← ccPlanning Academy · Real-time track
Reading the day’s signals
Slides done? Here’s the same idea in a bit more depth — the part worth keeping.
In depth: one number never tells the whole story
A red service level could mean a volume spike, an AHT creep, an adherence problem, or a one-off outage — and each needs a completely different response. Reading the day well means watching a small set of signals together and letting their combination point at the cause, rather than reacting to any single dial. Service level itself is the smoke alarm, not the diagnosis: it reliably tells you that something is wrong and almost never what.
The signals, and why they’re read in pairs
Beyond service level, the live leading indicators are the queue and longest wait — a growing queue and a rising oldest-contact time tell you trouble is building now, before the interval’s service level has even been calculated, so this is what you watch most closely. Forecast versus actual volume answers the demand-or-supply question: if actual is well above forecast the plan was right but the world changed; if volume is on plan and you’re still short, the problem is on your side. Occupancy shows how hard the available agents are working — very high means no slack and burnout risk, while low occupancy alongside a queue points at routing or skilling, not headcount. And adherence against staffing actual tells you whether the people you planned for are really on the phones.
The combination is the cure — and don’t over-watch
The diagnosis lives in the pattern: high volume with on-plan adherence means a demand surge, so add capacity; on-plan volume with low adherence is a conformance problem, so get people back on task; a queue with low occupancy is a routing or skill mismatch. The same red service level has different cures depending on what the other signals say. The discipline is to pick the handful of metrics that actually predict trouble and reveal cause, and watch those well — a wall of forty dials hides the five that matter and makes decisions slower, not better.
The principle to remember: watch a few signals together and let the combination diagnose the cause. Service level is the alarm; the queue, volume-vs-forecast, occupancy and adherence tell you what to actually do.
Quick quiz
Five questions. Pick an answer to each, then check your score.
1. Why is service level on its own a poor diagnosis?
A red SL could be volume, AHT, adherence or an outage — the other signals tell you which.
2. Which are the most useful leading (early-warning) signals?
Queue and oldest-contact time show trouble building now, before the interval’s SL is even known.
3. What does forecast-vs-actual volume tell you?
Volume above forecast = demand surge; on-plan volume but short = a supply-side issue.
4. A queue is building but occupancy is low. What does that suggest?
Queue + low occupancy points at routing/skills, not a raw headcount shortage.
5. What’s the danger of monitoring forty metrics at once?
Pick the handful that predict trouble and explain the cause — watch those well.