Proposal: rethinking the feedback course
To: Dana Whitfield, Director of People & Culture; Marcus Oyelaran, COO From: Learning & Performance consultant Date: 2026-10-06 · Re: Manager feedback request from the engagement survey
This is a portfolio artifact. Brightline Health Services and all names, figures and findings are fictional, constructed for portfolio purposes.
What you asked for
A 2-hour e-learning course on giving feedback, for all 70 managers, live next quarter — responding to the engagement survey finding that “employees don’t get useful feedback.”
What we found
We spent two weeks on discovery: six interviews, a 200-person pulse survey across all 12 clinics, a calendar audit, and a review of the review form, the one-on-one policy, the org chart and the manager scorecards.
Managers are not giving structured feedback — but the evidence says the causes are mostly in the system around them, not in their skills:
- Only 31% of managers hold one-on-ones at least every two weeks; there is no cadence standard, no template, and no rostered time. Spans of control have reached 12–18 direct reports.
- The annual review is the only formal feedback point, and the form collects ratings only — there is nowhere to record, or prompt, a real feedback conversation.
- Managers are measured on throughput only. Team development carries 0% weight on every clinic manager’s scorecard. Feedback work is invisible.
- The genuine skill gap is narrow: some managers — concentrated in the 35% promoted since the expansion — avoid difficult conversations and don’t structure feedback well.
Why this matters commercially: 90-day new-hire turnover is 22%; exit interviews repeatedly cite “no idea how I was doing.” At ~$9,400 per new-hire exit, that is roughly $160k/year in avoidable turnover cost. Getting to 15% recovers ≈$47–56k/year (7 points × 78 new hires ≈ 5–6 fewer exits) — and none of it moves if the system stays as it is.
The tell: usefulness scores barely differ between new and veteran managers (2.3 vs 2.6 out of 5). Experience is not the variable. Structure is: staff with a fortnightly one-on-one score 3.4/5 on useful feedback; staff without one score 1.9.
What we recommend
A blend that fixes the system first and trains the narrow skill — total manager time ~90 minutes over 30 days, instead of a 2-hour sitting:
| Component | What it is | Owner |
|---|---|---|
| 1. One-on-one cadence + template | Fortnightly 15-min 1:1s become the published standard; a one-page template ships with it; time rostered like any other clinical duty | Ops + HR |
| 2. Review-form change | Two qualitative questions added (“What did you achieve? What should change?”), mandatory to discuss | HR |
| 3. Span-of-control review | Spans of 12–18 are reported to the COO with options, for a leadership decision | Ops → COO |
| 4. “Feedback in 5 minutes” job aid | One printable page: structure, 3 phrasings, 3 defensive-recovery moves | L&D |
| 5. 15-minute branching scenario | Practice handling a difficult feedback conversation, with consequences — replaces the 2-hour course | L&D |
| 6. 30-day practice challenge | Weekly micro-goals and peer check-ins while the habit forms | L&D + HR |
| 7. Scorecard change (leadership decision) | Add team-development items to clinic manager scorecards | COO |
Cost and time comparison
| Compare | Requested 2-hour course | Recommended blend |
|---|---|---|
| Training-specific time | 140 hrs sitting + scheduling drag ≈ 210 hrs (clinical backfill) | ~90 min each ≈ 105 hrs, spread over 30 days |
| Recurring 1:1 time (the intervention itself) | Not built in — the course ends and the calendar reverts | ~2 h/manager/week at the median span of 13 (13 reports × fortnightly 20-min 1:1s ≈ 2.2 h/week — the document review’s own arithmetic). This row is not overhead on top of the intervention: it is the intervention. That time is already being spent today — unstructured — on firefighting, callbacks and redoing incomplete work; the cadence makes it deliberate, and it is the spend that displaces the rework and exit costs below. |
| Build cost | ~$18k (external SCORM build quoted) | ~$6k internal-equivalent (scenario + job aid + template) |
| Reaches which causes | Skill only (cause 4) | All four causes |
| Builds a lasting structure | No | Yes: cadence, template, form, scorecard stay after launch |
| Can be measured against turnover | Weakly | Yes (see below) |
Read the two time rows together: the 2-hour course bills ~210 one-off hours; the blend bills ~105 one-off hours plus the recurring 1:1 hours managers are already losing every week to the problems the blend fixes. The honest comparison is not “course hours vs blend hours” — it is what each option turns the time into: a sitting, or a working cadence.
How we’ll measure success
- Leading: % of managers at a fortnightly-or-better 1:1 cadence (31% → 80%); scenario completion with debrief (90%).
- Behavior: staff pulse item “I received useful feedback in the last 2 weeks” (2.4/5 → 3.5/5).
- Lagging: 90-day turnover (22% → 15% in 12 months).
- Design: staggered rollout by clinic so unrolled clinics act as a comparison group. Full plan: measurement-plan.
What we need from leadership
- COO: approve the cadence standard and rostered time; approve adding team-development items to scorecards; receive the span-of-control review.
- HR: approve the review-form change; endorse the template; co-own the 30-day challenge.
- Ops directors: protect the rostered one-on-one time in clinic schedules.
Risks
| Risk | Mitigation |
|---|---|
| “Less training” reads as “doing less” | This proposal ships more structure than a course would: standard, template, form change, scorecard items, practice. |
| Managers see 1:1s as admin burden | 15 minutes, templated, rostered; the challenge makes the first 30 days social, not solo. |
| Scorecard change stalls | Components 1–6 proceed without it; measurement isolates its effect via the staggered rollout. |
| The board expects “a course” by Q2 | Scenario + job aid + challenge launch inside the quarter; the cadence standard lands the same week. The board deliverable is the turnover trajectory, not an LMS package. |
Recommendation
Approve the blend. If the evidence had shown a broad skill gap, we would build the course — the discovery shows the opposite, and the turnaround metric is attached to the system, not to knowledge.