How to Launch an Employee Advocacy Programme

Plan the programme before you invite anyone

Agree the goal, the group, the rules, the topics and the starting numbers before launch. Programmes that start without them often see output drop within the first two months.

Build on original posts

Reshares of company posts add activity but rarely much reach. The posts that get read, and that AI engines cite, are written in the employee’s own words on their own profile. This shapes what you brief and how long each post takes to write.

Estimate reach before launch

In our own data the average active advocate has around 880 connections and publishes about three posts a month. Multiplied by the number of people you plan to start with, that gives a forecast you can present to your sponsor before committing budget.

Give the first 90 days one goal, one group and one review date

Tracking visibility, hiring and pipeline at once makes the results hard to read. With a single goal, the 90-day review is short and clear, which makes it easier to get the next group approved. Expect 12 to 18 months before the programme is fully established.

The failure pattern is consistent across companies with little else in common. Advocacy gets a slide at an all-hands, a dedicated channel, and a request to share the current campaign. Participation stays limited to a handful of people, most of them already working in marketing.

Within about six weeks the channel is quiet and the programme owner is chasing individuals by direct message. The usual explanation is low employee motivation.

The real cause is that the programme was announced without being set up: no agreement on its purpose, its participants, their subject matter, or the number that would show it had worked. Those questions then get answered later, under time pressure, by whoever happens to be in the thread.

Marketing leaders describe the symptoms in three ways, each tracing back to the same cause.

• LinkedIn output never gets past one or two committed people.
• Employees stop posting once the ideas run out.
• Leadership has no view of what is being published, by whom, or with what result.

A structured launch addresses all three, since each follows from a decision that was skipped.

You have decided to run employee advocacy. The evidence already settles three design decisions.

• Original posts. Semrush analysed 89,000 LinkedIn URLs cited by AI engines and found around 95% were original posts, according to Semrush’s study of cited LinkedIn URLsSemrushSemrush: We analysed 89K LinkedIn URLs cited in AI search (2026)Opens in a new tab. The programme therefore has to produce something of the person’s own, which changes what you brief and how long it takes.
• Personal profiles. Meltwater analysed 9.5 million AI citations and found roughly 75% of LinkedIn citations came from individual profiles, against about 25% from company pages, per Meltwater’s analysis of 9.5 million AI citationsMeltwaterMeltwater: 9.5M AI citations analysed, how LinkedIn content wins AI search (2026)Opens in a new tab. Resources go to the people, and the company page stops being the thing you optimise.
• Voluntary participation. LinkedIn sets three conditions for a programme that works, describing it as strategic, sustainable and organicLinkedInLinkedIn Marketing Solutions: What is employee advocacy and how do marketers succeed at it?Opens in a new tab, with taking part kept voluntary. Mandated participation produces posts with no readership, and makes the group harder to recruit a second time.

Two numbers from the same LinkedIn guidance are worth keeping for the business case you will need at the 90-day review: employees’ networks are collectively around ten times larger than a company’s follower base, and employee posts get roughly double the click-through rate of the same content posted by the company. Recruitment tends to be easier than expected as well, since the 2026 Edelman Trust Barometer puts “my employer” at 78% trust, the highest-scoring institution in the studyEdelman2026 Edelman Trust Barometer: “my employer” is the most trusted institution, at 78%Opens in a new tab.

Most programmes put this decision off. The goal determines who takes part, what they post about, and which numbers appear at the review.

Choose one of three goals. Each needs a different group, different posts and a different metric.

• Category visibility. You want your company named when buyers and answer engines describe your market, which is what AI search visibility covers. You start with your subject-matter experts, the posts take a clear position, and you measure reach plus how often the brand shows up in answers.
• Hiring. You want candidates to see what working here is actually like. You start with recent joiners and team leads, the posts are concrete about the work, and you measure applications and referral quality.
• Pipeline. You want the accounts sales is already working to see relevant people from your side. You start with sales and the leaders they sell alongside, and you measure meeting acceptance and earned media value against what the same reach would cost in ads.

All three are worth pursuing eventually. Pursuing all three in the first 90 days produces a review that supports no clear decision. Choose the goal your executive sponsor already cares about, state it explicitly, and treat the other two as secondary outcomes you report if they appear.

Write the goal as a single sentence you would be prepared to present at the 90-day review. If you cannot write that sentence yet, take two weeks to settle it before you launch.

Agree the time horizon with your sponsor in the same conversation. Ninety days is long enough to prove the mechanism works and to earn the next group, and too short to change how your market talks about you. Employee advocacy is a long-term programme, and the honest figure to put in front of leadership is 12 to 18 months before it is properly running. State it at the outset. A programme presented as a quarterly campaign will be assessed as one, and discontinued while the numbers are still small.

Pipeline is the slowest of the three to register. A reader who sees a post from one of your engineers in March may book a demo six months later without ever referencing it. That lag is normal, and it is why attribution has to be built in deliberately: UTM links on anything the group shares, a “how did you hear about us” field that somebody reviews, and a consistent habit of asking on sales calls. Set that up before the posts start, since the connection cannot be reconstructed once the deals are already in.

A company-wide invitation makes nobody in particular responsible for posting, and participation stays low as a result.

Start with a group small enough that you can name everyone in it, and select for four criteria.

• Something specific to say. Expertise the person already has, rather than a topic assigned to them.
• Honest feedback. People who will tell you which steps are getting in the way, rather than dropping out without explanation.
• Networks that barely overlap. Spread the group across departments, seniority levels and offices.
• At least one senior leader, publishing before anyone else is asked to. Visible participation from leadership is what makes the request credible.

Accept a declined invitation, and treat a hesitant yes the same way. Reluctant agreement rarely converts into published posts. Keep the first commitment small and bounded as well: an agreed number of posts over a set number of weeks is easier to accept than an open-ended arrangement.

Skip both extremes. Colleagues who already post every week do not need the programme, and colleagues with no interest in posting absorb a disproportionate share of your time. Start with the people in between, covered in lessons from companies empowering teams to post on LinkedIn.

With the names on a list, you can forecast reach. In our own data the average active advocate has around 880 connections, publishes about three posts a month, and draws 3 to 7% engagement on those posts, against 0.5 to 2% on a company page. Multiply those figures by the names on your list, and record the forecast in writing before launch. Track advocacy participation rate rather than total posts, so one prolific poster cannot hide a group that has gone quiet.

Decide what to call the programme internally before the first conversation. “Employee advocacy programme” is the category term used in marketing, and to the people being invited it reads as a policy rollout. Name it after the activity itself.

People rarely stop posting for lack of motivation. They stop because they have nothing specific to write about, and that is the easiest problem to fix.

Give each person a short list of subjects they already know something about, with a question against each one aimed at their actual work. “What did the customer say in that first call that changed how we scoped it?” produces a post where “post about our new release” does not. An open-ended request becomes a short, answerable one, and the answer is something only that person could have written.

A second source of hesitation is uncertainty about what employees are free to discuss. Settle that once, in writing, on a single page that marketing, legal and HR have all reviewed.

• Which subjects are open by default. Name them, including opinions the company is happy to be associated with. A short list of open subjects encourages writing; a long list of restrictions suppresses it.
• What to check before publishing. Usually customer names, unreleased product, financials, and anything regulated. Keep the list short, since every item adds delay.
• Who is responsible for what. Marketing owns the programme and the calendar. The employee owns their account, their voice and the decision to publish. Nothing reaches a personal profile without that person agreeing to it, which in Heyoo is a setting on their own profile rather than a rule they have to remember.

Get this right for the sake of the second group. If the first group feels that marketing is posting on their behalf without them, recruiting the next group becomes considerably harder.

“Marketing owns it” is not specific enough either. Name one person as programme lead, and choose someone suited to the work, since most of it consists of noticing a good post and acknowledging it the same day. Programmes with a named, responsive owner perform differently from those run out of a shared inbox.

In Heyoo, each person decides whether marketing may draft on their behalf, and whether anything drafted for them needs their approval before it publishes. The rules you wrote on the page become settings on the profile, so they continue to hold once the programme grows beyond the original group.

A group left to choose its own topics produces sound individual posts with no cumulative effect. One person writes about hiring, another about a conference, a third about a product release. Each post is fine on its own, and the company registers no particular area of expertise.

So decide, before you launch, on the few themes you want to be known for and what you think about each one. Content pillars give the group a shared subject. Points of view give them something to actually say about it, which is what turns a theme into a post somebody wants to read.

Three or four pillars are enough at launch. Beyond that, no single theme accumulates enough posts to become recognisable. Record three things for each one.

• The theme itself, in the words your buyers would use rather than your internal product names.
• The claim you are making about it, stated plainly enough that somebody could disagree.
• The claim you are arguing against, so the group knows which received wisdom they are pushing back on.

A pillar with no position produces descriptive posts that take no side, and those attract neither readers nor citations.

This is how advocacy campaigns work in Heyoo. Marketing sets the pillars and points of view once, then briefs a campaign against them. Everyone in the group receives a suggestion written in their own tone of voice and from their own angle, so ten people covering one theme produce ten different posts. Our own data puts the average time from opening one of those suggestions to publishing at around eight minutes.

Programmes usually lose their second budget because nobody recorded the starting position. The 90-day review can then report activity, but not improvement.

Four numbers, captured in week one, are enough.

• Company page reach over the last 90 days, so you can show employee reach next to it instead of on its own.
• The share of employees who posted anything on LinkedIn in the last month. This is your actual starting participation rate, and it is usually lower than expected.
• Branded search volume and direct traffic, which is where advocacy tends to show up first and where it is easiest to miss.
• How your brand is described when an answer engine is asked about your category. Save the answer so you can compare it later.

The fourth takes about twenty minutes and is the one most teams skip. Ask two or three answer engines the questions a buyer in your category would ask, and save the responses verbatim. A quarter later, that saved answer is among the most persuasive items in the review, whether it has changed or not.

Agree the reporting rhythm at the same time: monthly to the group, quarterly to the executive sponsor. Participation holds up where people can see their own numbers, and declines where the numbers sit in a spreadsheet that only marketing opens. Once posts are appearing regularly, the harder question is connecting them to revenue, which we worked through in making employee advocacy and thought leadership attributable.

In Heyoo this is what Post Analytics is for. Reach, engagement and participation are tracked per person, per team and per campaign, and traffic from advocacy posts is attributed with UTM links, so the monthly update to the group and the quarterly one to the sponsor come from the same place.

With the goal, the group, the rules, the pillars and the starting numbers settled, the launch itself is straightforward. It divides into three phases, each with a single objective.

Weeks 1 and 2: set up before inviting anyone

• Get the goal signed off by the executive sponsor, in the single sentence you wrote.
• Send the rules page to marketing, legal and HR and get it agreed.
• Record the four starting numbers.
• Speak to each person in the group individually. Ask what they would want to be known for.
• Load the pillars and points of view, and prepare enough material that nobody faces a blank page in week three.

Weeks 3 to 6: get everyone in the group to publish once

• Run one kick-off session for the whole group. Cover the goal, the rules, and what happens if a post is received badly, which is the question most participants have and few raise.
• Aim for one post per person in the first two weeks. How many people have published at least once is the only number worth watching in this phase.
• Ask the group to comment on each other’s posts. Early comments from colleagues make a first post considerably easier to publish.
• Resolve friction in the process in the week it appears. If a step takes a participant more than a few minutes, remove it or move it to marketing.

Weeks 7 to 12: settle into a rhythm and prepare the review

• Move to a regular cadence the group agrees to. Three posts a month each is a realistic steady state.
• Send the first monthly update to the group, with their own numbers in it.
• Select the next group from the people who asked to join after seeing colleagues post. That is the clearest available signal.
• Ask the group for anonymous feedback at the end of the twelve weeks: what helped, what got in the way, and what pace is realistic. Build the ongoing programme from those answers rather than from your own assumptions.
• Prepare the 90-day review against the numbers you recorded in week one.

The 90-day review decides whether the programme continues. Present three numbers, one example and one request.

• Three numbers, each shown against where you started: reach from the group next to company page reach for the same period, participation now against participation before, and the one number tied to the goal you picked in week one.
• One example. A concrete outcome that followed from a post: a candidate who mentioned it in an interview, a prospect who replied to a comment, a partner who made contact. Examples like this get repeated internally by executives, which charts do not.
• One request. The second group, named, with the resourcing it needs. Make the request in the same meeting, while the numbers are still on screen.

Between reviews, monthly reporting keeps the programme running.

Participants who can see what their posts achieved keep posting. Those who receive no feedback stop, and incentive schemes do not compensate for it. Report as far down this list as you can measure. Each step is harder to obtain than the one above it, and more valuable once you have it.

• Impressions and reach, per person and per team. The easiest figure to obtain, and the one participants care about most at the start, since it is their own.
• Comments and other interactions. Reach reports how many people saw a post. Comments indicate whether the argument registered, and with whom.
• Website traffic, tracked with UTM links on anything the group shares. This is the first advocacy number that reads as marketing rather than social media.
• Deals where a LinkedIn post is recorded as a source. Once sales records this, the programme carries a euro value, which makes it considerably harder to cut.

Do not wait until you can measure the bottom of the list. Report impressions from month one, then add each step as it becomes available. Steady month-on-month improvement convinces people more than sophisticated measurement.

Many programmes run their first months like this, and it works, though it takes a lot of time and is prone to errors. The person collecting the data has to chase every participant, and each employee has to report their numbers consistently, month after month. The sheet below comes from a company that collected its advocacy data manually in Excel to report results internally.

The effort scales with the number of posts, and the figures are only as current as the last manual update. The per-post detail that makes a report persuasive is also the slowest part to collect. Heyoo has official access to LinkedIn’s native analytics APIs, so the same post-level numbers arrive automatically and roll up across people, teams and campaigns.

Expect periods of inactivity. A participant who posted every fortnight and then stopped for six weeks has usually had a busy six weeks. Follow up, keep them on the list, and let them resume without a conversation about commitment. Removing people for inactivity turns a voluntary programme into one that can be failed, and discourages others from joining.

Put the reporting somewhere people already look. One slide in the all-hands is enough: what the group published, how far it reached, what it brought in, and who did it. Name the people.

That slide does three things. It shows the group that their work counted, it shows everyone else that employee posts contribute to business outcomes, and it recruits the next group, since people ask to join a programme that is visibly working.

Some of what the programme produces will never reach the report. Participants become more confident writing in public, and colleagues across departments start engaging with each other’s work. Report the numbers, since the numbers earn the budget, and track the rest, since that is what sustains participation.