hours reclaimed per week
pipeline from AI-surfaced closes
of all leads replied

The Problem
Backstage's commercial model only works inside a narrow window.
The pitch, "talk to us before you hire direct, we'll build the team in Moldova," lands when a company has publicly signaled it's hiring developers but hasn't started interviewing yet. That window closes fast. A CTO LinkedIn post about scaling the team sits at the top of the feed for a few hours, then disappears. A new senior role on TechMap or Adzuna gets discovered by the candidate market within a day. Miss the moment and the deal goes to whichever competitor caught it.
The constraint was structural. Two SDRs were manually researching the entire Benelux, DACH, and UK market. They could read what two people can read in a day, no more. The signals that mattered most were ephemeral, and the SDRs could either prospect or close, never both, which meant founders Sander and Emmo were being pulled into daily sales firefighting to keep the close work moving.
The shape of the problem was clear. Same headcount, more leverage, or hire a larger BDR team and accept the cost.

The Approach
The instinct in a situation like Backstage's is to hire.
Cap two SDRs against an entire region, hire two more. That gets you four people doing the same manual work, with the same attention ceiling and the same blindness to ephemeral signals. Doubling a constraint doesn't remove it.
The other instinct is to buy. A SaaS outbound tool, an off-the-shelf intent platform, an AI SDR. Those solve the volume problem but miss the harder one. The signals that actually convert for Backstage, a CTO posting about hiring, a new role appearing on a job board, are public for hours, not days, and no off-the-shelf vendor monitors them with the freshness or filtering precision the pitch requires. Generic outbound at scale would have produced more email volume against worse-timed accounts.
So we made the strategic call to flip the funnel. Instead of SDRs hunting for who might be hiring, the system would catch hiring signals the moment they went public and route them to the SDRs fully enriched. Same SDRs, completely different leverage.
The build was sequenced over three phases for a specific reason. Phase one had to prove the signals were real and harvestable before we built scoring on top of them. Phase two had to prove the scoring was directionally right before we wired it into outbound. Phase three had to put the whole engine in the hands of non-engineering operators, because a system the founders can't run themselves is a system that fails the moment we stop being in the room. Each phase de-risked the next.
The bet underneath all of it: the difference between catching a hiring signal in hours and catching it in days is the entire commercial advantage. Build the engine that compresses that window, and the same two-person sales team becomes structurally more leveraged than a competitor running a larger BDR motion manually.

The Solution
We built a fully autonomous lead-generation and outreach system for Backstage IT.
Phase 1, Signal sourcing
We built n8n scraping pipelines that hit LinkedIn, TechMap, and Adzuna in real time. Filters strip out recruitment agencies and FAANG-tier companies where the pitch doesn't apply. Deduplication and freshness policies make sure no signal gets prospected twice or after it's gone cold.
Apollo.io enrichment runs via Apify to attach firmographics, decision-maker profiles, and tech-stack detection to every account. The whole pipeline writes to a multi-tenant Supabase warehouse with companies and people tables, credentials isolated per-client via GCP Secret Manager.
Phase 2, Live monitoring, scoring, and outbound
A Python feed monitor scans the team's authenticated LinkedIn feed against 14 curated Dutch and English buying signals. "Hiring developers," "team groeit," "backlog groeit," funding announcements, leadership changes.
Every detected prospect gets a composite PriorityScore weighted 40% buying signal, 30% engagement, 30% relationship fit. Qualified accounts flow into Lemlist for personalized email outbound. Signal-driven LinkedIn outreach runs in parallel.
Phase 3, Operator handoff
A local GTD-style review interface lets SDRs triage signals without engineering involvement. Mark interesting, snooze for one day, three days, a week, or custom, wake when ready, add notes. 95% test coverage on the scoring and review logic.
As of late May 2026, the system runs entirely in the hands of Backstage's non-engineering operators, with no dependency on us to keep it shipping.

The Results
The same two SDRs now cover the addressable market in a way that two SDRs running manually never could.
The top of the funnel is full of accounts that have publicly signaled hiring intent in the last 72 hours, with decision-maker profiles and tech-stack data already attached. Reply rates climbed without personalization falling off, because personalization runs against real, current data instead of recycled templates.
The bigger shift is at the executive level. Sander and Emmo no longer firefight every day to keep the sales motion moving. The engine surfaces the highest-fit opportunities, the SDRs work them up, and the founders only step in for the top-tier closes the system identifies. The same two-person sales team now produces 40% of Backstage's total pipeline through AI-surfaced closes, a revenue stream that did not exist before the engagement.
Metric | Before | After | Impact |
|---|---|---|---|
Pipeline through AI-surfaced closes | 0% | 40% | New revenue stream that didn't exist before the engagement |
Time reclaimed per SDR per week | — | 16 hours (2 full days) | Reinvested in close work, proposals, customer relationships |
LinkedIn touches / SDR / day | Manual, attention-capped | 100, signal-driven | Real-time coverage of the addressable market |
Email capacity / SDR / day | 30, manual | 200 unlocked | 6.7× ceiling lift; capacity is now ahead of demand |
LinkedIn open rate | Industry avg 40-50% | 63.7% | Personalization quality survived the volume scaling |
LinkedIn reply rate | Industry avg 3.4% | 25% | ~7× the industry baseline |
Inbox delivery rate | Industry standard | 98%+ | Deliverability infrastructure held under scale |
Win / close rate | Baseline | Stable | Quality preserved — honest: no inflated "win rate jumped" claim |
Founder sales time | Firefighting daily | AI-surfaced top deals only | Executive-level time arbitrage; Sander + Emmo close the highest-fit opportunities the engine surfaces, not everything |
Key Results
hours reclaimed per week
pipeline from AI-surfaced closes
of all leads replied



