← ccPlanning Academy · Capacity planning track

Scenarios & sensitivity

Free visual lesson · about 5 minutes · short quiz at the end

ccPlanning academy · capacity planning

Scenarios & sensitivity

A long-range plan built on a single guess is a single point of failure.

The big idea

The further out you plan, the wider the uncertainty.

Next quarter’s volume, AHT and attrition are all estimates, and small errors compound over a year. Presenting one precise headcount number pretends to a certainty you don’t have. Good capacity planning works in ranges and tested scenarios.

Three scenarios

Low, expected, high.

Run the plan three ways: a conservative case, your best estimate, and a stretch case. Now you can say “we need 180 heads if growth lands as expected, 165 if it’s soft, 200 if it’s strong” — which is honest, and lets leadership decide how much risk to carry.

Sensitivity analysis

Which assumption moves the answer most?

Flex one input at a time and watch the required FTE. Often one or two assumptions — usually attrition or AHT — swing the plan far more than the rest. Those are the numbers to forecast hardest, monitor closely, and caveat loudest.

AHT attrition shrinkage volume

The widest bars are where the plan is most exposed.

What-if events

Model the big known unknowns.

A new product launch, a site closure, a large client win or loss, a planned IVR change — build these as discrete what-if overlays. Leadership can then see the capacity (and cost) implication of a decision before they commit to it. That’s planning as a decision tool, not a report.

Hedging the plan

Decide deliberately how much risk to carry.

You rarely staff to the worst case (too expensive) or the best (too risky). The scenarios let you choose consciously: plan to expected, hold a flex buffer (overtime, temps) for the high case, and have a slow-down lever for the low. The point is a deliberate hedge, not a hopeful single number.

Keep it legible

Three clear scenarios beat twenty.

It’s tempting to model every permutation. Resist it — a wall of cases is as useless as a single point. Low / expected / high, plus a couple of named what-ifs, is enough for a real decision. Clarity is the goal, not combinatorial completeness.

Why the hedge pays

180, 165 or 200?

Staff to the high case (200) and you carry ~20 idle heads if growth is soft. Staff to the low case (165) and you’re 35 short if it’s strong — with a lead time too long to recover.

So you plan to 180, hold a flex buffer (overtime, temps) for the upside, and keep a slow-down lever for the downside. The range turns one risky bet into a deliberate one.

The takeaway

Plan in ranges; know your sensitivities.

Run low/expected/high, find the one or two assumptions that move the answer most, overlay the big what-ifs, and hedge deliberately. Presenting a range with named risks is more honest — and more useful to a decision-maker — than false precision.

Now test yourself ↓

1 / 8

Slides done? Here’s the same idea in a bit more depth — the part worth keeping.

In depth: a single guess is a single point of failure

The further out you plan, the wider the uncertainty — and a year-long capacity plan is built entirely on estimates of volume, AHT and attrition, each of which can be a little wrong, with the errors compounding over the months. Presenting one precise headcount number to leadership pretends to a certainty you simply don’t have. Good capacity planning is honest about that, and works in ranges and tested scenarios rather than a single hopeful figure.

Scenarios and sensitivity

Run the plan three ways — a conservative low case, your best expected estimate, and a stretch high case — so you can say “180 heads if growth lands as expected, 165 if it’s soft, 200 if it’s strong,” and let leadership decide how much risk to carry. Then do sensitivity analysis: flex one input at a time and watch the required FTE. Usually one or two assumptions — most often attrition or AHT — swing the answer far more than the rest, and those are the numbers to forecast hardest, monitor most closely, and caveat loudest. On top of that, model the big known unknowns as discrete what-if overlays — a product launch, a site closure, a major client win or loss — so leadership can see the capacity and cost implication of a decision before they commit to it.

Hedge deliberately, and keep it legible

You rarely staff to the worst case (too expensive) or the best (too risky). The scenarios let you choose consciously: plan to expected, hold a flex buffer of overtime or temps for the high case, and keep a slow-down lever for the low. That’s a deliberate hedge rather than a single number you’re quietly hoping is right. And resist the urge to model every permutation — a wall of twenty cases is as useless as one point estimate. Low, expected, high, plus a couple of named what-ifs, is enough for a real decision; clarity is the goal, not combinatorial completeness.

The principle to remember: plan in ranges and know your sensitivities. Run low/expected/high, find the one or two assumptions that move the answer most, overlay the big what-ifs, and hedge on purpose — a range with named risks beats false precision every time.

Quick quiz

Five questions. Pick an answer to each, then check your score.

1. Why plan capacity in ranges rather than a single number?

Small errors in volume, AHT and attrition compound over a year — work in ranges and scenarios.

2. What are the three standard scenarios?

Three cases let leadership decide how much risk to carry — honestly.

3. What does sensitivity analysis tell you?

The widest-swinging inputs are the ones to forecast hardest and caveat loudest.

4. How should you handle a big known event like a product launch?

What-if overlays turn the plan into a decision tool, not just a report.

5. How many scenarios should you present?

A wall of cases is as useless as a single point — clarity beats combinatorial completeness.