Keep assessments original for every cohort, without lowering the bar

There's a quiet compromise that creeps into a lot of courses. The first year, you write a genuinely good case study, a solid exam, a project brief you're proud of. The second year, you reuse it and maybe you tweak the numbers. By the fourth semester, the same prompt has been handed in, photographed, shared and, now, fed into a chatbot more times than you'd like to think about.

You know it should change. But writing a fresh, equally rigorous version for every cohort is a serious amount of work, so the old one quietly stays.

The two bad options

When the pressure to refresh finally hits, most of us pick between two unsatisfying routes.

Reuse it and hope. Cheap on time, expensive on integrity. Past answers circulate, and a prompt that's findable online is a prompt that's already half-solved before the student starts. The assessment still looks rigorous. It just no longer measures much.

Rewrite it from scratch. Honest, but slow. And the quality drops, because you're rebuilding under time pressure at the end of term, not at your best. Do it across three courses and it simply doesn't happen.

What's missing is a third option: a fresh version, at the same standard, without the hours.

What actually leaked

It's worth being precise about what went stale, because it isn't the assessment.

Your learning outcomes didn't leak. Your standard didn't leak. The skill you're measuring — build an argument, read a balance sheet, defend a design decision under constraints — is the same skill it was three cohorts ago, and it's still the right thing to test.

What leaked is the surface: this company, these figures, this scenario, this exact wording. That's the part students can pass forward. And that's the only part you need to replace.

Once you separate those two layers, the job stops looking impossible. You're not rewriting an assessment every year. You're re-skinning one.

Keep the structure, vary the particulars

Here's the split, concretely.

Keep these — they carry the rigour:

• The outcome the task maps to, and the level you're testing it at.
• The shape of the decision the student has to make. A genuine trade-off with no clean answer

is what makes a case study hard; that shape is reusable forever.

• The evidence you ask for. A recommendation with justification, a worked model, a defence of a

method choice.

• The criteria you'll mark against. Same rubric, cohort after cohort. That's a feature: it's how

you compare this year's cohort to last year's honestly.

Vary these — they're what circulates:

• The context: sector, organisation, place, time period.
• The data: figures, sample, dataset, source documents.
• The angle: same trade-off approached from the other side of the table.
• The constraint: change what's scarce — budget, staff, time, information — and the reasoning

path changes with it.

A worked example. Your outcome is evaluate a pricing decision under incomplete information. Last year it was a mid-size hotel deciding on shoulder-season rates. This year it's a clinic pricing a new service line, with a data gap in the same place. Same decision, same depth, same rubric. Nothing anyone kept from last year helps.

"Isn't that what a question bank is for?"

Partly, and question banks are genuinely useful for recall. But they solve the small version of this problem. A bank gives you a hundred variants of a question that has one right answer, which works for definitions, calculations and multiple choice.

It doesn't give you a new case study. It doesn't give you a fresh project brief, or a scenario with a defensible decision inside it. The assessments that measure the most are exactly the ones a bank can't refill — and they're the ones sitting unchanged in your course folder.

The standard is the thing to protect

There's one honest risk in refreshing every cohort, and it's worth naming: drift. If this year's version is quietly easier than last year's, you haven't protected integrity. You've just moved the problem somewhere harder to see.

Two habits keep that from happening. Mark against the same rubric every cohort, so the standard lives in a document rather than in your memory of last June. And check the new version against the same outcome, at the same level — not against how it feels to read. Fresh and rigorous aren't opposites, but they don't stay aligned on their own.

If your rubric isn't stable enough to carry that job, that's the piece to fix first — before the assessment. A rubric specific enough to hold a standard is what makes everything above safe to do.

Where Quindaria fits

This is exactly the work Quindaria takes off your plate. Give it your course — only the description, the learning outcomes and the topics — and it generates assessments and activities grounded in that, at the level you set.

Run it again and you get a genuinely different version on the same topic: a new scenario, new data, the same standard. Not a watered-down shortcut, and not the generic thing the internet hands everyone. Built around your course, ready to use, and different every cohort.

Your course. Your style. Always aligned.

Generate two versions of the same assessment for one of your courses, and see how little they have in common.

Get early access →