An eight-week culture campaign finishes on a Friday. Three weeks later the Head of Internal Communication is asked to present its results to the executive committee. The team opens the analytics, finds 41,000 page views, 6,200 unique readers and a healthy spike in Viva Engage activity in week three, and then spends two days trying to work out whether that is good. It is not a data problem. Nobody wrote down, before the campaign started, what a successful outcome would look like. Everything after that point is retrofitting, and executives recognise retrofitting when they see it.
This is the step almost every internal communication programme skips, and it costs more than any tooling decision. A dashboard cannot tell you whether something worked if the definition of working was never fixed.
Why the objective is the measurement
The measurement industry inside internal communication has spent a decade arguing about metrics and almost no time on objectives. That is the wrong order. A metric is only meaningful relative to a claim you made in advance. Without the claim, every number is simultaneously good and bad, and the conversation defaults to whoever is most confident in the room.
Gallagher’s State of the Sector research repeatedly finds measurement named as the capability internal communicators most want to improve[1]. In practice a large share of that gap is not analytical. It is that the function is asked to prove the value of work whose intended effect was never specified. You cannot instrument an ambition.
There is a second cost. When success is undefined, the team defends itself with volume, which is the weakest available evidence. Our guide to internal communication analytics separates the four questions a number can answer, and the objective decides which of the four you are entitled to claim.
What a usable success statement contains
A success statement is one sentence written before the first asset is produced. It has four parts, and it is unusable if any of them is missing.
A named population. Not “the organisation”. The 3,400 people in operations across the eleven European plants, or the 620 people-managers, or the whole of the Spanish entity. The population is also your denominator, which is why this step forces a conversation with HR you will otherwise have later and under pressure.
A specific behaviour or state. What should be different at the end. Read the policy, completed the acknowledgement, attended one of the six briefings, can name the three priorities when asked in a pulse question.
A threshold. A number you commit to in advance, with the reasoning attached. Thresholds are uncomfortable precisely because they can fail, which is what makes them evidence.
A window. Measured when. Four weeks after the last asset, not “eventually”.
| Common objective | Why it cannot be measured | The usable version |
|---|---|---|
| Raise awareness of the new strategy | No population, no behaviour, no threshold | 65 per cent of the 2,900 EMEA staff read at least one of the three strategy posts within four weeks |
| Improve employee engagement | Measures a workforce state the campaign cannot move alone | Manager-cascade completion above 80 per cent in the twelve sites with the lowest pulse scores |
| Increase intranet visibility | Visibility is not an outcome for anyone | Unique readers in operations rises from 9 per cent to 25 per cent of that population by quarter end |
| Support the transformation | Support is not observable | 90 per cent of people-managers open the toolkit and 50 per cent run a team session, logged in the tracker |
Three objectives that are not objectives
“Raise awareness” is the most common. It fails because awareness has no denominator and no threshold, so any result confirms it. If awareness genuinely is the goal, convert it: awareness means a stated share of a stated population can recall a stated message at a stated time, and that is measurable through a two-question pulse.
“Improve engagement” is the most damaging, because it borrows a workforce measure that internal communication does not control and that moves slowly. Gallup’s long-running research treats engagement as a durable state shaped mostly by managers and the daily experience of work[2]. Adopting it as your campaign objective means accepting responsibility for something a communications campaign cannot move by itself. The argument is set out in full in our piece on mistakes in employee engagement analysis.
“Increase visibility of the intranet” fails a different test. It is an objective for the channel, not for the organisation. No executive has ever funded a function to increase the visibility of a website. Ask what the visibility is for, and the real objective appears underneath it.
Setting a threshold before you know the answer
The usual objection is that committing to a number in advance is guessing. It is, and a documented guess beats a post-hoc rationalisation every time. Three sources make the guess defensible.
Your own history. The most similar communication you ran in the last eighteen months, measured the same way. This is the strongest anchor and the reason to keep history that survives an intranet redesign or a product migration.
Sector reference points. Useful as a sanity check on order of magnitude, not as a target. Our note on internal communications benchmarks covers where comparison is fair and where it misleads.
The decision the number will drive. Ask what you would do differently at 40 per cent versus 70 per cent. If the answer is the same, the threshold is decorative and you should pick a different measure. This test kills more bad metrics than any framework.
Write the threshold into the campaign brief, share it with the sponsor before launch, and hold it. A threshold agreed in advance and missed is a finding. A threshold invented afterwards and met is a slide nobody believes.
What Microsoft 365 can confirm, and what it cannot
Native analytics is free, on by default and sufficient for a single team on a single surface. SharePoint site usage reports and per-post view counts genuinely answer whether a page was read, and Microsoft 365 usage reporting gives a solid picture of activity across the tenant[3]. If your success statement names one SharePoint hub and one behaviour on it, you may need nothing else.
The limit is the population clause. Native reporting has no view of your organisational structure, so it cannot express a result as a share of operations, of a country, or of people-managers. That is the part of the success statement executives care about, because it is the part that says who was reached and who was not. Structuring those attributes is covered in our piece on audience segmentation. Native reporting also counts per channel, so a campaign that ran on the intranet, the newsletter and in Teams produces three unlinked numbers rather than one reach figure.
Building the join in Power BI against the Microsoft Graph reporting API is a legitimate route with a known cost. 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.
This is where Tryane sits. It reads SharePoint, Viva Engage, Teams and your newsletter platform together, joins them to your organisational structure through Entra ID or an HR file, and keeps history across product transitions so last year’s campaign is still a valid baseline for this year’s threshold. 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.
The one-page version
Before the next campaign, write four lines and get the sponsor to initial them: the population, the behaviour, the threshold, the measurement date. Add one line naming what you will stop doing if the threshold is missed, which is what turns measurement into management. Then report against exactly those four lines and nothing else, with volume as context underneath. The report that results is shorter than what you produce today and considerably harder to argue with. Our guide to proving internal communications works to leadership covers how to present it, and the internal communication KPIs guide lists the measures worth committing to.
Frequently asked questions
What if leadership refuses to agree a threshold in advance?
Set one anyway and state it in the brief as your own working assumption. The purpose of the threshold is to make the result interpretable, and that works whether or not the sponsor formally signs it. In practice a sponsor who declines to agree a number will often engage once they see a specific one proposed, because arguing with 65 per cent is easier than inventing a target.
Is it acceptable to measure success on reach alone?
For most communications, yes, provided reach is expressed against a named population rather than as a raw count. Reach is honest, defensible and sufficient for routine reporting. Reserve behavioural and outcome claims for the few communications where a behaviour was explicitly requested and a system records it.
How do we measure success for messages that have no call to action?
Their success statement is a reach and attention statement, not a behaviour one. A stated share of a stated population read it, and the reading time is consistent with having read rather than glanced. Trying to attach a behavioural outcome to a message that asked for nothing is the fastest route to an attribution argument you will lose.
Should engagement survey scores be part of the success definition?
Only as a slow, shared context measure, never as the primary success criterion for a campaign. Surveys run once or twice a year, they lag, and they are influenced by pay, workload and management quality far more than by communication. Use them to frame the problem and use communication data to measure the intervention.
How far back should the baseline go?
At least four comparable communications, which for most teams means twelve to eighteen months. The practical obstacle is that native reporting retains a finite window and an intranet redesign or a platform migration typically resets what you can see. If the baseline matters to you, the history needs to be held independently of the products that produced it.
Sources
• Gallagher, State of the Sector
• Gallup, State of the Global Workplace
• Microsoft Learn, Microsoft 365 activity reports
Further reading
• A guide to internal communication analytics
• Calculating the ROI of internal communications
• Engagement surveys against communications data
• Executive reporting for internal communications
• An internal communications reporting template
Tryane runs a 15-minute working session with Heads of Internal Communication: bring one campaign you are about to launch and we will write its success statement with you, then show which parts your current Microsoft 365 setup can already evidence. Book a slot with Jérémy.