, not an outbound agency on monthly retainer.

Let's talk

Let's build your revenue system

I'm available for GTM engineering and RevOps projects. Tell me where revenue is leaking and I'll tell you what I'd build.