An ABR ladder is the set of pre-encoded versions of the same video at different bitrates and resolutions, designed so a streaming player can pick whichever rung fits the viewer's current network speed. A typical 1080p ladder might have 6 rungs: 240p at 400 kbps, 360p at 800 kbps, 480p at 1.5 Mbps, 720p at 3 Mbps, 1080p at 5 Mbps, 1080p at 8 Mbps. Every viewer downloads small segments (2–6 seconds) and the player switches up or down the ladder smoothly as the network changes.
The ladder is the single most important business decision in adaptive streaming. Too few rungs and many viewers get either buffering or unnecessarily-low quality. Too many rungs and your encoding and storage costs explode without proportional viewer benefit. Apple originally published a "recommended" ladder of fixed bitrates for HLS in 2010, and that fixed ladder became the industry default for years. Then Netflix in 2015 showed that per-title encoding — analysing each video's complexity and customising its ladder — saves ~20 % bitrate across the catalogue at the same perceived quality. That insight redefined the practice.
For a product team in 2026, the modern ABR ladder is never fixed. Three optimisation layers are now standard. Per-title encoding picks ladder rungs per video based on content complexity (a static lecture needs less than a snow-storm). Content-aware encoding (CAE) goes further with scene-level analysis. Context-aware encoding picks different ladders for different devices (phones don't need the 4K rungs). The pragmatic recipe: 5–8 rungs covering 240p to your top resolution, generated per-title via an automated optimisation tool (Mux's per-title, Bitmovin's CAE, AWS MediaConvert's QVBR, Netflix's open-sourced approach).
