Growceanu Customer.io implementation | Regulated EU crowdfunding
I advised on Growceanu's initial Customer.io implementation, then took over delivery. I audited the inherited workspace, rebuilt the flows, and launched the journeys. Growceanu is a regulated EU crowdfunding platform under ECSPR. Launch results: 1.5% email conversion from cold lead to KYC-verified client, and a >55% email opening rate.
For teams inheriting a half-built Customer.io account that is already sending under compliance review.
Email conversion from cold lead to KYC-verified client after the inherited-account rebuild.
Email opening rate at launch, reported with an Apple Mail Privacy Protection caveat.
Bounce investigation threshold I use as a hard-pause rule before the next send.
What the engagement covered.
Phase 1 advisory, then Phase 2 delivery on an inherited Customer.io account that was already sending.
Growceanu runs under Regulation (EU) 2020/1503 with Romanian and English cohorts. I advised until the first pre-launch campaigns were built, then owned the inherited-account audit, rebuilt the flows, and built the launch campaigns inside Customer.io.
This was remediation of an account that was already sending, with relationship objects for fundraising and compliance review on every template.
The stack already used Customer.io, including the classic drag-and-drop editor. People related to fundraising relationship objects. Sending ran from a subdomain that needed correct Apple Private Relay registration. I did not replatform from another ESP.
What was broken at the start.
The Customer.io workspace was inherited, already live, and failing in ways that did not show up in the UI.
Several onboarding branches still had messages set to Queue Draft, so profiles advanced while messages piled up in Drafts for months. KYC attributes existed in the schema but were empty on every profile. Object tokens could name the wrong company. Staging URLs had shipped in live emails. Locale segments were wrong, and list reputation risk was real on inherited cohorts.
Four failures hit at once: draft-only sending on live onboarding, staging URLs in nine live emails, locale matching only one full locale form, and conditions on unsynced attributes.
An old-database reminder later bounced at 9.5% (about 7 to 8% on the Romanian cohort), and thirty-three Apple Private Relay addresses bounced until the subdomain was registered correctly.
How the data model was rebuilt.
Person attributes and fundraising relationship attributes had to live in different places.
Timestamps that lived on the relationship returned nothing when queried as person fields. KYC fields were defined but empty because sync was broken. Language preference was unreliable, which made negative language segments unsafe for targeting.
Field categories I used (exact names withheld): a discrete investment-started event as the journey trigger after relationship-attribute triggers proved too late; a start timestamp on the fundraising relationship object; KYC attributes for targeting once sync worked.
Language preference was used only when populated. Company name on the relationship powered post-investment personalisation with trigger-scoped tokens instead of indexed object tokens.
How the rebuild ran, step-by-step.
Not a migration from another platform. Audit, fix data, rebuild journeys, then launch under a silent-failure checklist.
I inventoried every live flow, verified attribute location on live profiles, wrote findings with severity for sign-off, fixed data location and population, rebuilt journeys, ran QA, then launched. Relationship modelling and compliance review on every template dominated the timeline versus a greenfield event-only setup.
- Week 1AuditInherited account
Inventory live flows
I catalogued every live flow and each step's true sending behaviour, then verified attribute location and population on live profiles.
- +Find draft-only steps still on live branches
- +Spot staging or localhost URLs in templates
- +Confirm KYC and language fields actually sync
- Week 2Fix dataPerson vs relationship
Correct data location
I separated person attributes from fundraising relationship attributes and restored KYC sync so targeting conditions stopped matching everyone.
- +Move start timestamps onto the relationship object
- +Replace late relationship triggers with an investment-started event
- +Retire unsafe negative language segments
- Week 3RebuildJourneys + launch
Rebuild and QA
I rebuilt onboarding and post-investment journeys, then ran a silent-failure checklist before any launch send.
- +No draft-only steps on live flows
- +Tokens tested on multi-record profiles
- +Authentication checked from the exact sending domain
- Week 4LaunchGoverned sends
Ship with sign-off
I launched campaigns under two-person sign-off and a 3% bounce investigation rule, with hard-pause before the next flight if the threshold trips.
- +Compliance wording checked on every template
- +Apple Private Relay registration on the sending subdomain
- +Launch measured as cold lead to KYC-verified client
Before versus after, side by side.
Inherited half-built sending versus a governed rebuild inside the same Customer.io workspace.
A full ESP swap would have burned months without fixing Queue Draft behaviour, empty KYC sync, or compliance sign-off gaps that were already live in 2026.
| Before rebuild | After rebuild | Greenfield ESP swap | |
|---|---|---|---|
| Live sending behaviourDraft-only steps vs real delivery on inherited flows. | ✕ | ✓ | ~ |
| KYC attributes populatedSchema fields that actually sync to profiles. | ✕ | ✓ | ~ |
| Event trigger for investment startExplicit event instead of late relationship attributes. | ✕ | ✓ | ✓ |
| Silent-failure QA checklistStaging URLs, locale forms, token edge cases. | ✕ | ✓ | ~ |
| Two-person compliance sign-offStops corrected copy from staying unpublished. | ✕ | ✓ | ~ |
What changed after launch.
1.5% email conversion from cold lead to KYC-verified client, and a >55% email opening rate at launch.
I also measured more than 10% click-to-open. The opening rate is reported with an Apple Mail Privacy Protection caveat, so it is not used as a deliverability decision input.
I replaced late relationship-attribute triggers with an explicit investment-started event and introduced two-person sign-off after the compliance wording failure.
Ownership of flow and launch delivery moved to me after the advisory phase. QA and governance were installed, and launch journeys shipped. This page does not claim that every audit finding was fully remediated, only the changes stated here.
Related reading: the CRM implementation playbook and IP warmup planner.
Questions people ask.
Inherited accounts, bounce thresholds, open-rate caveats, and compliance constraints on this ECSPR engagement.
/01Was this a migration from another ESP?+
/02What broke silently before the rebuild?+
/03How do you treat open rate after Apple Mail Privacy Protection?+
/04What bounce threshold do you use?+
/05What compliance constraints shaped the copy?+
/06What did two-person sign-off fix?+
Need Customer.io implementation like this?
Most engagements start with a short call to review your workspace and constraints.
Bring bounce logs, compliance rules, and which flows are already live.