A community manager in a 9,000-person group looks after fourteen Viva Engage communities. Three are genuinely alive. Six tick over. Five have not had an original post in four months, and two of those five were launched by a director who still asks about them. She has no way of knowing which of her posts last quarter did anything, so she posts on the same schedule she has always used, into the same communities, at the same time of day.
That is the situation analytics is supposed to fix. It can, but not in the way most vendor material implies. Analytics does not raise engagement. It changes what the community manager does on Monday morning, and the change in engagement follows from the change in behaviour. The distinction sounds pedantic until someone asks you to prove the number.
Why a big uplift percentage should prompt a question
Internal communication is full of claims like “engagement rose by half after we introduced analytics”. Some of them describe something real. Most of them are unfalsifiable, and it is worth being precise about why, because the same weaknesses will appear in your own reporting if you do not guard against them.
Start with the denominator. If your engagement rate is active users over total members, then archiving five dormant communities removes their inactive members from the denominator and the rate rises without a single additional person doing anything. Membership numbers in Viva Engage move for administrative reasons all the time: group lifecycle policies, a tenant merge, a change to who is auto-enrolled[1]. A rate whose denominator is not fixed is not a measurement.
Then the control. Almost nobody in internal communication runs a holdout group, and for good reasons: withholding a safety message from a randomly selected quarter of a workforce is not an acceptable experiment. Without a control, a rise after an intervention is a rise after an intervention. It is not evidence that the intervention caused it.
Then everything else that moved in the same period. A reorganisation announcement, a results release, a redundancy consultation or a single unusually good CEO post will each move a monthly engagement figure more than a quarter of careful community management will. Seasonality alone is substantial in most organisations: August and late December do not resemble the rest of the year.
None of this means measurement is pointless. It means the honest claim is narrower and, in front of a sceptical executive, considerably more durable.
What analytics actually changes about the week
The value shows up as four specific behaviour changes, all of which are observable and none of which require an uplift claim.
It stops the posting into dead communities
The most common finding when a community manager first sees activity concentration is that a large majority of interaction happens in a small minority of communities. That reallocates effort immediately. Time spent producing content for five communities that nobody reads is time not spent in the three that carry the organisation’s actual conversations, and the native Viva Engage activity report will show you the concentration if you look for it[2].
It moves the posting time to the audience, not the calendar
Most internal posts go out when the communicator has finished writing them, which is late morning in the head-office time zone. In an organisation with shift patterns or several time zones, that is systematically the wrong moment for a large part of the workforce. Our note on identifying the best time to post on Viva Engage covers how to work this out per population rather than globally.
It finds the responders, not the readers
In a functioning community, a small group of people answer other people’s questions. They are usually not the people the communications team knows, and they are almost never the ones with the most senior titles. Identifying them, then supporting them, does more for community health than any campaign. This is the practical core of the argument in our piece on community managers and Viva Engage performance.
It gives you permission to close things
Closing a community that a director sponsored is a political act. Doing it with three quarters of activity data is a much smaller political act. The article on creating relevant Viva Engage communities is the constructive counterpart: fewer, better-scoped communities generally outperform a long list.
Claim versus evidence
| What people claim | What you can actually evidence |
|---|---|
| Analytics increased engagement by half | Activity rose in a period during which several things changed, including the analytics |
| Our engagement rate improved | The rate improved, and the denominator also changed. Both need stating |
| The campaign worked | The target population’s participation rose relative to a comparable population that was not targeted |
| Employees are more engaged | Participation in a channel rose. Engagement at work is a different construct, measured by survey |
| Posting at the new time is better | Posts at the new time reached a measurably larger share of shift populations across a full quarter |
What honest attribution looks like without a control group
You cannot randomise, but you are not left with nothing. Three approaches are defensible in front of a finance director.
Use a comparison population. If a campaign targets manufacturing and not the commercial function, the commercial function is an imperfect but real comparison. Movement in the target population that does not appear in the comparison population is the closest thing to evidence most internal communication teams will ever have.
Fix the denominator before you start, in writing. Decide what population each rate is measured against and where that population comes from, and do not change it mid-year. Most reconciliation arguments are denominator disagreements.
Report the gap, not the total. The measure that survives scrutiny is the distance between your best-served and worst-served populations. Gallup’s long-running workplace research has consistently found engagement to be weakest among frontline and deskless workers[3], and Gallagher’s State of the Sector research repeatedly identifies measurement as the capability internal communicators name as their biggest gap[4]. A narrowing gap is harder to produce accidentally than a rising total, which is precisely what makes it worth reporting.
What does realistic improvement look like? Honestly, the change teams describe most often is not a larger headline number. It is that a community that was invisible to a whole region stops being invisible, and that the monthly report starts prompting decisions rather than nods. If a supplier offers you a specific percentage before seeing your tenant, ask which denominator they used.
Where the native tooling stands
Viva Engage community insights are free, already switched on, and sufficient if you manage a single community and want to know whether your posts are landing inside it. For that job there is nothing to buy.
The limits appear when the question becomes organisational. Native insights do not join to HR attributes, so participation by function, site, language or contract type is out of reach without exporting and matching the data yourself. The retained reporting window is finite, which makes any trend spanning the Yammer to Viva Engage transition difficult to show. And Viva Engage activity sits apart from your SharePoint intranet and newsletter figures, so cross-channel reach cannot be assembled from it. The detail is in our note on the limits of Viva Engage analytics, and the background to the product transition is covered in what Yammer changed and what Viva Engage inherited.
Teams often try to close the gap with a custom Power BI build against the Microsoft Graph reporting API. That can work, with eyes open. 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 the native figures, with someone who owns it. Budget for the pipeline rather than the dashboard.
This is the layer Tryane occupies. It reads Viva Engage alongside SharePoint, Teams and your newsletter platform, joins that activity to your organisational structure through Entra ID or an HR file, and keeps history across product transitions so the Yammer years and the Viva Engage years read as one trend. Tryane is SOC 2 Type 2 certified and GDPR compliant by design, hosts in the EU by default with US data residency on request, deploys in a couple of hours over single sign-on with Azure AD or Entra ID, and installs no agent. For a like-for-like view, the comparison of Tryane against Viva Engage native analytics sets out where each fits.
Frequently asked questions
Can analytics increase Viva Engage engagement by 50 per cent?
No responsible answer to that question is a number. Analytics changes what community managers do, and behaviour change moves participation, but the size of the movement depends on your starting point, your population mix and what else happens in the same period. Treat any specific percentage offered before a supplier has seen your tenant as a marketing figure rather than a forecast.
Is Yammer still available?
No. Microsoft completed the transition of Yammer to Viva Engage and retired the Yammer branding and standalone experience. Communities, posts and membership carried across. Reporting continuity generally did not, which is why multi-year comparisons spanning the change are hard to produce from native tooling alone. The differences are set out in our comparison of Viva Engage and Yammer.
What is a good engagement rate for an internal community?
External benchmarks are weakly comparable, because they rarely publish their denominator and almost never match your population mix. Your own trend on a fixed denominator is more useful than any published figure. If you do want external reference points, read them alongside our note on internal communications benchmarks and check what each one is dividing by.
Can we prove a campaign caused a change in participation?
Rarely with certainty, and you should say so. What you can show is that the targeted population moved and a comparable untargeted population did not, over a period long enough to survive a single unusual post. That is weaker than causal proof and considerably stronger than a raw before-and-after figure.
Does Tryane replace Viva Engage analytics?
No. Viva Engage continues to record what it records. Tryane sits above it, reading Viva Engage alongside SharePoint, Teams and newsletters and joining all of it to your organisational structure, so you can answer segmented and cross-channel questions that no single native panel can. You keep the Microsoft 365 investment and add the measurement layer.
Sources
• Microsoft Learn, Viva Engage overview
• Microsoft Learn, Viva Engage activity report
• Gallup, State of the Global Workplace
• Gallagher, State of the Sector
Further reading
• What Yammer changed, and what Viva Engage inherited
• The limits of Viva Engage analytics
• Community managers and Viva Engage performance
• Tryane compared with Viva Engage native analytics
• Internal communications benchmarks for 2026
Tryane runs a 15-minute working session with Heads of Internal Communication, going through your current Viva Engage figures and showing which of them would survive a question about the denominator. Book a slot with Jérémy.