Employee engagement survey template

How the question set is built

The first block measures engagement itself: pride, intent to stay, discretionary effort and willingness to recommend the company. The other eight blocks measure drivers, the conditions that tend to raise or lower engagement. Separating the two lets you see which drivers move with the outcome in your own data. In most frontline organizations the manager items and the workload and schedule items do most of the work. For the definition and the main survey types, see the employee engagement survey glossary entry.

All items use the same five-point agree scale so they can be compared. Every statement is positive and covers one idea only. "My manager gives feedback and recognizes my work" is two questions, and you cannot tell which half the person disagreed with.

Scoring the results

Report each item as percent favorable: the share of people who chose agree or strongly agree. Neutral counts as not favorable. Theme scores are the average of the items in that theme. As a rough guide, many organizations treat 75 percent favorable or higher as a strength and under 60 percent as a priority, but your own trend matters more than any outside number.

The recommend question gives an employee Net Promoter Score: eNPS = percent of 9 and 10 answers minus percent of 0 to 6 answers. If 200 people answer, 70 give a 9 or 10 and 50 give 0 to 6, eNPS is 35 minus 25, or +10. Scores range from -100 to +100. Treat a single eNPS as a headline only; the driver items tell you what to fix.

Annual survey or pulse survey

A full annual survey with 30 to 45 items gives depth and lets you cut results by department, site and tenure. A pulse survey of 5 to 10 items, sent monthly or quarterly, shows whether things are moving between annual surveys. Most organizations do both: the annual survey to set priorities and a short pulse to track the two or three themes they chose to work on.

For a pulse, take the four engagement outcome items, two manager items and two items from the theme you are working on, plus one open question. Keep the wording identical each time so changes are real.

Surveys do not replace conversations. Pair them with a stay interview template for people you most want to keep, an onboarding feedback survey at 30 and 90 days, and an exit interview form for leavers, so you hear from every stage of tenure.

Running a survey for frontline and shift teams

A 600-agent contact center with three shifts will not get a 70 percent response rate if the survey goes only to company email. Send it by text and post a QR code in break rooms. Give hourly staff paid time to complete it, typically 15 minutes scheduled in the workforce plan, and make sure that time is recorded as hours worked.

Translate the survey into the languages your workforce uses. Report by shift as well as by team, because night-shift results are often very different and are hidden in a site average. A common benchmark for a well-run survey is a response rate of 65 to 85 percent; below 50 percent, the people who did not answer are often the least engaged, so results look better than reality.

Invite people with a short employee survey invitation email that says why you are asking, how long it takes, how anonymity works and when they will see results.

Anonymity, fairness and legal notes

Keep promises about confidentiality exactly. If you say results are only shown for groups of five or more, the survey tool must suppress smaller groups, including when demographic filters are combined. Do not try to identify the author of a comment.

Under the National Labor Relations Act, surveys that solicit grievances and imply promises to fix them during a union organizing campaign can be treated as an unfair labor practice. If organizing activity is underway, talk to labor counsel before launching a survey. Keep demographic questions optional and limited to what you need for reporting. Privacy laws outside the US, such as the GDPR, require a lawful basis and a privacy notice for survey data. This is informational, not legal advice.

Mistakes that kill response rates

The most damaging mistake is running a survey and doing nothing visible with it. Response rates fall at the next survey, and the item "things changed after the last survey" will tell you so. Share results within six weeks, even when they are uncomfortable, and pick a small number of actions per team. Other common mistakes: surveys longer than 15 minutes, managers asking their team how they answered, and changing item wording every year so nothing can be compared.