Back to work
// Content production

Video production workflow system.

For Influentia's Growth Lab, we built a Notion-based production system: update a video once, and every phase calculates its own status automatically, keeping calendars, task lists and timelines in sync.

Source of truth

Single

Manual status chasing

Gone

Non-applicable phases

Self-clearing

The situation before

Nobody could say what was blocked, who was overloaded, or what a video actually needed next without asking around, because production status lived in places the team had to update by hand.

Videos, people and production steps were tracked separately. Moving a piece of content meant updating several places, or hoping someone else would.

What was done, in progress or next was reconstructed in meetings instead of read from the system.

What we built

The system, concretely.

01

Linked production model

Four related databases, clients, team, videos and tasks, keep every piece of content in one connected structure.

02

Video as the single edit surface

The team changes global stage and production options on the video. Task rows are generated and driven from that, nobody maintains phase status by hand.

03

Automatic phase status

Each task computes completed / in progress / pending from the video's stage, including skips when a phase does not apply (no script, no brand, already optimized).

04

Work views the team actually uses

Per-client work tables, a live calendar, an auto-sorted task board, a personal task list and a Gantt timeline, all reading the same calculated state.

Conceptual architecture

How the pieces connect.

Video updated
Stage mapped to number
Tasks auto-status
Calendars & task views
First version liveWeeks, not months
The result

What changed.

  • One place to move a video, every phase view updates from that change.
  • Parallel production steps show as in progress together when they share a stage.
  • Non-applicable phases clear themselves instead of clogging the board.
  • Standard and studio flows run on the same model with different stage maps.
What happened next

The system kept evolving.

The first release locked the linked databases and automatic status. Later work layered client-specific flow rules and a companion daily notification path on top of the same model.

Platform limits that blocked native automation were worked around with precomputed stage numbers and a thin external layer, without asking the team to edit tasks directly.

Migrating off the old tracker followed the same discipline as everything else: a simulation pass, a diff report against what already existed, and a staged cutover, work already in progress first, then the next quarter's calendar, before touching older history.

See how the iteration model works →

What we delivered

Everything needed to keep running without us.

Linked Notion data modelAutomatic task status enginePer-client work calendarTask list & Gantt viewsTask generation templatesRole-based permissions for internal and external collaborators
More work

Other projects through the system.

Like what you see?

Let's build your version of this.

Tell us what you're trying to build or improve, and we'll show you how the system applies to your project.

Let's talk?