The New-Hire Onboarding That Isn't Sticking
A 1,200-employee logistics company redesigned onboarding six months ago. New hires now attend a two-day in-person session (16 hours) covering safety, systems, culture, and customer service, delivered by four different facilitators who each use their own slides. Post-session surveys average 4.6/5, so leadership considers it a success. But you notice: 90-day voluntary turnover for new hires is 28% (up from 19% before the redesign), first-line supervisors report new hires 'can't do the basics' for weeks, and the safety module-delivered last on day two when everyone is fatigued-has the lowest quiz scores. You've been asked to diagnose the problem without a training budget increase.
1. The surveys say 4.6/5 but outcomes are worsening. How do you diagnose what's actually wrong?
Read the worked answer
High satisfaction with poor outcomes is a classic signal that you're measuring the wrong thing. Diagnose across levels, not just reaction.
- Separate the data sources. Reaction (4.6/5) measures how people felt, not whether they can perform. Pull Level 2 (quiz/knowledge), Level 3 (supervisor observation of on-the-job behavior), and the turnover data together.
- Look for structural causes, not just content causes. Four facilitators with different slides means inconsistent content-new hires literally get different training depending on cohort. That alone could explain 'can't do the basics.'
- Test the sequencing hypothesis. Safety last on a fatigued day two is a known attention problem; low quiz scores there are consistent with placement, not necessarily content.
- Interview a sample. Talk to 6-8 recent hires and their supervisors about what they couldn't do in weeks 1-4. Ask supervisors to name the specific 'basics' that are missing-this defines the real performance gap.
- Check whether the gap is training or environment. If supervisors lack time to coach or systems are hard to access, more training won't fix it. Confirm before redesigning.
2. You have no additional budget. What delivery changes would you make, and how would you sequence them?
Read the worked answer
With no new money, the levers are sequencing, standardization, and shifting work off the live session.
Sequence:
- Standardize facilitator materials first (lowest cost, highest leverage). One shared deck and facilitator guide eliminates cohort-to-cohort variance. This is a coordination fix, not a spend.
- Move safety earlier and to a higher-attention slot; pair it with a short practice/check rather than a lecture at the end of day two. Reallocate existing time rather than adding hours.
- Redistribute content. Move reference-heavy material (systems how-tos, policy) to job aids or short pre-work the new hire does before day one, freeing live time for practice and culture.
- Add spaced follow-up using existing channels (supervisor huddles, a weekly email or LMS nudge) so learning continues past the two days-this addresses the 'weeks to competence' complaint without new cost.
- Brief supervisors on the specific 'basics' and give them a short observation checklist so on-the-job reinforcement is consistent.
Measure as you go: track safety quiz scores and a supervisor observation rating before/after, so you can show whether the sequencing and standardization changes moved behavior-not just satisfaction.