TAC HRTAC HR
A scene from a Saudi workplace
HR templates

Pulse Survey Policy Template

The long annual opinion survey arrives late. Answers are gathered at the end of the year, the results are shown two months later, and whatever the employee complained about has either gone stale or the employee has left. A short recurring survey asks a few questions every few months, so the manager sees the direction of opinion while it is still moving, and the employee sees that what he said arrived and that something changed because of it. None of this works without two conditions: confidentiality that people believe, and a visible reply to the results. This policy writes down both conditions and the rhythm between them. It is a guidance template that you adjust to your own company.

Last reviewed
On this page

01What a survey policy is for, and who it covers

The policy aims to measure employees' opinion of the working environment, management, pay and development at a regular pace, and to turn what appears into decisions that are announced. It applies to all employees or to one branch or department, depending on the wave.

A survey is not a complaints channel and not an appraisal of an individual. Anyone with a specific complaint follows the complaints route, and anyone who wants to raise something with a manager follows the open door policy. The survey gathers the general direction, and it is never meant to find out who said what.

02The pulse survey policy text

[Company name]
Pulse Survey Policy
Version: [number]      Effective date: [date]

1. Rhythm and size
1.1 The company runs a short survey every [__] months, with no more than [__] questions, and it can be completed in under [__] minutes.
1.2 Once a year the company runs a wider survey, which does not fall in the same weeks as a short wave.
1.3 No new survey is sent before the results of the previous wave have been published and management has replied.

2. Responsibility
2.1 HR designs the questions, launches the wave and presents the results, and senior management approves the questions before launch.
2.2 No other party takes part in processing the answers.

3. Confidentiality
3.1 A survey is anonymous unless it is announced otherwise before it opens, and in an anonymous survey the company does not keep the identity of the person who answered.
3.2 The result of a question or group is not shown if fewer than [__] people answered in it, so that the people behind it cannot be worked out by subtraction.
3.3 No attempt is made to find out who gave an answer by any means, and whoever does so breaches this policy.
3.4 Employees are told before each wave that taking part is optional and that an answer has no effect on their appraisal.

4. Questions
4.1 Questions are phrased neutrally with no hint of the answer, and none asks about two things at once.
4.2 No question asks what would identify the respondent in a small group (for example the job title of a post held by one person).
4.3 The core questions stay the same between waves so they can be compared, and two or three are added as needed.

5. Opening the survey and taking part
5.1 A survey is open for [__] days, and employees are reminded once, halfway through.
5.2 The time taken to answer counts as working time, and nobody is obliged to answer.
5.3 HR measures the response rate and does not chase non-respondents one by one.

6. Results and reply
6.1 Results are published to employees within [__] days of closing: the response rate, the average of each question and the share of each answer.
6.2 Management commits to [__] action items, each with an owner and a date, and announces them with the results.
6.3 The next wave shows what changed because of the previous one.

7. Free-text comments
7.1 Free-text comments are read as groups of themes and are not published verbatim if they would reveal the writer.
7.2 If a comment contains a serious allegation (harassment or a financial violation), HR refers the matter to the route that handles it without trying to find out who wrote it, and invites employees to use the complaints route.

8. Records
8.1 The questions of each wave, the statistics of the results, the action plan and its follow-up are kept for [__].

Approval of the policy
[Approver's name and title]      Signature: ________      Date: [date]

--------------------------------------------------

Short survey

Wave: [number]      Period: from [date] to [date]      Audience: [everyone / branch / department]
This survey is [anonymous]. Taking part is optional and will not affect your appraisal.

1. I understand what is expected of me in my work.        [1 2 3 4 5]
2. My direct manager listens to my views.                  [1 2 3 4 5]
3. I have the tools I need to do my job.                   [1 2 3 4 5]
4. I see room to grow in this company.                     [1 2 3 4 5]
5. I would recommend a friend to work here.                [1 2 3 4 5]
6. What one thing, if it changed, would improve your work? [free text]
(1 = strongly disagree, 5 = strongly agree)

03Notes on the clauses of the policy

Clause 1: short and regular beats long and rare

Five questions that take two minutes get an answer from most employees, while forty questions are finished only by the angry and the enthusiastic. Comparison from one wave to the next is what gives the numbers their value, which is why the core questions stay fixed. For the detail see the employee satisfaction survey.

Clause 3: confidentiality is shown by action, not by words

An employee believes in confidentiality when he sees that results are not shown for a small group and that his manager does not call him after the survey to ask "was it you?". The rule of a minimum number of respondents guards against leaking by subtraction: if three of four in a department answered, the department's result exposes the fourth. The program applies a minimum participation to an anonymous survey and does not show its results below it.

Clause 4: a neutral question gets a useful answer

"Don't you agree that the manager listens to you?" hints at the answer. "My direct manager listens to my views" is a sentence the employee agrees or disagrees with. A question that joins two things ("pay and benefits") yields an answer nobody can attribute to either.

Clause 5: taking part without pressure

Chasing non-respondents by name turns an anonymous survey into a loyalty test. What you want is an overall response rate, a general reminder and an explanation that persuades employees their answer is read.

Clause 6: the reply is what keeps the survey alive

The biggest cause of falling response is that the employee sees nothing change. So results are published with small action items, an owner and a date, even if the item is only "we are reviewing the break schedule at the southern branch". For the detail see giving employees feedback.

Clause 7: the free comment is both a source and a risk

Comments are the richest part of a survey, and an employee may write in them what he would not say openly. They are gathered into themes, nothing that reveals the writer is published, and a serious allegation is sent to its own route.

04More articles

A short survey is launched every 3 months, the first wave on 2027-02-07, then 2027-05-07, 2027-08-07 and 2027-11-07. Each stays open 14 days, so the first closes on 2027-02-21 and the results are published 7 days later, on 2027-02-28. 10 employees were invited and 8 answered. One five-point question had the answers in the table, and the summary uses the program's question-summary function:

ScoreRespondents
5 strongly agree2
4 agree3
3 neutral2
2 disagree1
1 strongly disagree0

The response rate is 80% (8 of 10) and the average of the question is 3.75 out of 5. The share of favourable answers (4 and 5) is 62.5% and of unfavourable ones (1 and 2) is 12.5%. These figures are read with the previous wave, not on their own. These are illustrative inputs. The figures come from the program's survey-summary function.

05When this policy is not applied

An individual complaint

Someone who wants to challenge a decision about him does not do it in a survey but through the complaints and grievances route, where he is answered by name.

Appraisal and promotion

Survey answers do not enter any person's appraisal, promotion or penalty.

A very small company

In a department of three employees survey confidentiality cannot hold against subtraction, so its result is merged with another department's or is not shown in detail.

06Difficult cases and common mistakes

A bad result about one particular manager

It goes to the manager's own manager and to HR, and is discussed with the manager as a general trend of the department with ideas for improvement, not as testimony against his team members, and the team is never asked "who said it".

Employees ask for a survey with names

A survey can be non-anonymous if that is announced before it opens and for a clear purpose, such as choosing a date or recording a view between two options. An anonymous survey is never converted into an open one after it has started.

Common mistake: a survey with no reply

The worst outcome is that employees are asked and then see nothing. If the company is not ready to act on the results, it is better not to ask.

Common mistake: reading the average without the distribution

An average of 3.5 may mean everyone is neutral or that half are content and half are angry. Look at the spread of answers before you decide.

Common mistake: waves with no gap

A survey every month repeats the same questions and tires employees. A rhythm of every three months or longer suits better unless an event calls for an urgent question.

07Steps to adapt the policy to your company

  1. 1

    Choose the core questions

    Five to eight covering clarity of role, the relationship with the manager, tools, growth and recommending the company, and keep them fixed.

  2. 2

    Set the rhythm and the length

    A wave every three months, for example, open two weeks and published a week after closing.

  3. 3

    Choose a minimum number of respondents

    A number below which no result is shown, and remove any question that exposes respondents in a small group.

  4. 4

    Write the reply template

    One page showing what you said, what we will work on, who is responsible and when.

  5. 5

    Announce the policy to employees

    Before the first wave, with a notice that explains confidentiality, that taking part is optional, and why the surveys are run.

08Pulse survey policy checklist before approval

9 items

09Frequently asked questions

  • Five to ten, answered in under five minutes, and usually one free-text question. The shorter it is, the higher the response.

  • Make it anonymous unless you have a clear purpose that needs names, and never turn an anonymous one into an open one once it has begun. Confidentiality increases honesty on sensitive subjects.

  • There is no single rate right for every company. Watch the trend between waves, and if it falls ask whether people saw a reply to the previous wave.

  • Publish the response rate, the averages, the shares and the action items. Publishing free comments verbatim needs anything that reveals the writer removed.

  • That is an exit interview held at resignation with different questions, and it is not a recurring wave. For the questions see ten exit interview questions.

  • You can compare the trend of satisfaction with the rate of resignations over the same period, without claiming one causes the other. To read more see resignation reasons and retention.

Related pages

Inside TAC HR

Pulse surveys in TAC HR

In the «الإعلانات والاستبيانات» (announcements and surveys) screen of the employee requests service you build a survey with sections and questions of eight types: single choice, multiple choice, yes or no, a scale from 1 to 5 or 1 to 10, ranking, date, number and free text. You set its audience, all employees or one branch or department, add an optional end date, and switch on the anonymous mode. The employee answers from his phone. In anonymous mode the person behind an answer is not stored at all, and what is kept is only who took part without his answer, so you see the number of participants, and the screen does not show the results of an anonymous survey until participation reaches a minimum. Results come as a summary for each question: the spread of answers and the average of a scale. Management publishes the action plan in the «المجتمع» (community) section as an announcement with a read receipt showing who has seen it. You will find other templates on the templates page.

  • Employee requests — An employee request system with an approval route per request type and a leave balance checked before approval