Why is intent to stay considered the best predictor of turnover?

Meta-analyses by Steel and Ovalle (1984) and Griffeth, Hom, and Gaertner (2000) found that intent to leave correlates with actual turnover at r = 0.50 and r = 0.38, respectively — the strongest single-item predictor in the turnover literature. That said, Cohen et al. (2016) found the relationship can diverge at the organizational aggregate level. Use intent to stay as a leading indicator and leading indicator only: it tells you where risk is concentrated, not that everyone who scores low will definitely leave.

How many questions should a morale survey have?

Aim for 20–26 items for a quarterly morale pulse, keeping completion to under 7 minutes. The core is a 5-item morale block, a 5-item intent-to-stay block, 1–2 loyalty items, and 3 open-ended flight-risk probes. Longer surveys reduce completion rates without improving actionability — every question should earn its place by informing a decision you can actually make.

Should the morale survey be anonymous?

Most morale surveys are confidential (data held by a third party, reported only in aggregate) rather than truly anonymous, because you need demographic segmentation to make the data actionable — for example, identifying which team or tenure band is at highest risk. The critical protection is the reporting threshold: never report results for groups smaller than n=5, which prevents re-identification of individual responses. Be transparent in the survey intro about which standard you use.

How often should we run a morale and intent-to-stay survey?

Quarterly is the most effective cadence for a morale and retention-risk pulse. It gives you four data points per year — enough to see trend direction — while still allowing two to three weeks between close and results share-back to act before the next wave. Monthly is too frequent for most teams to absorb and act on results; annual is too infrequent to catch flight risk early enough to intervene.

What is the difference between a morale survey and a stay interview?

A morale survey is a structured, confidential questionnaire sent to all employees at once — it gives you aggregate trend data and surfaces systemic flight risk. A stay interview is a one-on-one conversation between a manager and an individual employee, designed to surface that person's specific retention anchors and concerns. Stay interviews are not anonymous, not scalable to the whole team in a single wave, and not suited for aggregate benchmarking. Use the morale survey to find the 'where' (which team, which segment); use stay interviews to go deep on the 'why' with specific individuals at risk.

Employee Surveys

What Questions Should Be on an Employee Morale Survey?

An employee morale survey should measure three things: current team sentiment, intent to stay, and the eNPS-style loyalty signal — because intent-to-stay is the strongest single predictor of actual turnover (meta-analytic r = 0.38–0.50). A 20–26-item survey with 5-pt Likert morale items, intent-to-stay items, an eNPS-style recommend question, and open-ended flight-risk probes gives you a leading retention indicator, not a lagging one. Replacing a salaried employee costs six to nine months of salary (SHRM), so catching morale dips early is cheaper than backfilling the role.

21 QuestionsFlight-Risk / RetentionQuarterly Pulse5-pt Likert + 0–10 + Open

The Question Bank

Copy-Ready Questions, Grouped by Theme

Every group uses the scale that fits it. Copy one question, a whole theme, or the full set straight into your survey tool.

Filter by theme21 questions

Current Team Morale

5-pt Likert1 Strongly disagree → 5 Strongly agree

Measures how employees feel about the team's mood and energy right now — the sentiment baseline you track over time. Items draw on compiled best-practice morale constructs (QSET-021, not a named instrument).

1.
Morale on my team is high right now.
QSET-021 (compiled best-practice)
The direct morale signal — the single item most correlated with team-level sentiment trends when tracked quarterly.
2.
I feel positive about coming to work each day.
QSET-021 (compiled best-practice)
Affective tone of daily work experience; low scores here often precede intent-to-leave spikes.
3.
My team's energy and enthusiasm have been strong recently.
Team-level vitality item; distinct from individual morale — catches collective dip before attrition surfaces.
4.
I feel proud to work for this organization.
Organizational pride predicts both advocacy and retention; decline is an early flight-risk signal.
5.
I feel that my contributions are recognized and valued.
Recognition deficit is consistently among the top drivers of voluntary departure — pairing it with morale anchors the action: recognition is a lever you can pull.

Intent to Stay

5-pt Likert1 Strongly disagree → 5 Strongly agree

The flight-risk core. Intent to stay is the strongest single predictor of actual turnover — meta-analytic r = 0.38 (Griffeth et al. 2000) to 0.50 (Steel & Ovalle 1984). Items are compiled best-practice (QSET-021) anchored to that validated construct.

1.
I rarely think about looking for a job elsewhere.
QSET-021 (compiled best-practice; construct anchored to Griffeth et al. 2000; Steel & Ovalle 1984)
Reverse-keyed intent-to-leave item — the closest proxy for actual turnover intention in short survey form.
2.
I expect to be working at this organization 12 months from now.
QSET-021 (compiled best-practice)
Time-anchored retention expectation; the 12-month horizon matches operational planning cycles and has strong predictive validity.
3.
I see a long-term future for myself here.
Measures longer-horizon commitment beyond the 12-month window; captures employees who are staying short-term but mentally out.
4.
I feel motivated to grow within this organization rather than elsewhere.
Growth orientation within the org is a leading retention signal; its absence surfaces the 'stuck but looking' employee segment.
5.
If I received a comparable offer elsewhere, I would seriously consider staying here.
Tests the strength of psychological attachment — not just whether they plan to stay, but whether a competing pull could dislodge them.

Loyalty & Advocacy

0–10 scale0 Not at all likely → 10 Extremely likely

An eNPS-style recommend item plus a Likert advocacy item. The 0–10 rating generates a loyalty signal you can trend over time. Note: the eNPS scoring calculator and interpretation bands live on the eNPS page — link there for your score.

1.
On a scale of 0 to 10, how likely are you to recommend [organization] as a place to work?
QSET-004 (eNPS — adapted from Net Promoter Score, Reichheld & Bain, 2003)
The standard eNPS item (Reichheld/Bain). Lock wording and scale for ≥12 months to preserve trend comparability; never average individual scores — aggregate to %Promoters − %Detractors.
2.
I would recommend this organization as a great place to work to a friend or colleague.
5-pt LikertQSET-021 (compiled best-practice)
A Likert companion to the 0–10 item, useful when you want % favorable alongside the NPS-style score. Tracks advocacy in the same morale survey sweep.

Work Environment & Conditions

5-pt Likert1 Strongly disagree → 5 Strongly agree

Situational factors most likely to tip a borderline employee toward leaving: workload, psychological safety, manager relationship, and clarity. These are the levers you can pull between survey waves.

1.
My workload is manageable and sustainable.
Workload strain is one of the top modifiable attrition drivers; a low score here predicts burnout-driven departures within the next quarter.
2.
I have the resources and support I need to do my job well.
Resource adequacy is both a morale and a retention driver — employees who feel set up to fail exit sooner.
3.
My manager supports my wellbeing and professional growth.
Manager quality mediates the relationship between morale and intent to leave; this item surfaces team-level manager risk early.
4.
I feel safe raising concerns or disagreeing without fear of negative consequences.
Psychological safety predicts willingness to stay and speak up; its absence is strongly correlated with quiet quitting and eventual departure.
5.
Decisions that affect my work are communicated clearly and in a timely way.
Communication clarity during change is a consistent morale predictor; gaps here spike flight risk, especially during reorgs or policy shifts.
6.
How satisfied are you with your current work arrangement (on-site, hybrid, or remote)?
Work arrangement satisfaction matters: ~6 in 10 exclusively-remote employees say they are extremely likely to job-search if remote flexibility is removed (Gallup, 2025 — STAT-015). Surface the RTO flight-risk signal early.

Flight-Risk Open-Ends

Open textFree text

Two targeted open-ended probes that surface the 'why' behind low quantitative scores. These items are compiled best-practice (QSET-021 and QSET-004). Qualitative responses should be thematically coded before your results share-back.

1.
What, if anything, is making you consider leaving this organization?
QSET-021 (compiled best-practice)
The single most diagnostic retention question — ask it every wave, even when quantitative scores look fine. Voluntary attrition nearly always has a proximate cause that surfaces here first.
2.
What one change would make you significantly more likely to stay?
QSET-004 open-end (adapted)
Action-oriented framing; produces more actionable responses than a generic 'suggestions' prompt. Pairs with the recommend-item open-end.
3.
What do you value most about working here — what keeps you?
QSET-021 (compiled best-practice)
Identifies retention anchors, not just risks. Understanding what's holding people helps you protect those factors during change. Code stays as often as you code flight risks.

Decision Guide

When Should You Use This Survey?

Match the survey type and cadence to your situation.

🤝

You want to catch flight risk before people hand in notice

Use

Run this morale and intent-to-stay pulse quarterlySegment intent-to-stay by team and tenurePair quantitative scores with the open-end flight-risk probes

Avoid

Waiting for your annual engagement census to surface the signal

Intent-to-stay predicts actual turnover (r = 0.38–0.50), but the correlation only helps you if you measure it before people leave. A quarterly morale pulse gives you 4 data points per year vs 1.

🚪

You want to understand why people are leaving after they have already decided to go

Use

Use an exit interview or exit survey insteadLink exit themes back to morale survey signals from prior quarters

Avoid

Using this morale survey as a post-departure debrief

This survey is the leading indicator; exit interviews are the lagging one. Run both — morale-retention catches risk early, exit interviews confirm the root cause after the fact.

🏠

You have a remote-work policy change coming and want to quantify attrition risk

Use

Add the work-arrangement satisfaction item to this pulseCrosstab intent-to-stay by current work arrangementNote: ~6 in 10 exclusively-remote workers say they are extremely likely to job-search if remote flexibility is removed (Gallup 2025)

Avoid

Announcing policy changes without first surveying sentiment — you will lose the retention signal

A pre-announcement morale pulse gives you a baseline; a post-announcement pulse lets you see the actual impact on intent to stay. Two data points tell a story; one data point is just noise.

📊

You want to hold managers accountable for team retention

Use

Report intent-to-stay and morale scores by team (n≥5 threshold required)Pair manager-level flight-risk data with manager-effectiveness scoresLink to stay-interviews for the follow-up conversation

Avoid

Reporting team-level flight-risk data to managers without coaching or a clear action plan

Managers who see their team's flight-risk data without support become defensive, not responsive. Pair the data with a structured stay-interview guide and a clear 'what to do this week' action.

⚡

You want a single loyalty number to trend over time

Use

Use the eNPS item in this surveyVisit the eNPS page for scoring bands and the interactive calculator

Avoid

Building a custom scoring model on top of the morale items

The eNPS question has a defined formula, known benchmarks, and ≥20 years of use. The morale items are a flight-risk index, not a single scored metric — report them as % favorable per item and track trends.

Benchmarks

What "Good" Looks Like

Scores only mean something against a benchmark. Here are the numbers worth measuring against.

6–9 months of salary

Cost to replace a salaried employee (e.g. $30k–$45k for a $60k role); Gallup estimates 50–200% of annual salary by role level.

SHRM; Gallup — STAT-009

r = 0.38 to 0.50

Meta-analytic correlation between intent to leave and actual turnover — the strongest single predictor of voluntary departure.

Steel & Ovalle (1984); Griffeth, Hom & Gaertner (2000) — STAT-011

20% globally engaged in 2025

Global employee engagement rate — down from a 2022 peak of 23%. With morale sliding, flight-risk surveys are an early-warning system, not a nice-to-have.

Gallup, State of the Global Workplace 2026 — STAT-001

~6 in 10 exclusively-remote employees

Share of exclusively-remote US workers who say they are extremely likely to job-search if remote flexibility is removed — a concrete policy flight-risk signal.

Gallup, 'Hybrid Work in Retreat? Barely' (Sept 2025) — STAT-015

24–25% lower turnover (high-engagement orgs)

Top-quartile engaged business units see 24–25% lower turnover vs bottom-quartile in high-turnover industries; 59–65% lower in low-turnover industries.

Gallup Q12 Meta-Analysis (10th/11th ed.) — STAT-004

Do It Right

Survey Design Best Practices

The methodology that separates a survey people answer honestly from one they ignore.

Run morale and intent-to-stay on a recurring pulse — not just in an annual census

Intent to stay is a leading indicator: it predicts who will leave before they hand in notice. Tracking it quarterly (not just annually) surfaces dips while there is still time to act. Pair this pulse with your annual engagement census — continuous listening means one deep census plus lighter recurring pulses. Don't survey faster than you can act; fatigue comes from inaction, not frequency.

METH-004 — ContactMonkey; Vantage Circle; Perceptyx

Lock eNPS wording for at least 12 months

The recommend item (0–10) is only comparable across waves if wording and scale are identical. Never average individual scores — aggregate to %Promoters minus %Detractors. Suppress results for any group below n=5 (the 'Rule of 5') — a team of four is too small to report without re-identification risk. Do not tie eNPS to manager compensation; Reichheld explicitly warned against linking NPS scores to pay.

METH-010 — Sopact; Lattice; METH-007 — 15Five

Apply a minimum reporting threshold (n≥5) to intent-to-stay crosstabs

Flight-risk data sliced by team or manager is highly sensitive. If you report intent-to-stay results for a group of three or four, the individuals can often infer who said what. Set your reporting threshold before you launch — most vendors default to n≥5 — and never release results for groups below that threshold, even informally in a slide deck.

METH-007 — 15Five; WorkTango; CultureMonkey

Close the loop within two weeks of results

Morale surveys that produce no visible action accelerate the decline they were designed to catch. Within about two weeks of results, share what you heard and commit to specific actions — even if the action is 'we heard you but can't change X, and here's why.' Only 51% of employees report that actual improvements result from their survey feedback (Perceptyx, 2026). Being in the other half is the real differentiator.

METH-008 — Perceptyx; Gallup; Culture Amp

Avoid These

Survey Mistakes That Wreck Your Data

Measuring morale once a year and calling it a retention program

An annual survey tells you how people felt six months ago on average. Intent-to-stay works as a leading indicator only when measured frequently enough to catch dips while you still have time to intervene. By the time your annual results are in, the people at highest risk may have already accepted other offers.

Instead: Run a quarterly morale and intent-to-stay pulse alongside your annual census. Four data points per year vs one is the difference between a retention program and a post-mortem.

Asking intent-to-stay questions but never acting on low scores

Only 51% of employees report that actual improvements result from survey feedback (Perceptyx, 2026). Running a flight-risk survey and taking no visible action is worse than not running it — it signals that you measured the problem but chose to ignore it, which accelerates the departure of the employees you most want to keep.

Instead: Schedule a 'what we heard / what we'll do' communication within two weeks of results. Even acknowledging a problem you cannot immediately fix is better than silence.

Surveying groups too small to be anonymous — and reporting anyway

A team of four cannot provide truly anonymous intent-to-stay data. If you report flight-risk results for groups below n=5, individuals can often identify their own or colleagues' responses, which destroys trust and suppresses future participation — precisely the opposite of what you need for sensitive retention data.

Instead: Set your reporting threshold at n≥5 before you launch. Suppress sub-threshold groups entirely; do not round up or share informally in slide decks.

Double-barreled flight-risk questions that mix two constructs

Asking 'Are you satisfied with your pay and career growth opportunities?' combines two independent drivers in one item. A respondent satisfied with pay but dissatisfied with growth has no valid answer. The result: noisy data that cannot tell you which lever to pull.

Instead: One item, one construct. Ask about pay in one item and career growth in another. Use the validated QSET-021 items as a starting point — they are deliberately single-construct.

Confusing morale with engagement — and writing the wrong questions

Morale is current affective state ('how does the team feel right now?'); engagement is a driver of discretionary effort ('do people bring their best?'). The distinction matters because the interventions differ: low morale often responds to manager communication and recognition; low engagement needs deeper driver work (growth, meaning, role fit). Blending the two produces an unfocused survey and an unactionable result.

Instead: Keep this survey focused on sentiment and intent to stay. For engagement drivers, use your annual engagement census (Q12-anchored driver groups). Link the two — morale tells you where to look; engagement tells you why.

Tying eNPS scores to manager compensation

Fred Reichheld, who developed the NPS methodology, explicitly warned against linking NPS scores to manager pay. When managers know their bonus depends on the recommend score, they coach responses, exclude low-scoring employees, or inflate their team's results — destroying the validity of the metric you are trying to track.

Instead: Use eNPS as a diagnostic and trend tool, not a performance metric tied to rewards. Surface low scores in coaching conversations, not in pay review.

Sources & Licensing

Where These Questions Come From

Validated instruments have owners. Here's what's adapted from what — and how to use each one without stepping on a license.

eNPS / Net Promoter Score

The recommend item ('On a scale of 0 to 10, how likely are you to recommend [organization] as a place to work?') adapts the Net Promoter Score methodology. The question wording is generic and freely reproducible. 'Net Promoter Score' and 'NPS' are registered trademarks of Reichheld, Bain & Company, and Satmetrix — attribute the methodology when publishing.

Source: Reichheld, F. (2003). The one number you need to grow. Harvard Business Review, 81(12), 46–54.

Morale & Intent-to-Stay Items (QSET-021)

The intent-to-stay and morale questions in this survey are compiled best-practice items — they are NOT a named, proprietary instrument and carry no license restriction. The intent-to-stay construct is academically validated: cite Griffeth, Hom & Gaertner (2000) and Steel & Ovalle (1984) for construct validity, but do not attribute these specific items to any commercial instrument.

Source: Griffeth, R. W., Hom, P. W., & Gaertner, S. (2000). A meta-analysis of antecedents and correlates of employee turnover. Journal of Management, 26(3), 463–488.

Ready to Send

Launch & Follow-Up Templates

The invite, the reminder, and the results share-back — the messages that drive response rates.

Morale Pulse — Email Invitation

Subject: 5-minute check-in — your voice shapes how we respond Hi [First Name], Every quarter, we run a short morale and intent-to-stay survey so we can act on what's actually happening — not what we assume. This one takes about 5 minutes. Your responses are confidential and reported only in aggregate groups of 5 or more. No individual response is ever attributed to you. [Link to survey] The survey closes on [Date]. We will share a summary of what we heard and what we plan to do within two weeks of closing. Thank you, [Manager / HR Lead Name] [Company]

Send from the direct manager or CHRO — visible sponsorship is the single biggest driver of response rate. Include the anonymity guarantee in the first three lines.

Frequently Asked Questions

Morale measures current affective state — how the team feels right now, and whether people are thinking about leaving. Engagement measures drivers of discretionary effort (Q12, UWES constructs like vigor, dedication, and absorption). A satisfied, loyal employee can be disengaged; a highly engaged employee can still have low morale during a tough quarter. Run a morale pulse quarterly to catch flight risk early; run an engagement census annually or biannually to diagnose the underlying drivers.

Keep Exploring

Related Survey Guides

Employee Surveys

What Questions Should Be on an Employee Engagement Survey?

34 employee engagement survey questions grouped by validated driver — Q12-based, UWES-anchored, with the right scale and benchmarks. Copy-ready for your annual census.

Read guide

Employee Surveys

What Is eNPS and How Do You Calculate It?

Learn what eNPS is, how to calculate it with real benchmarks, and get the exact question wording — plus an interactive calculator to read your score instantly.

Read guide

Employee Surveys

What Questions Should You Ask in an Exit Interview?

The complete exit interview question bank: 30 questions across 7 themes — reasons for leaving, pay, manager, culture, growth, rehire intent — with templates.

Read guide

Browse All Survey Guides

A Survey Only Helps If You Act On It

The fastest way to tank your next response rate is to collect feedback and do nothing. Actify turns survey findings into action — recognition, engagement activities, and wellbeing benefits employees actually choose.

Book a 20-min Demo See How It Works

No credit card required. 15-minute setup.