Extraneous Load: The Tax Your Processes Charge Before Work Begins
Sweller's cognitive load theory separates the difficulty of the work from the difficulty your systems add to it. Most process improvement targets the wrong one.
Separating the work from the friction
A credit analyst assessing a marginal case is doing genuinely hard cognitive work. That is intrinsic load, and it is what you employ them for. Switching between four systems to assemble the file, re-keying a reference number, remembering which of two dashboards shows the current figure — that is extraneous load, and it is pure waste.
Both feel like "the job" from inside. Staff rarely distinguish them when asked, which is why interview-based process review systematically under-detects the second category.
Why extraneous load is invisible in the numbers
Extraneous load does not appear as a line item. It appears as slightly longer handling times, slightly higher error rates late in shifts, and experienced people who are inexplicably much faster than new ones — because they have absorbed the friction into habit rather than because the process is sound.
It also caps capability. Working memory spent on navigation is unavailable for judgement. Reduce extraneous load and you have not merely saved seconds; you have released capacity into the part of the task that actually requires a person.
Where to cut first
Target duplicate data entry, cross-system switching, information positioned away from the decision it informs, and any figure a person must hold in mind while navigating to another screen.
The measurement that matters is not task completion time but error rate under load — late shift, high volume, after interruption. That is where extraneous load converts into cost, and it is where the return on removing it shows up first.
Key takeaways
- Distinguish inherent task difficulty from difficulty your systems impose.
- Extraneous load consumes the same working memory judgement requires.
- Staff cannot reliably separate the two, so observation beats interviews.
- Measure error rate under peak load, not average completion time.
Continue reading
Tell us about the decision you're trying to get right.
We work with organisations on the cognitive layer beneath their decisions, systems and people processes.
Book a consultation