Why staff need a policy and what it is not
Everyone who works for an organisation is, to some degree, visible on social media as someone who works there. Most of the time that is good for the organisation: people talk about their work with pride, answer questions in their field and bring the brand into conversations it could never start itself. Problems come from a small number of predictable situations. A photo taken at work shows something it should not, such as a customer, a patient, a screen or a document. A member of staff praises a product without mentioning that they work for the company that sells it. Someone replies to an angry customer from their personal account and the reply becomes the story. A journalist messages a junior employee and quotes the answer.
A policy exists so that people know, before any of those situations, what the organisation expects and where to turn. The personal branding guide tells professionals posting in their own name to respect confidentiality, client agreements and any social media policy; this tutorial writes that policy.
It is worth being clear about what the policy is not. It is not the voice guide: how the official accounts sound and how often they post is the job of the brand voice and cadence guide. It is not the playbook for replies, moderation and complaints, which the community management guide covers. It is not a list of everything staff must never do, which nobody reads twice. And it is not an employment contract or a disciplinary procedure. How an employer may act in relation to staff accounts, and what happens when the policy is broken, depend on the law where the organisation operates and on its own employment arrangements; those questions belong with the employer's own HR and legal advisers and are outside this tutorial. The strategy section places the policy alongside the other decisions about who speaks for the brand.
Before you start
- The voice guide for official accounts. The policy points to it rather than repeating it, so it should exist first; the brand voice and cadence guide builds one.
- A list of the official accounts and who runs them. The tutorial on securing social media accounts builds an account registerThe brand's list of every account it holds on each platform, dormant ones included, with handle, use, status, admins, attached email and phone, and owner.; the policy names which accounts are official and who may post from them.
- The situations that have already happened. Ask managers, the social team and whoever handles press for the posts, replies and approaches that caused trouble or uncertainty. They are the best source of examples.
- The approval route. Know who approves what on the official accounts; the approvals and compliance reviews guide sets an approval matrix by risk for marketing generally.
- A named owner and a RACI. Use a RACI matrix to name who is responsible for drafting the policy, who is accountable for it (usually a senior marketing or communications lead), who is consulted (HR, legal or compliance, and representatives of staff) and who is informed (everyone it covers).
- A related policy to borrow structure from. If the organisation has one, the AI use policy shows the same approach: decisions written down briefly, with the method kept in linked guides.
The steps
Decide who the policy covers
Write down who the policy applies to before writing any rule. It should cover everyone who could be seen as speaking for or about the organisation: employees, contractors and temporary staff, and people at agencies or freelancers who have access to the official accounts. Say whether it covers volunteers, board members or ambassadors, if the organisation has them. The point of stating it is practical: an agency account manager posting from the brand's account and a part-time member of staff posting from their phone need different parts of the policy, and each should be able to find the part that applies to them. Keep the scope to social media and similar public online spaces, such as forums, review sites and community groups, and leave messaging between colleagues to other policies.
Separate official accounts from personal ones
The most useful line in the whole policy is the one between official and personal accounts. Official accounts are the ones the organisation owns and runs; only the named people in the account register post from them, and they post in the brand's voice under the voice guide and the approval route. Personal accounts belong to the people who hold them; staff speak for themselves, and the policy says so plainly. Ask staff who mention their employer in a profile to make clear that their views are their own, and ask them not to create accounts that look official, such as a location or team account using the brand's name or logo, without going through the people who run the official ones. A post on a personal account can still reach (ReachUnique people who saw an ad.) customers, colleagues and journalists, which is why the following steps cover the few things that apply to personal accounts too.
Write what staff may say about the organisation, with examples
Start with what is welcome, because most staff want to know what they can do, not only what they cannot. Sharing the organisation's public posts, talking about their own work, celebrating colleagues and answering questions in their field are usually all fine. Then set the few boundaries that matter: nothing confidential, nothing about customers, nothing that presents a personal view as the organisation's position, and no replies to complaints about the organisation from a personal account. Write each rule with a short example of a post that is fine and one that is not. Examples do more work than definitions: "a photo of the team at the stand, taken where no customer or screen is visible" teaches more than a paragraph about appropriate content. Where the organisation wants to go further and actively encourage staff to post, the guide to employee advocacy programmes covers the programme, and the policy stays the rulebook it points to. The employee advocacy launch checklist checks that every participant is pointed to this policy before the programme starts.
Require disclosure when staff promote the employer's products
When a member of staff recommends the organisation's products or services, readers are entitled to know they work there. The site's guides that cite the FTC's endorsement guidance, listed in this page's sources, state that a material connection with a business should be disclosed clearly and conspicuously; the influencer contracts and disclosure guide covers it for creators, and the reviews and social proof guide states that endorsements from staff need disclosure. In the UK, the advertising codes administered by the ASA require marketing to be obviously identifiable, and the CMA addresses hidden advertising, as the influencer contracts guide states. Staff praise is Social ProofEvidence that other people have chosen, used or approved something, used to reduce a buyer's doubt. (see the glossary entry), and it is only honest proof when the reader knows who is speaking. So the policy asks staff to say plainly that they work there when they promote what it sells, in the post itself rather than only in a profile, and treats any reward for posting, such as a prize, points or a bonus, as a further connection that needs disclosing too. This summary is educational, not legal advice; check the position in each market with whoever handles compliance.
Keep confidential information and other people's personal data out of posts
Two kinds of information stay out of every post, on official and personal accounts alike. The first is confidential information: anything the organisation has not made public, such as results, plans, prices before launch, client names under agreement and internal documents, including those visible in the background of a photo. The second is other people's personal data: customers, patients, clients and colleagues, their names, faces, details and stories, and photos taken at work in which they appear. Under the GDPR, listed in this page's sources, personal data needs a lawful basis, as the personal data in analytics guide summarises; the policy does not need to explain the law, only to set the rule that staff do not post other people's personal data and route any request to feature a customer or colleague through the official accounts. When customer posts are reshared by the brand, the permission process for customer contentContent about a brand made by customers or creators rather than by the brand itself. applies, which the official accounts handle, not individual staff. How the official accounts then ask colleagues to appear, record what they agreed to and act when someone leaves is set out in the guide to featuring employees in social media content.
Route customers and journalists to the right people
Staff will be approached on social media by people who think they can help. Customers with questions or complaints should be sent to the official channel, with a short, friendly line staff can use, rather than answered from a personal account; the official team can then help them properly under the community management guide's response standards. Journalists, bloggers and anyone asking for a comment on behalf of the organisation go to the named spokesperson or press contact, without a comment from the member of staff. Put the official channel and the press contact's name or role in the policy itself, so nobody has to search for them.
Name the escalation route for problems
Things will go wrong: a post from an official account that was a mistake, a member of staff who becomes the target of a hostile pile-on because of where they work, a customer posting a photo of an employee, a post someone regrets. Name one route for all of them: who to tell, how to reach them out of hours, and what to do in the meantime, which is usually to stop and not reply until the named person has seen it. Make clear that a member of staff who is being harassed because of their job should report it and will be supported, not judged. The community management guide sets out the escalation path for the official accounts; the policy connects staff to it.
Decide who speaks in a crisis
When something serious happens to the organisation, many people will want to help by posting. The policy should say that during a crisis only the named people speak for the organisation, that the official accounts carry the organisation's statements, and that staff may share those statements but should not add their own account of events, speculate or reply to criticism on the organisation's behalf. Name the role that decides when a situation counts as a crisis, and point to the organisation's crisis plan for who writes and approves the statements; the tutorial on writing a social media crisis plan builds one. The rule is simple because it has to be remembered under pressure: in a crisis, staff point people to the official statement.
Write it in plain language and test it on staff
Draft the policy in short sentences, in the second person, with the examples from the earlier steps. Keep it short enough to read in one sitting; put links to the voice guide, the approval route and the crisis plan at the end rather than copying them in. Then test the draft on a few people it covers, from different roles, by asking them what they would do in the situations collected before you started. Where their answer differs from what the policy intends, the policy is unclear, not the person. Send the tested draft for the legal and HR review named in the RACI; that review is the employer's own and this tutorial does not substitute for it.
Launch, train and set a review date
Publish the policy where staff will find it, introduce it in a short session or message that walks through the examples, and include it in onboarding for new staff and in briefing for agencies and freelancers with access to the official accounts. Ask managers to raise it when someone in their team starts posting about work more often. Set a review date and keep a short log of situations the policy did not answer; at each review, use that log, along with changes to the platforms the organisation uses and to its own structure, to update the rules and the examples.
When the steps are done, the finished document is short. The table lists its parts, what each says and where the detail lives, so that the policy stays brief and the linked documents carry the method.
| Section | What it says | Where the detail lives |
|---|---|---|
| Who it covers | Employees, contractors, agencies and freelancers with access, and anyone else named | The policy itself |
| Official and personal accounts | Which accounts are official, who runs them, and that personal accounts are the person's own | The account register |
| What staff may say | What is welcome, the few boundaries, and examples of each | The policy itself |
| Disclosure | Staff who promote the products say they work there; rewards for posting are disclosed too | The influencer contracts and reviews guides |
| Confidential and personal information | Nothing unpublished; no customers, patients, clients or colleagues without the official route | The official accounts' permission process |
| Customers and journalists | The official channel for customers; the named press contact for journalists | The policy itself |
| Problems and escalation | Who to tell, how to reach them, and what to do meanwhile | The community management guide's escalation path |
| Crisis | Only named people speak; staff share official statements | The crisis plan |
| Review | The owner, the review date and the log of unanswered situations | The RACI |
A home care provider's policy (illustrative)
A home care provider's policy
Corvessa Home Care, an invented home care provider with 140 carers and 12 office staff, has no social media policy until a carer posts a photo taken in a client's kitchen, with the client's medication and a care plan visible on the table. The post is removed within the hour, but the client's family sees it first. The operations director is made accountable for a policy, with the marketing coordinator drafting it and HR consulted.
The draft fits on 2 pages. It separates the 3 official accounts, run by the marketing coordinator and one office manager, from staff's personal accounts. It bans any post about clients or taken in a client's home, with 2 examples of photos that are fine (the team at a training day, a carer's own garden) and 2 that are not. It asks staff who share the provider's job adverts or services to add that they work there, and it names the operations director as the only person who answers press questions. Before launch, the coordinator tests the draft on 6 carers with 5 situations; 2 of the situations get the wrong answer, because the draft said "service users" where carers say "clients", so the wording changes. The policy goes into induction for new carers, and the first review is set for 12 months later.
Verify it worked
- Read the policy as an agency freelancer, then as a part-time employee. Each can find the part that applies to them.
- Check the account section. It names the official accounts and who may post from them, and says personal accounts belong to the person.
- Look at each rule. It has a short example of a post that is fine and one that is not.
- Look for the official channel, the press contact and the escalation route, including how to reach someone out of hours. All appear in the policy itself.
- Ask a few staff from different roles what they would do in the situations you collected. Their answers match what the policy intends.
If it goes wrong
- Testers give answers the policy did not intend. The wording is unclear, not the person. Reword it using the terms staff actually use, then retest.
- Staff ignore the policy or stop posting about work. It reads as a list of prohibitions. Add what is welcome, with examples.
- Staff praise a product without saying they work there. Disclosure sits only in a profile. Ask for the connection in the post itself, and cover rewards for posting too.
- Customers or journalists get answers from personal accounts. Nobody told staff where to send them. Put the official channel and the press contact in the policy.
- Photos taken at work show clients or screens. The examples are too general. Add examples from the organisation's own settings and route requests to feature people through the official accounts.
Next steps
- Read the employee advocacy programmes guide if you want to encourage staff to post, not only set their limits.
- Read the influencer contracts and disclosure guide to compare how disclosure is handled for creators with the wording you ask of staff.
- Read the AI use policy guide to write a companion policy in the same short style, with the method kept in linked guides.
Common mistakes
- A list of prohibitions with no examples. People cannot tell what is allowed, so they either ignore it or stop posting about work altogether. Show what is welcome and what is not.
- Blurring official and personal accounts. If the policy does not say which accounts speak for the organisation, every post by every member of staff becomes a possible statement.
- Forgetting disclosure. Staff praise without saying they work there reads as independent opinion. Ask for the connection to be stated in the post.
- Incentives without disclosure. A prize or bonus for sharing is a further connection, and the posts it produces need to say so.
- Photos taken at work. Most breaches of confidentiality and personal data start with a photo. Give specific examples from the organisation's own settings.
- No route for customers and journalists. Staff answer because nobody told them where to send people. Put the official channel and the press contact in the policy.
- Treating the policy as legal advice or a disciplinary code. It sets expectations for social media; the legal and employment questions belong with the employer's own advisers.
- Writing it once. Platforms, roles and situations change. Keep a log of what the policy did not answer and review it on its date.
