Diagnosis and data collection
Step 1 / 3·Case study: Starbucks closes 7,100 stores to retrain
0/3 steps done
Learning objectives — after this module you will:
- Draw a TNA lesson from Starbucks' retraining shutdown decision
- Preliminarily distinguish training problems from non-training problems
- Choose a suitable data-collection method
Case study: Starbucks closes 7,100 stores to retrain
See organization-level TNA lead to a bold decision.
On the afternoon of Feb 26, 2008, Starbucks did the unprecedented: it simultaneously closed about 7,100 U.S. stores for over 3 hours to retrain roughly 135,000 baristas on espresso skills. Lost revenue was estimated at about $6 million in that afternoon alone.
Why? Howard Schultz had just returned as CEO and saw a gap at the ORGANIZATION level: during over-rapid expansion, coffee quality — the brand's core experience — had declined. This wasn't 'training for fun': it stemmed from a real business signal (falling quality, a diluted experience) and targeted a specific capability (correct espresso extraction technique).
The TNA lesson: (1) the biggest training needs usually surface at the organization level through business metrics, not through training requests; (2) a worthwhile training decision must tie to something the organization truly values — here, brand quality; (3) well-timed training is a strategic statement, not a cost.
A nuance to remember
Retraining helped Starbucks refocus on quality, but its turnaround also relied on many other changes (product, stores, strategy). Good TNA doesn't overclaim: it identifies exactly the part training can solve, and is honest about the rest.

Key takeaway: Starbucks 2008: a business signal (falling quality) → an organization-level training need → action on the right core capability.
