Case Study · CRM & Engagement

CRM now drives 17% of Joyn's monthly video views — with marketing teams shipping campaigns code-free.

Joyn's 20M-MAU streaming service spanned 9+ platforms, but its CRM team was bottlenecked by engineering: profiles were fragmented across anonymous and registered users, data sat in silos, and every campaign needed a dev sprint. I architected a Braze engagement engine that turned the data stack into a plug-and-play CRM, making marketing managers fully self-sufficient.

Role Led end-to-end architecture & rollout — identity resolution, event tracking, warehouse sync, and reverse-targeting

20M MAUs 9+ platforms Braze engagement engine 17% of monthly video views
Streaming CRM Identity Resolution Marketing Automation

Reaching 20M viewers with the right message at the right moment

Engagement is everything for a streaming service this size — but the right message can only land if every device, profile, and viewing signal feeds one connected CRM.

SCALE

One message across 9+ device types

20M monthly viewers stream across mobile, desktop, and smart TVs from Apple, Android, LG, and Samsung — every one needs consistent, timely outreach.

9+ platforms = one CRM must cover them all
AUTONOMY

Marketing was blocked on engineering

Every campaign required developer work, so the CRM team couldn't move at the pace the business needed to retain viewers.

Dev dependency = campaigns stuck behind sprints
RETENTION

Keeping viewers coming back

With 60% of users registering each month, turning casual browsers into returning, registered viewers was the core retention goal.

60% register monthly = a habit waiting to be built

Fragmented identity, siloed data, manual execution

Three structural problems stood between Joyn and a CRM its marketers could run on their own.

IDENTITY

Anonymous and registered profiles were split

A viewer browsing anonymously and the same person after sign-up looked like two different users, breaking targeting and personalization.

Split profiles = data lost at registration
SILOS

Viewing signals trapped in the warehouse

Offline video views and watchtime — the richest signals for personalization — lived in the data warehouse, disconnected from the CRM.

Siloed signals = no behavioral targeting
MANUAL WORK

Heavy lifting on every launch

Without orchestration, even routine campaigns required manual setup and engineering hand-offs that slowed everything down.

Manual execution = no room to scale

A single source of truth, wired into a plug-and-play CRM

Drawing on prior enterprise CRM work with Eloqua, I built Braze into a layered engagement engine — each step unlocking richer targeting.

1
Unified identity resolution anonymous → registered

Established one seamless user identity across CRM and analytics, handling the anonymous-to-registered transition so no profile data was lost at sign-up.Braze · identity sync

2
Real-time event tracking

Built the event-tracking architecture that triggers instant in-app messages and push notifications off live viewer behavior.Event tracking · push & in-app

3
Warehouse sync for offline signals

Piped offline video views and watchtime from the data warehouse into Braze daily, enriching every profile for deeper personalization.Warehouse sync · watchtime

4
Recommendations & reverse-targeting

Enabled algorithmic content recommendations plus "reverse recommendations" — finding the ideal audience for a specific title and launching campaigns to them programmatically.Recommendations · reverse-targeting

Autonomy for teams, measurable growth for the business

17%
of monthly video views
attributed to CRM operations
100%
marketing self-sufficiency
No code, no dev sprints
20M
MAUs reached
60% registered ecosystem
9+
platforms unified
Mobile, desktop, smart TV

From an engineering bottleneck to a self-serve growth engine

Marketing managers ship cross-channel campaigns solo

CRM managers now launch sophisticated email, push, and in-app campaigns independently — no code, no waiting on engineering sprints — across all 9+ platforms.

Reverse-targeting for any title

For any piece of content, the system identifies the perfect audience and launches a targeted campaign to them automatically.

Profiles enriched daily

Offline watchtime and video-view signals flow into the CRM every day, keeping personalization grounded in real viewing behavior.

Let's talk

Tackling something like this?

Tell me your biggest data or AI bottleneck and I'll send you how I'd approach it — the same way I worked this problem. No résumé dump, no pitch. Just a useful conversation.

Start the conversation →