
How Strix surfaced 435 high qualified leads for under €1
An autonomous lead engine that surfaced 435 high qualified, ICP-scored e-commerce leads from a single run, for a total infrastructure cost of around €3.
- Industry
- Digital Commerce
- Region
- Netherlands
- Company size
- 250
- Timeline
- 12 weeks
Strix is a Dutch digital-commerce agency that builds and scales e-commerce operations for ambitious brands. Its growth depends on reaching the right decision-makers at in-market e-commerce brands, and finding them by hand was the bottleneck. What's Next built an autonomous lead engine that surfaces real buying signals and delivers ranked, verified leads to sales every week.
The challenge
Agency growth runs on conversations. Not cold outreach blasted at a purchased list, but the right conversations, with decision-makers at e-commerce brands who are actually in-market right now. For Strix, a Dutch digital-commerce agency that builds and scales e-commerce operations for ambitious brands, finding those people reliably was the bottleneck.
The manual alternative is what most agencies default to: someone scrolling LinkedIn, exporting a list, spending hours enriching contacts and guessing whether a CMO at a brand they vaguely recognise actually fits the profile. It's slow, expensive, and impossible to scale without headcount.
The questions leadership couldn't answer:
Which brands in our market are actively shopping for a new commerce platform or agency right now? Who are the decision-makers we should be talking to this week, ranked by fit? How much is it costing us, in time and opportunity, to find leads by hand? And what would a steady, repeatable lead flow actually look like at our scale?
The consequence of staying on the manual path wasn't theoretical. Every hour a growth hire spent on prospecting was an hour not spent on pipeline. And the leads they found weren't necessarily the right ones, just the findable ones.

How we approached it
The brief wasn't "build a lead gen tool." It was: find a way to surface the people already signalling intent in the market, and deliver them to sales ready to act on.
Step 1: Map where buying signals actually live. Before writing a line of automation, we audited where genuine e-commerce intent shows up. Not job boards. Not generic company databases. The live signals: who's engaging with category content on LinkedIn, who's just stepped into a new commerce role, which brands are publicly signalling a platform migration, which companies just received funding. Seven distinct signal types, each pointing at a different stage of intent.
Step 2: Stress-test the LinkedIn engagement channel. The highest-signal source was also the newest idea: tapping the audience actively engaging with e-commerce content on LinkedIn, rather than cold prospecting. We ran a controlled test. 900 people captured from a single engagement pool. Then we applied the filter, enrich and verify sequence to that raw set and measured yield: how many survived each stage, and how many arrived as genuinely qualified leads at the end.
Step 3: Design for a single unified pipeline. Every signal source had to flow through the same enrichment, verification and scoring layer. Not seven separate tools. One engine. The reason: if adding a new signal source means rebuilding the pipeline, the system doesn't grow. If it means pointing one more input at the same engine, it compounds.
Step 4: Build the precision layer last, not first. Verification, confirming that a decision-maker genuinely belongs to the company we think they do, was the step most lead tools skip. It's also the step that determines whether sales trusts the list. We built it in as a required stage, not an optional one.
The discovery phase didn't just define what to build. It's why the build didn't generate a lead list Strix's sales team would quietly stop looking at.

What we built
What's Next built Strix an autonomous lead engine: seven live signal sources feeding one automated pipeline that captures buying signals, enriches them against company and contact data, verifies decision-maker identity, scores against Strix's ideal-customer profile, and delivers a ranked weekly digest. No manual prospecting. No per-seat tool bill that scales against you.
LinkedIn engagement intelligence is the engine's standout source. Instead of building a list and hoping it's in-market, we tap the pool of people already engaging with e-commerce content on LinkedIn, the ones reading, reacting and commenting on category posts right now. That engagement is the signal. We capture them, filter out competitors, internal staff and irrelevant profiles, then run the remainder through the pipeline. From a single engagement pool of roughly 900 people, the engine delivered 435 qualified leads, a conversion rate of around 51% from raw signal to scored, sales-ready contact.
Role change and platform intent signals round out the LinkedIn layer. A decision-maker stepping into a new commerce role has a budget conversation in their first 90 days. A brand publicly signalling a platform migration is already shopping. Both signals sit in public data, and the engine surfaces them before the competition does.
Website visitors, look-alike audiences, funding and M&A events, and manual imports complete the seven sources. Each one represents a different intent stage. Together, they mean Strix's pipeline doesn't depend on any single channel going quiet.
The cost picture is what makes the architecture genuinely defensible:
Approach | Cost per lead (est.) | Cost at 435 leads |
|---|---|---|
Manual prospecting (2 hrs at €75/hr loaded) | ~€150 per qualified conversation | ~€65,000 |
B2B SaaS lead tool (Apollo, Lusha, per contact) | €1.50 to €5.00 per contact | €650 to €2,175 |
What's Next owned stack (LLM + data APIs, pay-per-use) | ~€0.007 per lead | ~€3 per run |
The deeper point is ownership. SaaS lead tools charge per seat and per contact: the more you use them, the more you pay. The What's Next stack is a pipeline Strix owns outright. Increasing volume doesn't increase the per-unit cost. Running it twice a week instead of once doesn't trigger an upsell conversation.
The verification layer is the piece most lead tools skip. Before any lead reaches the weekly digest, the engine confirms that the decision-maker is genuinely associated with the company it thinks they belong to. Mismatched contacts, a common failure mode when enrichment scales, are removed automatically. Sales receive only contacts they can trust.
The weekly digest is the delivery format by design. Leads arrive scored and ranked. No dashboard to log into, no manual export, no interpretation required. The list lands, and the first call goes out.
435
The outcome
Strix's first live engagement run produced 435 qualified, scored e-commerce leads for a total infrastructure cost of roughly €3. Recognisable Dutch retail and e-commerce brands were in the set. The weekly operating cost to keep the engine running across all seven sources sits at around €1 per day.
The operational shift is straightforward. The time a growth hire previously spent on manual prospecting, sourcing, enriching and second-guessing fit, no longer exists as a task. That capacity is now in pipeline conversations, not list-building.
The strategic shift is what scales: every run makes the scoring model sharper. The engine doesn't just produce leads, it produces data on which signals convert and which don't. Over time the filter tightens, the conversion rate from raw signal to qualified lead improves, and the cost per qualified conversation drops further.
Seven signal sources are live. The architecture supports new sources without rebuilding the pipeline. When a new intent signal becomes available, a new data partnership or a new LinkedIn signal type, Strix adds it in and it flows through the same engine.
Total project timeline: 12 weeks. The discovery phase, where we mapped and stress-tested the signal sources before building, was what made the 51% signal-to-lead conversion rate achievable rather than accidental.
For us it's about quality, not volume. The engine surfaces the right customer profiles and takes the work of finding them off our hands. Our team spends its time on the right conversations, with the right people at the right moment.

Want results like this?
Let's talk about what AI can ship for your team.
- Free 30-min advice call
- No obligation, no pressure
- A specialist, not a sales rep


