How Much Reach Does Employee Advocacy Actually Add? The Maths

On this page

The multipliers on every other page cannot be applied to your company
The four numbers that decide the answer
The worked example: forty employees, twelve posters
Overlap is the biggest correction and every calculator drops it
The crossover: how many posting employees beat the company page
Where the model breaks
What to report instead of reach
When advocacy is not a reach play at all

The short answer

Advocacy reach is four numbers multiplied together: people who actually post, their median connections, the share of a network that sees a given post, and an overlap discount because your employees know each other. Run those and twelve weekly posters produce roughly 840 gross impressions a week on stated assumptions, about 460 unique people after overlap. Five weekly posters is where advocacy passes a two thousand follower company page.

The multipliers on every other page cannot be applied to your company

A multiplier without a denominator is not a number, it is a shape. The advocacy reach multipliers that appear on almost every ranking page do not state the sample size, the measurement period, the industry, or what they were measured against. Applying an unstated ratio to your own headcount gives you a forecast you cannot defend the first time somebody in finance asks a follow up question.

You have seen the figures. This page is not going to repeat them, and the reason is not squeamishness. The published multipliers disagree with each other, several of them trace back to the same small set of vendor analyses, and not one of the pages quoting them states how many companies were in the sample or over what period. A ratio in that condition cannot be multiplied by your headcount to produce anything meaningful.

What you can do instead is build the number from four inputs you can read off your own accounts in about twenty minutes. It produces a smaller figure than the vendor calculators do. It also produces a figure you can put in a board pack and still be standing behind in six months.

One unsourced multiplier costs you the whole report

If a number in your deck came from a vendor blog and you cannot name the sample, take it out before somebody else does. A finance director only has to catch one unsourced ratio to start discounting everything else on the page, including the numbers you measured properly.

The four numbers that decide the answer

Advocacy reach is four terms and one correction. Posters, connections, delivery rate, and an overlap discount. Every advocacy calculator on the internet uses the first three and drops the fourth, which is why their outputs are large and unusable.

The Supersonify Net Reach Model

Four inputs, read from your own accounts rather than borrowed from a vendor. Gross weekly impressions equal P times C times d times posts per person per week. Net unique people equal gross multiplied by one minus o.

P, the postersNot headcount, not enrolment, and not the number who attended the kickoff. The count of people who actually published something in the last thirty days. Around 3% of members post more than once a week according to aggregate 2026 LinkedIn statistics reports, which is the base rate you are recruiting against, so assume roughly a third of your enrolled list until your own data says otherwise.

C, the connectionsThe median first degree connection count across those posters, which you can read off their profiles in ten minutes. Use the median rather than the mean, because one hyper connected salesperson distorts the average, and that person is usually among the least likely to post consistently.

d, the delivery rateThe share of a poster's network that sees a given post. This is not a published constant and it moves with the post, the format and the day. Take it from your own advocates by dividing their impression counts by their connection counts. Until you have that, run the model at 5% and at 15% so you can see exactly how much of your answer is resting on a guess.

o, the overlap discountThe share of the modelled audience that is already counted somewhere else, because your colleagues are connected to each other and most of them already follow the company page. It is the largest correction in the model and it is absent from every calculator, which is the single reason those calculators produce numbers nobody in finance believes.

Two of the four inputs are facts you can look up today. One is measurable within a month of running the programme. Only the overlap discount has to be estimated at the start, and the section below gives you a way to stop guessing it.

The worked example: forty employees, twelve posters

Twelve weekly posters produce 840 gross impressions a week and about 462 unique people once overlap is applied. Over a year that is 43,680 impressions and 624 posts, which cost 312 hours of employee writing time at thirty minutes each. Here is every input, so you can replace them with yours and get a different answer honestly.

Input	Assumption used here	Replace it with
Headcount	40	Your actual headcount, which barely matters
Posters, P	12	People who published in the last 30 days
Median connections, C	700	The median across those twelve profiles
Delivery rate, d	10%	Their own impressions divided by their connections
Posts per person per week	1	Your real cadence. A fortnightly cadence halves everything below
Overlap discount, o	45%	Measure it with the method in the next section

The inputs. Every figure in the middle column is an assumption stated for the sake of the worked example, not a benchmark and not a measurement.

Line	Working	Result
Gross impressions per week	12 x 700 x 0.10	840
Gross impressions per month	840 x 4.33	3,637
Unique people per week after overlap	840 x 0.55	462
Gross impressions per year	840 x 52	43,680
Posts produced per year	12 x 52	624
Employee writing hours per year at 30 minutes each	624 x 0.5	312

The arithmetic, shown in full so you can check it rather than trust it.

Say the annual number out loud before you present it. Forty three thousand impressions a year is a small number in media terms, and anybody who has bought paid distribution will recognise it as small. That is the honest starting position, and it is a stronger place to argue from than a borrowed multiplier that collapses under one question.

The comparison that matters is not impressions against impressions. These impressions land in the feeds of people who chose to connect with a named individual, and the reply arrives in that individual's notifications where a human can answer it. LinkedIn's own targeting data says four out of five members drive business decisions, though that is a claim about the platform's total membership rather than about the particular seven hundred people your advocate happens to know.

The 312 hours is the other half of the answer, and it is where most of the real money sits. Software is usually the smallest line in an advocacy budget, which is examined in what advocacy software actually costs.

Overlap is the biggest correction and every calculator drops it

Overlap is the share of your modelled audience already counted somewhere else in the same model, and it is large in exactly the companies that run advocacy programmes. Your colleagues are connected to each other. They met the same customers, attended the same three conferences, and most of them already follow the company page.

There are four sources of it and they compound. Colleague to colleague connections, which are close to total inside a small company. Shared customers and prospects, which is highest among precisely the buyer adjacent people you selected for. Shared events, where a whole team connects with the same attendee list in one week. And page followers, since employees and their contacts are usually the first people to follow a company page.

Overlap does not reduce impressions. Ten people seeing a message ten times is still a hundred impressions. It reduces unique people, and unique people is what a reach claim actually asserts. Reporting gross impressions as reach is not a rounding error, it is counting the same person up to a dozen times and calling it an audience.

How to measure your own overlap in one afternoon

Pick three advocates from different teams

Different teams matters. Three people from the same team produce an overlap figure that flatters nobody and misleads everybody, because their networks were built in the same rooms.

Have each publish their own post on the same day

Same day, same topic, different words. Spread them across the day rather than the same hour, so you are measuring network overlap rather than one crowded slot in the feed.

Wait 48 hours, then list the reactors and commenters on each post

Both lists are visible on the post itself. Copy the names into three columns of a spreadsheet. This is the tedious part and it takes about twenty minutes for three normal posts.

Count the names appearing on two or more lists

Divide that count by the total distinct names across the three columns. That percentage is your engagement overlap, and it is the first real number you will have about your own network structure.

Treat it as a floor, not a measurement

Engagers are a biased subsample. They are the most loyal and most active part of each network, so they overlap more than the full networks do. Use the figure to sanity check the discount you assumed, and never present it as a measured audience overlap.

Do not report engagement overlap as network overlap

It is a proxy with a known and directional bias. Anybody who spots the difference in a review will discount every other number you presented, and they will be right to. State it as a floor, state the bias, and it becomes a credible input instead of a liability.

What to take away

The published advocacy multipliers cannot be applied to your company, because none of them states the sample, the period or the denominator behind the ratio.
Advocacy reach is posters multiplied by connections multiplied by delivery rate, then discounted for overlap, and the overlap term is the one every calculator quietly drops.
On the assumptions in this post, twelve weekly posters produce about 840 gross impressions a week and roughly 460 unique people after a 45 percent overlap discount.
Five weekly posters is where advocacy passes a 2,000 follower page posting three times a week, and 22 posters is where it passes a 10,000 follower page.
You can measure your own overlap in an afternoon by counting repeat names across three advocates' engagement lists, as long as you report it as a floor rather than a measurement.

The crossover: how many posting employees beat the company page

Five weekly posters passes a 2,000 follower page posting three times a week, on the assumptions used here. The formula is page followers times page delivery rate times page posts per week, divided by employee connections times employee delivery rate. Everything else is arithmetic.

The page assumptions in the table are a 5% delivery rate and three posts a week. Page delivery is set lower than the 10% used for individuals for a structural reason rather than a mystical one: a page post competes against people in the same feed, and a follow is a weaker relationship signal than a mutual connection. Change that assumption and the whole table moves, which is exactly why it is written down.

Company page followers	Page impressions per week	Weekly posters needed to match it	What it means
500	75	2	Two people outreach the page, so treat the page as a credibility surface rather than a channel
1,000	150	3	A small group still beats the page comfortably
2,000	300	5	The crossover most small B2B companies are actually sitting at
5,000	750	11	Eleven weekly posters needs funded time and a named coordinator
10,000	1,500	22	Twenty two weekly posters is a real programme with a real budget line
25,000	3,750	54	At this size advocacy adds to the page, it does not replace it

Crossover headcount by page size. Assumptions: page delivery 5% over three posts a week, employee delivery 10% over one post a week, median 700 connections. Rounded up to whole people.

Read the table the right way round. The crossover moves with your page, not with your headcount. A company with a weak page reaches it with five people. A company with a strong page never reaches it and should stop selling advocacy internally as a reach play, because the first person to run this calculation will find the same thing you just did.

The 54 figure at the bottom is the useful one. Sustaining fifty four weekly posters against a platform where around 3% of members post more than once a week is not something that happens by encouragement. It happens with funded time or it does not happen, which is the argument for running it without a platform at all, covered in advocacy without buying software.

Where the model breaks

Three inputs break it, and you should say so before somebody else does. The delivery rate is not a constant, the poster count decays, and impressions are an input rather than an outcome.

Delivery rate moves with the post, the format, the day and whatever the ranking model is doing this quarter. Never quote a single figure. Run the model at two rates and present the range, because a range that contains the truth is worth more than a point estimate that does not.
Poster count decays. Enrolment is not participation, and participation falls sharply around month three unless the thirty minutes is funded and protected. Re-count P every month rather than reusing the number from the launch deck.
Impressions are an input. Nobody has ever been invoiced for one. The model sizes the opportunity, and it should never be the last slide.

36%

year on year growth in video watch time on LinkedIn, which is a concrete reminder that format mix shifts delivery rates. A delivery figure you measured on text posts last year is not the figure your model needs this year.

LinkedIn via Search Engine Journal, 2026

The practical response to all three is the same. Recalculate quarterly with your own current numbers, keep the assumptions visible in the same table as the results, and let the model shrink when the inputs shrink. A forecast that never moves is a forecast nobody is checking.

What to report instead of reach

Report unique people reached with the overlap discount shown, posts published, and conversations started, then convert to cost per conversation. Cost per conversation is the only advocacy line that sits beside a paid media line without needing a translator, which is why it is the one that gets a second year funded.

Metric	Where it comes from	Does it survive a budget review
Gross impressions	Platform analytics, summed across advocates	No. It double counts people and everybody in the room knows it
Unique people after overlap	This model, with the discount stated	Yes, because you disclosed the correction before anybody asked for it
Posts published	A manual count	Yes. It is an output and nobody can dispute a count
Conversations started	Advocates log inbound messages referencing a post	Yes. It is the number sales also recognises
Cost per conversation	Programme hours times blended rate, divided by conversations	Yes, and it is directly comparable to your cost per lead on paid
A multiplier versus the page	A vendor blog	No, and it discredits the rest of the page it appears on

Which advocacy metrics survive a budget review and which ones damage the report that contains them.

Cost per conversation also makes the channel comparison honest in both directions. Some quarters advocacy will lose that comparison, and knowing which quarters those are is worth more than a metric that always looks good. The same discipline applied across channels is covered in LinkedIn ads against Google ads.

When advocacy is not a reach play at all

If your page is strong and your headcount is small, advocacy will not add meaningful reach and you should stop selling it internally as though it will. The crossover table above tells you that in one line, and finding out from your own arithmetic is considerably cheaper than finding out from a sceptical CFO.

It still earns a budget line, for a reason that has nothing to do with volume. A message carried by a named person with a job title can be replied to, argued with, and asked a follow up question. A page post cannot do any of those things in a way that a buyer finds credible, because nobody believes they are talking to a company.

That is also where the value concentrates. 76% call LinkedIn the most effective channel for thought leadership, according to Content Marketing Institute data cited in 2026, and thought leadership is carried by people with names and track records rather than by logos. Advocacy is a credibility mechanism that produces some reach, not a reach mechanism that produces some credibility, and the two get budgeted very differently.

So use the model for sizing and not for selling. Size the programme, show the assumptions, state the crossover, and let the reach number be as small as it honestly is. The programmes that survive their second budget review are the ones that never over-claimed in the first.

Questions people ask next

What is a realistic reach number for employee advocacy?

It depends on four inputs, and any figure quoted without them is decoration. Twelve weekly posters with a median 700 connections and a 10% delivery rate produce about 840 gross impressions a week, or roughly 460 unique people after a 45% overlap discount. Run it with your own numbers before promising anything.

Why do employee advocacy calculators produce such large numbers?

They multiply headcount by average connections and skip two corrections. They count enrolled employees rather than people who actually post, and they apply no overlap discount, so the same colleague is counted once for every advocate connected to them. Fixing both usually cuts the output by more than half.

Does employee advocacy reach more people than a company page?

It depends entirely on the size of your page. On the assumptions in this post, two weekly posters beat a 500 follower page, five beat a 2,000 follower page, and it takes 54 to beat a 25,000 follower page. Calculate your own crossover before framing advocacy as a replacement for the page.

How do I measure the overlap between employee networks?

Have three advocates from different teams post on the same topic on the same day, then list everybody who reacted or commented on each post and count the names appearing on two or more lists. That percentage is a floor rather than a measurement, because engagers overlap more than full networks do.

How many employees do I need before advocacy adds meaningful reach?

There is no headcount threshold, only a ratio to your page. Two posters can beat a small page and twenty two are needed to beat a page with 10,000 followers. Count people who published in the last thirty days rather than people who enrolled, since enrolment overstates the real figure substantially.

Sources