Skip to main content
Your progress
0 of 5 lessons complete0%
T4 / M4.6 / Engineer TIER / ~10 min

Forecasting cloud spend: module quiz

M4.6 module quiz

Ten questions. 80% to pass (8 of 10). Open book, unlimited retakes.

Answers are collapsed under each question. Answer first, then check.


Q1

A top-down forecast of $1.351M grown by 12% is:

A. $1.550M
B. $1.513M
C. $1.620M
D. $1.470M

Show answer

Correct: B. Arithmetic slips in forecast decks are how forecast credibility is lost, and they are always caught by whoever is being asked to fund the number.

Q2

Top-down forecasting is strongest when:

A. The org is stable and historical growth is a fair guide
B. Team-level detail is already readily available today
C. The teams disagree on the numbers
D. A major migration is planned soon

Show answer

Correct: A. It misses team-specific dynamics by construction, which is exactly what bottom-up is for.

Q3

Bottom-up forecasting’s characteristic bias is:

A. Random error in the inputs
B. Systematic under-counting of the shared services
C. Currency drift over the period
D. Systematic over-estimation by teams protecting headroom

Show answer

Correct: D. Every team adds a margin, and the margins compound at the org level. Knowing the direction of the bias is what lets you correct for it.

Q4

Hybrid forecasting reconciles the two by:

A. Averaging the two figures
B. Taking the higher of the two figures that were given to you
C. Running a conversation that resolves the discrepancies explicitly
D. Taking the lower of the two

Show answer

Correct: C. Averaging hides the disagreement, which is the most informative part: where the two methods diverge is where somebody knows something the other method does not.

Q5

Forecast accuracy should be measured:

A. Against the budget that was originally set for that period
B. Against actuals, per horizon, with variance direction tracked
C. Only at year end, once closed
D. As one org-wide percentage

Show answer

Correct: B. Direction matters more than magnitude: consistent under-forecasting and consistent over-forecasting have different causes and different fixes.

Q6

Accuracy targets should:

A. Loosen as the horizon lengthens
B. Be the same at every horizon
C. Not be set
D. Tighten as the horizon lengthens

Show answer

Correct: A. Holding a 12-month forecast to a 1-month standard guarantees failure and trains everyone to ignore the metric.

Q7

A confidence band communicates:

A. That the forecast itself is simply unreliable this time
B. The best and worst case scenarios
C. The honest range within which the outcome is expected to fall
D. The margin of error in the input data

Show answer

Correct: C. Presenting a single number implies a precision nobody has, and the first time it is wrong, the band you did not give is what you get blamed for not giving.

Q8

Under pressure to give a single number, the right response is:

A. Give the number with its band and the assumptions that would move it
B. Give the midpoint alone, without any qualification whatsoever
C. Refuse to give a number
D. Give the conservative end only

Show answer

Correct: A. The band is not hedging: it is the actual state of knowledge, and stating the assumptions makes it actionable rather than evasive.

Q9

Forecast confidence should:

A. Stay constant right through the whole of the period
B. Widen as the period progresses
C. Narrow as the period progresses and actuals accumulate
D. Be set once and left alone

Show answer

Correct: C. A forecast that never updates as evidence arrives is a prediction rather than a forecast, and re-baselining mid-period is a feature.

Q10

The calibration loop means:

A. Adjusting the forecast until it matches the budget
B. Averaging across multiple methods
C. Recalculating it every month
D. Using measured prior variance to correct future forecasts

Show answer

Correct: D. A team that consistently comes in 8% under can subtract 8%. That is a better forecast than a more sophisticated model that ignores its own track record.


What’s next

Back to Forecasting cloud spend.

Start with the bill.

Foundations takes about five hours. The first lesson is nine minutes.

Open curriculum. No login. No paywall. 290 lessons across 7 courses, three publicly verifiable credentials. Read it on the train, take the exam on a Saturday, list the credential on your résumé Monday.

5h median time to finish Foundations
0 logins, paywalls, or marketing forms
open curriculum, public credential verifier
Multi-cloud automation· Production-ready in 30 min· SOC 2 · ISO 27001· 20–60% off the bill, first month· 4 platforms · 1 console· Multi-cloud automation· Production-ready in 30 min· SOC 2 · ISO 27001· 20–60% off the bill, first month· 4 platforms · 1 console·