Did the time employees take to complete mandatory compliance training get worse after the new training platform was introduced?

The audit question
Fifteen employees were measured before and after a new training platform. Completing it now seems to take longer. Is that a real slowdown — or normal variation?

Scenario

A company rolls out a new compliance training platform. Soon after, team leads start mentioning that the mandatory course seems to take people longer to finish. To check whether that impression is real, the auditor pulls completion times — in days — for fifteen employees trained on the old platform and fifteen on the new one, and runs both sets through the tool.

Data

Days (data0)
4
7
3
5
6
8
4
5
3
7
6
4
5
8
3
Days (data1)
9
14
4
8
11
13
17
10
12
8
15
14
9
11
18

What the tool returned

The auditor pasted both datasets into the tool and ran the analysis. The following result came back:

Analyser
🖊️
Before the change (data0), the metric averaged 5.2, typically ranging between 3.0 and 8.0. Since the change (data1), it now averages 10.0. With 94% confidence the true figure sits between 8.7 and 11.5. 100% of samples tested confirmed an increase of approximately 4.8 — giving good assurance that this increase is true, not a random fluctuation.
Size of the change
4.8
Verdict
Strong increase
Model: Poisson (auto-detected) n₀ = 15 · n₁ = 15 · ESS = 421

Audit conclusion

Training that used to take around five days now takes around ten. Comparing the fifteen completion times from before the platform change with the fifteen from after shows a slowdown too large and too consistent to be normal variation — this is a real deterioration, not an impression.

The slowdown is recorded as a confirmed finding. Next steps: raise it with the platform owner, investigate what changed — course length, platform usability, reminder flow — and re-measure after corrective action. Before attributing the full slowdown to the platform itself, confirm both groups had comparable workloads and identical course content.

Tool usage benefits

Bad news is the easiest news to argue away — "people were just busy," "it's a one-off." ChangeVerifier closes that escape route: it shows the slowdown is systematic, sizes it at roughly five extra days per employee, and attaches a confidence level that makes the finding hard to dismiss. The same tool that confirms improvements also catches deteriorations early — before a vague impression becomes a year of lost productivity. From pasting 30 numbers to reading the verdict took under a minute, with no statistics background needed.