Module 2/6 25 min
Isolate the impact: training or the market?
Step 1 / 2·Three isolation techniques
0/2 steps done
Learning objectives — after this module you will:
- Understand why you must isolate impact before taking credit
- Grasp 3 techniques: control group, trend line, adjusted estimation
- Choose the technique that fits the data context
Learn ~8 min
Three isolation techniques
Know how to separate training's contribution from other factors.
After sales training, sales rose 20%. But in the same period the company cut prices 10% and a competitor ran out of stock. Taking all 20% for yourself is credibility suicide before the CFO. Three isolation techniques, from strongest to lightest:
3 isolation techniques
| Technique | How it works | When to use |
|---|---|---|
| CONTROL GROUP | Compare the trained group with a similar untrained group in the same period | Strongest; feasible with phased rollout (batch 1 trained first = test group, batch 2 = control) |
| TREND LINE | Plot the metric's trend BEFORE training, project it forward; the actual exceeding the projection = estimated impact | When you have stable historical data and no other major changes at the same time |
| ADJUSTED ESTIMATION | Ask participants/managers: what % of the improvement is from training? × confidence (%) — take the conservative figure | When neither a control group nor trend line is possible; be transparent it's an estimate |
The conservative principle
When choosing between two numbers, ALWAYS take the lower. A 45% ROI report fully believed by the CFO beats a 145% one doubted entirely. The measurer's credibility matters more than the number's beauty.
Key takeaway: Control group > trend line > adjusted estimation; always choose the conservative figure.