The annual intranet review usually arrives as a deck of site usage charts. Page views by site, unique visitors by month, most-visited pages, a search term list. Every number is accurate. And when the CCO asks the only question that matters, which is whether the intranet is working, the deck cannot answer it, because every chart in it describes the site rather than the organisation the site exists to serve.
That is a scoping failure, not a tooling failure. SharePoint records what it records perfectly well. The problem is that an intranet performance analysis has to start from a population and a question, and most of them start from whatever report was easiest to export. Here is how to run it the other way round.
Decide what performance means before you open a report
An intranet in a large organisation is doing three jobs at once, and they fail independently. It distributes news that the organisation needs people to see. It answers questions people already have, mostly about policy, process and tools. And it acts as the front door to everything else, from expense claims to the safety library. A single “is the intranet performing” question collapses all three, which is why it never gets a satisfying answer.
Split them and each becomes measurable. Distribution asks whether a specific piece of news reached the population it was written for. Answering asks whether people found what they came for, which is a search and journey question rather than a page-view question. Front door asks whether the intranet is where the organisation starts, or whether people have quietly reverted to asking their manager. Different questions, different evidence, and only the first one is really about news pages at all.
The reviews that fail almost always fail at the denominator
“The article got 4,200 views” is not a performance statement. It becomes one the moment you can say what it is 4,200 out of. In a 12,000-person group with 40 sites, an article aimed at the whole organisation that reaches 4,200 people is a poor result. The same 4,200 for a message aimed at the commercial function is close to complete coverage.
Setting the denominator is unglamorous work and it is where the value is. You need the addressable population for each message, which usually means the headcount minus the people who genuinely have no digital access, split by the attributes you will be asked about: function, site, country, language, contract type. Get that right once and every subsequent number in the review means something. Our note on which intranet metrics matter goes through the specific SharePoint figures that mislead when they are reported without one.
The measurement plan, surface by surface
SharePoint is not one surface, and the reporting differs by surface in ways that matter to what you may claim.
| Surface | What native gives you | What to actually report |
|---|---|---|
| Organisation news and the home site | Site usage, page views, unique viewers, time on page. | Reach against the addressable population for that message, with the segment breakdown. |
| Hub and department sites | Per-site usage reports, owned by the site owner rather than by you. | Whether the department channel duplicates or complements the central one. Overlap matters more than volume. |
| Search | Query volume, top queries, abandoned queries. | Failed and abandoned queries, which is your best evidence of what the intranet does not answer. |
| Viva Connections feed and mobile | Separate analytics for the Connections experience[1]. | Frontline and deskless reach, which the desktop intranet numbers systematically understate. |
| Tenant level | The SharePoint site usage report in the Microsoft 365 admin centre[2]. | Scope control: which sites are intranet and which are team collaboration. Never report the tenant total as intranet traffic. |
Two details from that table are worth dwelling on. The first is that tenant-level SharePoint usage includes every team site and every document library in the organisation, which is mostly collaboration and has nothing to do with your intranet. Reporting it as intranet performance inflates the number by a factor that varies month to month, and the boundary is discussed properly in why the OneDrive and SharePoint confusion corrupts your intranet numbers.
The second is search. Failed queries are the most under-used evidence in the whole stack. A month of search logs will tell you exactly which policy nobody can find and which tool nobody knows the name of, and fixing those is usually cheaper and more visible than a redesign. Microsoft’s page diagnostics and portal health guidance covers the technical side of keeping those pages fast enough to be usable[3].
What changed by 2026, and what it does to your numbers
Three shifts have altered what a page view means, and none of them show up as a warning in the reporting.
Consumption moved off the page. A growing share of intranet content is now read inside the Viva Connections feed, inside a Teams tab, or surfaced in a Copilot answer, without anyone loading the page. The page view undercounts, and it undercounts unevenly across populations. What happens to your figures when an assistant answers before the user reaches the intranet is the subject of chatbots and the intranet.
The mobile and frontline path is separate. Viva Connections has its own analytics surface, and if you are not reading it, your frontline reach number is not low, it is missing. That is a different and much more embarrassing error to make in front of an operations director.
Reporting windows did not get longer. Native retention is finite, and it is shorter than the horizon of most intranet programmes. A three-year redesign narrative cannot be told from native data alone, which is exactly the narrative a redesign business case needs. Measuring the impact of a SharePoint intranet redesign covers the before-and-after design.
The honest position on native tooling
SharePoint native analytics is free, already switched on, and sufficient for a site owner who wants to know whether their own pages are being read. For a single communication site with a single audience, it does the job and you should not spend anything. Read it properly first: using SharePoint native analytics explains where each number comes from, and the limits of SharePoint native analytics sets out precisely where it stops.
The gap is the same one every time. Native does not join to organisational attributes, so the segment breakdown that makes a denominator useful is not available without exporting and joining the data yourself. It is per-site, so cross-site and cross-channel unique reach cannot be computed. And the history is short.
The usual answer is a Power BI build against the Microsoft Graph reporting API. It can be done, and several organisations run it well, but go in with your eyes open. The reporting API is not consistently reliable and produces days with missing data[4]. Customers have observed KPI values in Power BI that do not reconcile with the native SharePoint reports for the same period, and that discrepancy tends to surface in the meeting where you least want it. A dependable build needs a robust ingestion process, scheduled reconciliation against the native figures and a named owner. Budget for the pipeline rather than the dashboard, a point made at length in building SharePoint analytics with Power BI.
A cadence that survives a reorganisation
Analysis that only happens once a year is not analysis, it is an annual argument. A workable rhythm has three layers. Weekly, the publishing team looks at reach against target segment for anything that mattered, and nothing else. Monthly, the IC team reviews failed searches and the two or three populations where reach is furthest below plan. Quarterly, the head of function takes one structural question to the leadership team, supported by a trend rather than a snapshot.
The reorganisation point is not incidental. Large organisations restructure, and when they do, the old segment definitions stop matching the new ones. If your measurement layer holds history against the attributes as they were, you can restate the trend. If it does not, your baseline disappears at exactly the moment leadership wants proof that communication held the organisation together through the change. Gallagher’s State of the Sector research consistently finds measurement to be the capability internal communicators name first as their gap[5], and continuity through change is a large part of why.
This is where Tryane sits. It reads SharePoint alongside Viva Engage, Teams and your newsletter platform, joins that activity to your organisational structure through Entra ID or an HR file, and keeps unlimited history so a trend survives a redesign, a migration or a restructure. SOC 2 Type 2 certified, GDPR compliant by design, EU hosting by default with US data residency on request, deploys in a couple of hours over single sign-on, no agent to install.
Frequently asked questions
What is a good page view number for an intranet news article?
There is no useful absolute figure, which is why benchmarks of raw views mislead. Report reach as a share of the addressable population for that message. A well-targeted piece reaching 60 per cent of its intended segment is a strong result; the same view count against the whole organisation may be a weak one.
Can I measure frontline intranet reach from SharePoint alone?
Only partially. Frontline consumption often happens through Viva Connections on mobile, which reports separately, and some of it happens through channels that are not SharePoint at all. A frontline reach figure built only from SharePoint site analytics understates the audience and hides which populations are genuinely unreached.
How long does SharePoint keep analytics data?
The native retention window is finite and shorter than most intranet programmes. Plan for it: either export and store the data yourself on a schedule, or use a layer that retains history independently. Discovering the limit when someone asks for a three-year trend is the expensive way to find out.
Should search be part of the intranet performance review?
Yes, and it should carry more weight than it usually does. Failed and abandoned queries are direct evidence of unmet demand, they are cheap to act on, and improvements show up quickly. Most teams report top queries, which is far less useful than the queries that returned nothing worth clicking.
Does Tryane replace SharePoint analytics?
No. SharePoint continues to record what it records. Tryane sits above it and joins that activity to your organisational structure across channels, so you can report reach by function, site, country and language, compute unique cross-channel reach, and hold a trend that outlives a redesign. The comparison with SharePoint native analytics sets out the four gaps in detail.
Sources
• Microsoft Learn, Viva Connections analytics
• Microsoft Learn, SharePoint site usage report
• Microsoft Learn, portal health and page diagnostics
• Microsoft Learn, known issues with Microsoft Graph
• Gallagher, State of the Sector
Further reading
• Using SharePoint native analytics properly
• The limits of SharePoint native analytics
• The top KPIs to measure intranet success in 2026
• Viva Connections analytics and intranet adoption
• Measuring cross-channel internal communications
Tryane runs a 15-minute working session with Heads of Internal Communication, taking your current SharePoint intranet reporting and showing what the same numbers look like once they carry a denominator. Book a slot with Jérémy to schedule yours.