Lostness metric

Description

A navigation score from 0 upwards combining revisits and unnecessary pages, computed from distinct pages visited, total visits and the minimum pages needed for a task.

For each participant and task, count N, the number of different pages visited; S, the total pages visited including revisits; and R, the minimum number of pages needed. Lostness L = sqrt[(N/S - 1)^2 + (R/N - 1)^2]. The first term grows with revisits, the second with visits to pages that were not needed; a perfect path scores 0. Smith observed that participants scoring under 0.4 showed no signs of being lost, while those above 0.5 clearly appeared lost. The metric is useful in navigation and information architecture studies alongside task success, because it captures how efficiently people found their way and not just whether they arrived. It needs a known optimal path and logged page sequences from task-based sessions.

Attributes

Loss type
Latent
Stage
Evaluation
Effort
Days

Steps

  1. Define each task's optimal path and set R, the minimum number of pages needed.
  2. Run task-based sessions and log every page visit per participant per task, using a tool, logger or video.
  3. Count N (distinct pages) and S (total visits including revisits).
  4. Compute L = sqrt[(N/S - 1)^2 + (R/N - 1)^2] for each participant and task.
  5. Classify scores below 0.4 as not lost and above 0.5 as lost; report mean L and the share lost per task or design.
  6. Compare designs or re-measure after changes, alongside task success and time.

Sources

  • Measuring the User Experience (Albert & Tullis, 2022)
  • Validating Product Ideas (Sharon, 2016)

Also known as

  • Lostness
  • Lostness score
  • Smith's lostness measure
  • Navigation lostness L

Categories

  • Content
  • Information Architecture & Content