Case 10 · Content Thumbnail Automation

Repeatable design.
Repeatable production.

Problem
Every week, four thumbnails in a fixed format — and their YouTube schedules — were made by hand.
Decision
Separated reading from typesetting, and reduced human review to a single point: the title.
Verified outcome
In live operation it produced four thumbnails and four YouTube schedules, and both first-run incidents became prevention rules.
Title layout from an automatically generated thumbnail
Actual generated output · title-layout excerpt
My role
Problem definition & interviews · decision rules · template & review criteria · operation · AI collaboration
Implementation
Apple Vision OCR · local typesetting (macOS) · YouTube Data API · Python
Input
2 photosExtract the required production information
Output
4 thumbnailsReproduce a recurring template
Operation
4 schedulesmatched human-made schedules on 19 fields
DESIGN DECISIONS

Flexible interpretation.
Consistent layout.

I first surveyed 100 finished thumbnails from 26 weeks. With effectively no design choices to make, the problem was redefined from “design generation” to “reading and typesetting automation”.

01

Extract the information

Read six required information fields from source photos and prepare production inputs.

02

Apply measured templates

Use established title positions, spacing and template criteria to generate the thumbnails.

03

Compare the result

Inspect pixel-level differences between the source and generated layout, then refine the output. Local typesetting instead of an external design service means no accounts, monthly fees or usage caps.

04

One check over an unattended run

Instead of a Saturday-night automatic run, a person confirms only the title and line breaks; thumbnails and YouTube schedules then run in one step. If any thumbnail fails, nothing is scheduled.

FIRST RUN

First-run incidents,
next week’s rules.

In a rehearsal with a real bulletin, the four schedules matched human-made ones on 19 fields. The first live run still surfaced two problems the tests hadn’t.

INCIDENT 01

Executables offloaded to the cloud

Cloud sync had evicted helper tools and fonts from the device within a week, so the run stopped. Every run now downloads missing files first.

INCIDENT 02

A clipped title on a public schedule

A long title entered on one line was found clipped on an already-scheduled thumbnail. With approval, all four were regenerated and replaced; titles now scale to fit, and below 75% generation refuses to run.

RULE

Ask once more at input

Keeping 11 or more non-space characters on one line now prompts for a line break again — closing the root cause: the renderer never checked for overflow.

Original · manualGeneratedOriginal · manualGenerated

The design principle

Recurring production criteria moved into code and review rules, leaving one point for human judgment. Failures found in operation became rules the next run enforces on its own.

Evidence & scope

Based on existing automation records and source/generated images. The application was church-bulletin content; public images show only the title-layout area. Recorded position differences were 14–22px for type A and 6–14px for type B, rather than exact replication. The 19-field match refers to test schedules from a real-bulletin rehearsal (since deleted), with visibility and title prefix intentionally different. Operation records cover the first live run only; no success rate for later runs is claimed. Church name, sermon titles and speaker details are not published.