Designing Assessments That Resist Cheating
July 6, 2026 · 6 min read
Change the task before you buy the camera
The most cheating-resistant assessment isn't the one with the most surveillance. It's the one where copying an answer doesn't help, because the task asks for something a student can only produce by doing the work. Redesign the question first. Reach for monitoring last, if at all.
I've argued the surveillance side of this elsewhere, that proctoring tools feel like control right up until they're bypassed, and the platforms that get assessment right spend the budget on design instead. This post is the other half: what "design instead" actually means when you sit down to write the assessment. Because "make it cheat-proof" is not a feature you install. It's a set of choices about the task.
What makes a task resist copying
- Ask for application, not recall. A question with one lookup-able answer is a question worth cheating on. "Explain why this approach fails for this specific dataset" has no answer key to paste.
- Make it personal to the work. Tie the task to something the student produced earlier, their project, their data, their earlier choice. You can't copy an answer that depends on your own prior work.
- Randomize from a real bank, not a token shuffle. A question bank helps only when the variants are genuinely different problems, not the same question with the numbers swapped. Swapped numbers are a speed bump; different problems are a wall.
- Assess the process, not only the product. A short reflection on how they got there, or a staged submission, is far harder to fake than a final answer.
- Use time deliberately. A tight timer on a recall test just rewards whoever has the answer key open. A generous window on an applied task rewards the person who actually engages.
| Instinct | What it really buys | Better move |
|---|---|---|
| Add a proctoring tool | Deterrence until it's bypassed | Redesign the task to be copy-proof |
| Shuffle question order | Minutes of friction | A bank of genuinely different problems |
| Shorten the timer | Pressure on everyone, honest or not | Apply time to applied work, not recall |
| Lock the browser | An arms race with the student | Assess process alongside product |
The honest limit
You can't make cheating impossible, and I won't pretend a well-designed task does. A determined student with enough time can still get help. What good design does is change the economics: it makes cheating harder than just doing the work, and it stops punishing the honest majority for the actions of a few. That's a better goal than "impossible," and it's actually reachable.
For the case against leaning on surveillance, read what actually works on online exams. Assessment integrity also depends on knowing who's really taking the exam, which is where the student-data security checklist comes in. When assessment is central to your product, custom LMS development builds the question banks, staged submissions, and process capture around how you actually teach, rather than bolting a camera onto a generic quiz.

Written by Choaib Mouhrach
Founder & Senior Software Engineer
I design and build custom learning platforms for organizations with complex training and certification workflows. Instead of stitching together plugins and third-party tools, I create systems tailored to how each business operates, reducing administrative overhead while improving the learner experience.
Your learning product deserves its own platform.
If you want to deliver a learning experience built around your product, your learners, and your goals, you are in the right place. We build platforms that give you the control and flexibility to grow without limits.