The Cost of Miscommunication: What You Can Prove, and What You Cannot

By Nicolas Saliba · 8 min read

Every budget cycle, a version of the same slide circulates: poor communication costs large organisations some large sum per employee per year. It is a comforting number if you run internal communications, and it is the number most likely to end your meeting badly. A finance director will ask where it came from. The honest answer is usually a self-reported survey of a few hundred managers, extrapolated across a headcount that has nothing to do with yours, some years ago.

The instinct is to find a better statistic. The better move is to stop making a claim that size and start making one you can defend. This article is about which claims those are, and what has to be true for them to hold.

Why the big numbers do not survive contact with finance

The circulated figures share a construction, and once you see it you cannot unsee it. Someone surveys a sample of employees or managers and asks them to estimate hours lost to unclear communication. Self-reported hours are multiplied by an average salary. That per-person figure is then multiplied by a headcount, often a national or global one, to produce a headline.

Three things go wrong in that chain. Self-estimated time loss is not measured time loss, and people are poor at both recall and attribution. The average salary applied is rarely the salary of the people who reported the loss. And the multiplication treats every hour as fully recoverable, as though perfect communication would return all of it as productive output, which no organisation has ever observed.

None of that makes the underlying concern false. Communication failure genuinely costs organisations money. It makes the specific figure unusable in a room where someone does arithmetic for a living. If your credibility rests on a number that collapses under one question, you have handed the sceptic in the room the win.

The attribution problem, stated properly

Here is the uncomfortable part, and being straight about it is what earns the rest of the argument.

Communication sits upstream of almost everything an organisation does, which means it is confounded with almost everything. When a business unit that communicates well also performs well, at least four explanations fit the data equally. Communication improved performance. Good performance made communication easier, because there was good news to share. A capable leadership team produced both. Or the unit’s work is simply easier to communicate about.

Gallup’s meta-analytic work on engagement and business outcomes is careful about this distinction, reporting associations across large numbers of business units rather than asserting simple causation from any single case^[1]. That is the right posture. It is also more modest than the claims usually made on internal communications slides.

A second problem compounds it. The measure of “communication quality” is itself usually a survey question answered by the same people whose engagement you are correlating it with. Two survey items answered in the same ten minutes by the same person will correlate. That correlation is not evidence of business impact, and the article on engagement surveys versus communications data works through how to use each for what it is good at.

The claims that do hold up

Narrow the claim and the evidence gets stronger fast. A defensible claim has three properties: a named mechanism, a cost that someone outside communications already tracks, and a comparison that is not simply before and after.

Claim	Evidence required	How strong it is
Poor communication costs the group millions a year	None available	Not defensible. Do not use it.
Communication drives engagement, which drives performance	Correlational survey data	Weak on its own. Reasonable as context, not as proof.
An ambiguous instruction generated 340 extra service desk tickets in a fortnight	Ticket categories before and after, plus the clarified reissue	Strong. Named mechanism, cost already tracked by someone else.
The policy took 41 days to reach 90 per cent acknowledgement, and the tail sat in two sites	Acknowledgement workflow joined to site	Strong. Directly measured, and it points at an action.
Sites that received the manager briefing pack reached compliance 11 days sooner	Staggered rollout across comparable sites	Strong, if the sites really are comparable.
A safety instruction did not reach the night shift at two plants	Reach by shift and site	Strong, and this one rarely needs a monetary figure at all.

The pattern in the strong rows is that the cost is already being counted by someone else. Service desk volume belongs to IT. Compliance lag belongs to legal or risk. Rework belongs to operations. When you borrow a number that another function already defends, you inherit its credibility instead of having to manufacture your own.

Designing so the comparison means something

Before and after is the weakest design and the most commonly used. Three practical alternatives get you a long way without pretending to run a clinical trial.

Stagger the rollout. If a manager briefing pack goes to twelve sites in March and twelve in May, the May cohort is a usable comparison group for eight weeks. This is the most under-used technique in internal communications and it costs nothing.
Pre-register the intended audience. Write down, before publication, who the message is for. Without that, reach percentages are calculated after the fact against whoever happened to see it, which is not a measurement.
Hold a segment back briefly. Not always ethical or possible, and never for safety or compliance content. Where it is legitimate, a two-week holdout on a non-critical campaign produces the cleanest evidence you will ever get.

Be honest about what these designs cannot do. Sites differ. People talk to each other across the comparison boundary. A staggered rollout tells you about the eight weeks, not about the year. Stating those limits yourself is what stops someone else stating them for you.

The measurement you need underneath all of it

Every strong claim above depends on one variable that most organisations cannot produce: who was actually reached, described by the attributes that matter to the business. Not how many views the page got. Which functions, which sites, which shifts, which contract types.

Native tooling in Microsoft 365 does not hold this. SharePoint site analytics reports on the site, and Viva Engage community insights report on the community, both honestly and both for free, and for a single-surface question they are sufficient^[2]. What they do not do is join activity to organisational structure, which is precisely what turns “3,600 views” into “the night shift at two plants did not see it”.

Gallagher’s State of the Sector research has repeatedly found measurement to be the capability internal communicators name as their largest gap, and the reason is structural rather than a matter of skill^[3]. Deloitte’s human capital research points the same way, describing organisations that hold rich workforce data but struggle to connect it across functional boundaries^[4].

This is the layer Tryane occupies. It reads SharePoint, Viva Engage, Teams and newsletter platforms together, joins that activity to your organisational structure through Entra ID or an HR file, and holds history across product transitions so a multi-year claim is reproducible. SOC 2 Type 2 certified, GDPR compliant by design, EU hosting by default with US data residency on request, single sign-on through Azure AD or Entra ID, no agent to install, deployment in a couple of hours. For the metrics themselves, the guide to internal communication KPIs and the note on calculating internal communications ROI set out what belongs on the slide.

What to say when you are asked the cost

The sentence that works is smaller than the one you wanted to say, and it lands harder. Something close to: “I cannot tell you what poor communication costs this group, and I would be sceptical of anyone who claims they can. What I can tell you is that the reissue of the January instruction generated 340 additional service desk tickets, that we reached 90 per cent acknowledgement eleven days faster in the sites that got the manager pack, and that we currently have no visibility of the night shift at two plants. Here is what I want to do about the third one.”

That answer concedes the thing that was never provable, delivers three things that are, and ends on an action. It is also the answer a senior finance person is most likely to respect, because it sounds like someone who understands evidence. The article on proving internal communications works to leadership develops the wider case, and the benchmarks piece covers what a reasonable comparison actually looks like.

Frequently asked questions

Is there a reliable figure for the cost of poor internal communication?

Not one that will survive scrutiny. The circulated figures come from self-reported survey estimates multiplied by headcount, and each step in that chain adds error. A specific cost tied to a specific failure, using a number another function already tracks, is far more useful and far more defensible.

Does that mean internal communication cannot show a return?

It means the return is shown at the level of specific outcomes rather than as a single organisation-wide figure. Time to compliance, avoided rework, reduced repeat queries and reach into hard-to-reach populations are all measurable, and several of them convert to money using rates finance already accepts.

Are engagement survey results useless for this?

No, but they answer a different question. Surveys tell you how people feel and are subject to recall and mood effects. Communications data tells you what actually happened. Used together, the survey explains the behaviour and the behavioural data grounds the survey. Used alone, either can mislead.

What is the single most useful thing to start recording?

The intended audience for each message, written down before it is published. Almost no organisation does this, and without it every reach percentage is calculated after the fact against whoever happened to arrive, which is not a measurement of reach at all.

How do we handle a leadership team that wants the big number anyway?

Give them the credible small numbers alongside a plain statement of what cannot be proven. In our experience the specific claims land better, because they are checkable, and because they come with an action attached rather than a plea for budget.

Sources

• Gallup, employee engagement and business outcomes

• Microsoft Learn, SharePoint site usage and portal health

• Gallagher, State of the Sector

• Deloitte, Human Capital Trends