Pulak Singh

Two Lane

One thumbnail template for a road-trip channel. The sign, the headline and the yellow bar never move, so only the photograph and the words change.

Type
YouTube thumbnail system
Role
Art direction
Deliverables
Channel mark, thumbnail template, four episode thumbnails
Tools
Photoshop
Typefaces
Acumin Pro
Year
2026

Concept project — art directed by Pulak

Episode 10 thumbnail: The Ghats after dark, with light trails on a hill road at night

The brief

Two Lane is a concept travel channel about India’s back roads, ridden on a motorcycle. The template is used here for four episodes: Spiti, NH 44, the Konkan coast and a night ride in the Ghats.

Episode 7 thumbnail: The cold desert, Spiti Valley
Episode 8 thumbnail: 1,000 km in one day, NH 44
Episode 9 thumbnail: Slow road to Goa, Konkan coast

One frame

Every thumbnail has the same parts in the same places: a sign top left, a two-line headline with one line on yellow, and the episode number bottom left. A channel is built on repetition. Viewers learn the frame and spot the next episode among other channels.

Yellow is the colour of road markings, which suits a channel named after a road, and it stands out on every photograph in the set: sky, sea, rock and night.

The template, with a sign top left, a ten-character headline and red safe zones in the right-hand corners

Built for small screens

Most thumbnails are seen small. Two lines of up to ten characters keep the headline 16 px tall when the thumbnail is only 168 px wide. The right-hand corners stay empty for YouTube’s duration badge and hover buttons.

The episode 8 thumbnail at 360 pixels wide
360 px
The episode 8 thumbnail at 246 pixels wide
246 px
The episode 8 thumbnail at 168 pixels wide
168 px

Two keys

On dark photographs the type is white over a soft black gradient from the left. On bright photographs it turns to ink and the gradient comes off.

The dark key and the bright key side by side

Production

Built in Photoshop at 2560 × 1440 and exported at 1280 × 720, the size YouTube recommends. Photographs are placed as smart objects, so a crop can change without losing quality, and the safe zones sit in a hidden layer group in every file.

The photographs are concept images, generated from a written brief for each episode: the place, the light and where the rider sits, so the type has room.