A Guide to Internal Communication Analytics: The Four Questions Behind Every Number

By Nicolas Saliba · 8 min read

A CCO asks the Head of Internal Communication one question at the end of the quarter: is the intranet working? The team arrives with numbers. Two hundred and forty thousand page views, a 34 per cent newsletter open rate, 1,900 reactions across eleven Viva Engage communities. None of it answers the question, and everyone in the room works that out inside ninety seconds. The conversation drifts into anecdote, because anecdote is all that is left.

The numbers are not wrong. They are accurate measurements of things nobody asked about. Internal communication analytics breaks at the point where four genuinely different questions get flattened into a single figure, and the figure is usually whichever one the platform puts on its front page. This guide separates the four, says where each is answered inside Microsoft 365, and gives the order to build them in.

The four questions behind every internal communication number

Every metric an internal communication function produces is evidence for exactly one of four questions. They run in sequence, each harder than the one before it, and each close to meaningless without the one before it.

Distribution. Did the message leave the building? News posts published, emails sent, Teams posts made, communities posted into. This is production data. It measures your team, not your audience.

Reach. Did it arrive in front of the people it was written for? This is where a denominator appears, and where most reporting quietly gives up.

Engagement. Did anyone do anything with it beyond receiving it? Reading time, replies, reactions, click-through to whatever the message asked people to do.

Outcome. Did something change in the organisation because of it? Policy acknowledgement, form completion, near-miss reporting on a plant floor, movement in a pulse question about clarity of direction.

Nearly every unproductive measurement conversation has the same shape: someone offers evidence for question one while an executive is asking question four. Naming the question you are answering before you show the slide resolves more meetings than any dashboard redesign.

Question	What it measures	Where it sits in Microsoft 365	The usual failure
Distribution	Volume your team produced	SharePoint publishing history, Teams post counts, send logs	Presented as an achievement
Reach	Share of a named population exposed	Site usage reports, per-post views, email open data	No agreed denominator
Engagement	Depth of interaction, by population	Page analytics, reactions, replies, link clicks	Aggregates hide concentration
Outcome	Change in behaviour or belief	HR and operational systems, pulse surveys	Attributed to communication alone

Question one: distribution, the number that flatters your team

Distribution is the only question every internal communication team can already answer, which is why it dominates monthly reporting. It is not useless. Publication volume tells you whether a campaign actually ran, whether a site is being maintained or abandoned, and whether the German entity is producing local content or waiting for headquarters.

What it cannot do is stand in for effect. A slide reading “142 news posts published this quarter” invites exactly one response from a finance-minded executive, and it is not a favourable one. Use distribution as context underneath a reach figure, never as the headline. If you are building your first reporting pack from nothing, our note on where to start with internal communication analytics sets out the shortest path to something defensible.

Question two: reach, where the denominator decides the answer

Reach is the first genuinely useful measure, and the one most organisations get wrong for a structural reason: they have a numerator and no denominator. SharePoint site usage reports and per-post view counts give you unique viewers^[1]. They do not give you the population that post was written for, because Microsoft 365 does not know which population that was.

So 6,400 unique viewers becomes an impressive-looking number with no scale attached. Against a 7,000-person tenant it is remarkable. Against 48,000 it is a head-office readership. And the same figure hides the distribution question that matters most: whether those viewers are concentrated in three functions or spread across the sites you were actually trying to reach.

Fixing this is unglamorous. Agree a workforce file with HR, write down who is in scope and who is not, freeze the definition, and publish reach as a percentage of a stated population every time. The structuring work is covered in our piece on audience segmentation for internal communications. Everything above this line in the four-question sequence depends on getting it right.

Question three: engagement, the most abused word in the discipline

Engagement is used to mean three incompatible things: interaction with a piece of content, participation in a channel, and the psychological state Gallup measures in the workforce^[2]. Conflating the first and third is the single most common analytical error in this field, and it is worth reading the common mistakes in employee engagement analysis before you put an engagement score in front of a board.

Content engagement is a legitimate measure with a narrow meaning: of the people who received this, how many did something with it. Reactions and replies are weak signals, easily inflated by a few enthusiastic community managers. Reading time and click-through on the specific action a message requested are stronger, because they are harder to produce accidentally.

Read engagement by segment or do not read it at all. A 12 per cent aggregate interaction rate that turns out to be 31 per cent in corporate functions and 2 per cent in operations is not a mediocre result. It is two different results, and only one of them needs work.

Question four: outcome, attempt it last and attribute it honestly

Outcome is the question executives want answered and the one you should build last, because it is the only one where you cannot control the evidence. Communication is never the sole cause of an organisational change, and claiming otherwise costs credibility that takes years to rebuild.

The defensible version is narrow and paired. Choose one behaviour the communication explicitly asked for, measure it in the system that owns it, and show it next to reach in the same population. Completion of a mandatory training module, uptake of a benefits change, submission of safety observations at a specific site. When reach in a site rises and the behaviour rises with it, you have a correlation worth presenting and a caveat worth stating. Gallagher’s State of the Sector research keeps finding that measurement is the capability internal communicators name as their biggest gap, and this is the layer where the gap actually bites^[3]. Our guide to defining success before you measure it covers how to fix the objective first so the outcome question has something to point at.

What Microsoft 365 answers natively, and what it does not

Native analytics is free, already switched on, and genuinely sufficient for one team running one surface. A community manager who wants to know whether posts land inside their community should use Viva Engage community insights and stop there. A site owner tracking a single SharePoint hub gets what they need from site usage reports. Nothing in this article argues otherwise, and our walkthrough of reading SharePoint native analytics properly shows how much is available before you spend anything.

The native layer stops at the tenant boundary of each product. It does not join activity to organisational attributes, so the reach question cannot be answered by function, country, site or contract type without exporting and joining the data yourself. It reports per channel, so a person reached by the intranet and the newsletter and a Teams post counts three times and a person reached by none of them counts as nothing. Teams is a particular blind spot, covered in our piece on Microsoft Teams as an internal communication channel. And the retained reporting window is finite, which makes multi-year narrative difficult exactly when you want to show progress.

Teams often try to close the gap with a Power BI build on the Microsoft Graph reporting API. That can work, and it is more demanding than it looks. The reporting API is not consistently reliable and produces days with missing data. Customers have observed KPI values in Power BI that do not reconcile with the native SharePoint reports for the same period. A dependable build needs a robust ingestion process and scheduled reconciliation against native figures, so budget for the pipeline rather than the dashboard. The specific failure modes are in our note on SharePoint analytics in Power BI.

The order to build this in

Do not build all four at once. Agree the denominator, then publish reach as a percentage of a stated population for your five most important communications. That alone changes the quality of the conversation with an executive audience. Add segmented engagement next, because segmentation is what turns a number into a decision about where to act. Add cross-channel deduplication when your audience genuinely uses more than one channel, which our piece on cross-channel measurement works through. Attempt outcome last, on one narrow behaviour, with the attribution caveat written into the slide.

This is the layer Tryane occupies. It reads SharePoint, Viva Engage, Teams and your newsletter platform together, joins that activity to your organisational structure through Entra ID or an HR file, and keeps history across product transitions so a reorganisation does not reset your trend line. Tryane is SOC 2 Type 2 certified and GDPR compliant by design, hosts in the EU by default with US data residency on request, and deploys in a couple of hours over single sign-on with Entra ID, with no agent to install. For the metric set itself, our guide to internal communication KPIs lists what survives executive scrutiny.

Frequently asked questions

What is the difference between reach and engagement?

Reach is the share of a defined population that was exposed to a communication. Engagement is what the exposed population did next. Reach needs an agreed denominator and is harder to produce honestly. Engagement without a reach figure underneath it is uninterpretable, because a high interaction rate on a message almost nobody saw is not a good result.

Which single metric should we start with?

Reach against a stated population, for one important communication. It forces the denominator conversation with HR, it is understood by executives without explanation, and every more sophisticated measure depends on it. A fourteen-metric framework adopted in week one usually collapses by month three because half the metrics have no data source.

Can we answer all four questions with native Microsoft 365 analytics?

Distribution and per-channel engagement, yes. Reach against an organisational population, not without joining the activity data to workforce attributes yourself. Outcome sits outside the communication tools entirely, in HR and operational systems. Native reporting is the right answer for a single team on a single surface and runs out at the organisational question.

How long should our reporting history go back?

Longer than the native retention window, which is the practical problem. Native reporting keeps a finite period, so a comparison spanning a Yammer to Viva Engage transition or an intranet redesign is usually unavailable. A multi-year trend needs history held independently of the products that generated it.

Does Tryane replace our native analytics?

No. SharePoint, Viva Engage and Teams keep recording what they record. Tryane sits above them, reads them together with your newsletter platform, and joins the whole set to your organisational structure so the reach and segmentation questions become answerable. You keep the Microsoft 365 investment and add the measurement layer for what native does not cover.

Sources

• Microsoft Learn, SharePoint site usage report

• Gallup, State of the Global Workplace

• Gallagher, State of the Sector