Retention and Regretted Attrition

Scenario

Example organisation: in one year, a 12-person team loses four engineers, three of them regretted. Each exit is treated as a surprise. The manager backfills, but delivery drops for months and the remaining engineers feel overloaded and start looking too.

What you see on the team

• Resignations from people who "seemed fine".
• Exit interviews mentioning growth, recognition, pay or the manager.
• Delivery drops after each departure; knowledge lost.
• A spiral: departures increase load, which causes more departures.

The common mistake

Treating attrition as bad luck and responding only after resignations, instead of tracking risk and investing in retention continuously.

Why it happens

Retention work is quiet and its success is invisible: people simply stay. The cost of a departure is spread over months and rarely added up. Gallup notes managers account for up to 70% of the variance in team engagement (Gallup) — retention is largely in the manager's hands.

The playbook

Estimate the real cost so retention gets the attention it deserves:

annual_salary = 110_000
loaded = annual_salary * 1.3                  # benefits, taxes, equipment
weekly = loaded / 52

vacancy_weeks = 12                            # open role until the new hire starts
ramp_weeks = 16                               # new hire reaches full productivity
team_drag_weeks = 8                           # others cover, interview, onboard (0.5 person)
recruiting = 0.20 * annual_salary             # agency fee or internal recruiting cost

lost_output = vacancy_weeks * weekly + ramp_weeks * weekly * 0.5 + team_drag_weeks * weekly * 0.5
total = lost_output + recruiting
print(f"lost output:      ${lost_output:,.0f}")
print(f"recruiting:       ${recruiting:,.0f}")
print(f"total per exit:   ${total:,.0f}  (~{total / annual_salary:.1f}x annual salary)")
print(f"three exits/year: ${3 * total:,.0f}")

lost output:      $66,000
recruiting:       $22,000
total per exit:   $88,000  (~0.8x annual salary)
three exits/year: $264,000

Know each person's risk factors through 1:1s: stalled growth, below-market pay, low recognition, poor fit with work, conflict, burnout, a better offer.

Hold "stay conversations", not just exit interviews:

- What keeps you here? What would make you leave?
- What's one thing that would make your job more satisfying?
- Is there work you'd like to do that you aren't doing?
- How are you feeling about your growth and your pay?

Act on the answers: growth opportunities, recognition, pay adjustments with data, workload, team changes.

Track regretted attrition as a team health metric and discuss causes with your manager.

When someone resigns, have an honest conversation; counter-offers rarely fix growth or relationship problems. Plan knowledge transfer and celebrate their contribution.

What to say

Stay conversation:

"This isn't a performance conversation — I just want to understand what keeps you here and what might not. If you were offered a job elsewhere tomorrow, what would make it tempting?"

Trade-offs

• Retention efforts can't keep everyone; some departures are healthy for the person and the team.
• Counter-offers may fix pay short term, but if the cause is growth or management they rarely last.
• Asking about leaving can feel awkward; asked sincerely, most people appreciate it.

Signals it's working

• Regretted attrition falls; departures are rarely surprises.
• You can name, for each person, what keeps them and what might not.
• Exit reasons shift away from growth, recognition and management.
• Teams recover from departures faster (less knowledge concentration).

Key points

• A regretted departure can cost close to a year's salary once vacancy, ramp-up, team drag and recruiting are counted.
• Know each person's risk factors; hold stay conversations, not just exit interviews.
• Act on growth, recognition, pay data and workload before resignations.
• Track regretted attrition and its causes as a health metric.