How do you measure internal communication effectiveness?
The dashboard is green. The announcement went out on Sunday, most people opened it, the intranet post has the most views of the month. On Thursday a shift supervisor calls HR to ask what the new attendance rule actually says. Nothing in the dashboard was wrong. It measured delivery. It did not measure whether anyone understood.
The short answer: measure five things, in this order. Did the message reach every group, including frontline and off-site staff (reach)? Did people understand it in the language they read best (comprehension)? Do they remember it and act on it a week later (retention)? Was it about their work (relevance)? Are your channels coordinated, or adding noise (channel load)? Opens, clicks and views belong in the report as evidence of delivery — not as evidence that communication worked.
Why do open rates say nothing about understanding?
Because an open is a technical event, not a human one. An email platform counts an open when a tracking image loads. It cannot tell whether the reader skimmed the subject line, read the second paragraph, or understood the deadline in it.
Most teams still report exactly that. In Gallagher's 2025/26 survey of internal communication and HR professionals, most of them in the US and UK, 44% described their measurement as basic output tracking and 26% as campaign-level reporting. Only 16% measured outcomes or sentiment, and 12% looked at business impact (n = 474 answering that question). Most leaders do not ask for impact: respondents said their leaders mainly ask for sentiment (56%) and activity or reach (52%), and only 27% ask for business-impact data.
The numbers themselves are less solid than they look. Published open-rate benchmarks come from email-tool vendors' own customers, and pixel-based opens are distorted by privacy features and image blocking. Reading time is short even when the open is real. Microsoft's 2024 Work Trend Index found, from Microsoft 365 usage data, that 85% of emails are read in under 15 seconds. Its survey covered no Middle East market.
The measurement profession drew this line years ago. AMEC's Integrated Evaluation Framework, built for public communication and applied here by analogy, separates outputs (what you put out that reaches people) from out-takes (what audiences take from your communication), outcomes (the effect on them) and impact (the results your communication caused, in full or in part). An open rate is an output. Understanding is an out-take. The report your CEO wants is about outcomes.
What should you measure instead?
Five measures cover what matters, and none of them needs new software to start. The aim is a picture of what people understood and did — not a longer list of activity. The table gives the question each one answers, a metric you can report, and the reason it earns its place.
| Measure | The question it answers | A metric you can report | How to collect it | Why it matters |
|---|---|---|---|---|
| Reach | Did every group get it, including frontline, night shift and off-site staff? | Share of each target group reached through at least one channel within 3 working days | Channel data, plus a supervisor confirmation for teams without email | An average hides the warehouse — the group that never got the message disappears in the total |
| Comprehension | Did people understand what changed and what to do? | Share answering two or three check questions correctly, by group and language | A 60-second check after each key message | Understanding is the point; an open is only the receipt |
| Retention | Do they remember it and act on it a week later? | Recall at 7 days, plus one behaviour you can see in operational data | A follow-up micro-poll; the system the message was about | Behaviour is what pays for the effort |
| Relevance | Was the message about their work, or only about the company? | Share answering yes to "Does this affect my work?"; share of messages tagged as work-related in a content audit | A one-question yes/no check and a monthly content tag | Irrelevant messages teach people to skip the next one |
| Channel load | Are channels and senders coordinated, or is it noise? | Messages per employee per week, by channel; number of senders; duplicate sends | A channel calendar and a four-week message log | Most communicators see overload rising: 83% of Gallagher respondents say information overload is a growing problem (n = 631) |
One more measure is worth adding when the budget is questioned: cost per employee who acted — the campaign budget divided by the number of people who did what the message asked. It turns a communication plan into a figure a finance director can compare with other spending, and it is a measure we use when we review campaigns. Where you can, compare with a site or a period the campaign did not reach, so you count the people who acted because of it. Set each measure when you set the objective — not after the campaign — and build the measures into your plan from the first draft.
Which tool measures which internal communication channel?
No single tool measures every channel. Each platform reports what it can see, which is usually delivery. The useful question is what each channel cannot tell you, and which simple instrument fills that gap. Before you measure a channel, be clear what each channel is for. A channel judged on a job it was never meant to do will always look weak.
| Channel | What its own data gives you | What it cannot tell you | The tool that fills the gap |
|---|---|---|---|
| Email and newsletters | Opens and clicks (pixel-based) | Whether the content was understood | A three-question check sent 2–3 days later |
| Intranet or employee app | Page views, time on page, reactions | Whether frontline staff can reach it at all; understanding | An access audit by group and a one-question in-app poll |
| WhatsApp groups | Read receipts in small groups, replies | Anything at scale; who was never added | Supervisor confirmations and a short poll link or QR code shared in the group |
| Town halls, in person or online | Attendance | What people took away | A two-question exit poll by QR code, and the themes of the questions asked |
| Line managers and team briefings | Whether the briefing happened | Whether the message survived the cascade | A cascade check: ask five people per team for the key point within 48 hours |
| Posters, screens and printed notices | Nothing | Recall | A spot recall check on site, a week after the notice goes up |
| The whole system | — | How the channels work together | A communication survey, focus groups and, when the stakes justify it, a full audit |
Two patterns from Gallagher's 2025/26 data are worth knowing before you choose. Only 40% of respondents were satisfied with their ability to reach all employees regardless of location or work type (n = 571). And in frontline-heavy organisations, in-person town halls were rated the single most effective channel (51%), and virtual town halls were rated 18 points lower. Microsoft's frontline research points the same way: 63% of 9,600 frontline workers surveyed in eight markets in 2021, none in the Middle East, said messages from leadership did not make it to them.
In many Egyptian workplaces the channel that actually reaches a production line, a hospital ward or a sales force is a supervisor's WhatsApp group. Measure it as a channel — not as an informal side effect. That means knowing which groups exist, who is in them, and whether the official message reached them unchanged.
How do you test comprehension in Arabic and English?
Test in the language and the words people actually read, and give every question one right answer. A comprehension check is a short quiz, not an opinion question, and that is its advantage: a question with a right answer is not pulled up or down by mood or by habits in how people use a rating scale. Research on response styles finds that answers on agreement scales shift with culture and even with the language of the questionnaire (Harzing, 2006, across 26 countries; no Arab country is confirmed in the sample). A knowledge check avoids most of that.
- Write the check in Arabic first. Translated questions carry the English sentence structure with them, and readers notice. Our guide to Arabic-first employee surveys explains why.
- Ask about the action, not the announcement. "From which date does the new attendance rule apply?" tests understanding. "Did you read the message about attendance?" tests nothing: most people will say yes.
- Add an honest exit. Include "I am not sure" as an option, so guessing does not inflate the score.
- Keep it to 60 seconds. Two or three questions, one link or QR code, no login.
- Sample, do not survey everyone. A few dozen answers per group can show a large gap between sites, shifts or languages. By standard sampling arithmetic, a gap of about 25 points or more is large enough to trust. Smaller gaps and precise percentages need more answers.
- Report the gap, not the average. A line such as "head office 91%, night shift 40%" (illustrative figures) is a finding. "Average 78%" hides it.
What can you measure this month with no new software?
You can run a full measurement cycle in four weeks with a free form builder, a QR code and a spreadsheet.
- Week 1 — choose. Pick the three most important messages of the coming month. For each, write the groups it must reach, the one thing people must know, and the one thing they must do. Start a simple log of every message each group receives, by channel.
- Week 2 — build. Write two or three check questions per message, in Arabic and English. Put them in a free form and make a QR code for teams without email.
- Week 3 — send and check. Send the messages as planned. Within two to three days, run the check with a sample from each group, and ask supervisors to confirm reach for teams you cannot see in the channel data.
- Week 4 — recheck and report. Seven days after each message, ask one question again to measure retention. Then report on one page: reach, comprehension and retention for each group, plus messages per employee per week from your log. Lead with the largest gap and the one change you will make.
That one page answers the question the dashboard could not. It also gives you a baseline, so next quarter's report can show movement rather than activity.
When do you need a full internal communication audit?
When the problem is the system rather than a single message. Signs to watch for: the same misunderstanding keeps causing rework; nobody can list every channel and its owner; several departments send to the same people without coordinating; a restructuring or major change is coming; or leadership no longer trusts the numbers you report. Change is a frequent trigger. In Gallagher's 2025/26 survey, change-management communication was the skill most often rated critical for the next year (57%, n = 587), while 61% of teams had no change communication approach in their strategy.
A full audit measures what a self-check cannot: how channels, content and leaders work together. Our IC Quality Index assesses channel architecture, message clarity, leadership communication capability and governance, and delivers an IC maturity scorecard, a channel map with content-flow diagnostics, friction hotspots with quick wins, and a 12-week prioritised improvement plan.
If you want a first reading before you commit to anything, start with the free IC Quality Index self-check. It takes about five minutes, shows your weakest measure, and says plainly what it is: an indicative self-assessment of your practices — not an audit of what employees understood.
Frequently asked questions
What are the most important internal communication KPIs?
Five cover most needs: reach by group, comprehension rate, 7-day retention or action, a relevance check, and messages per employee per week. Report each by group — site, shift, language — because the average hides the people the message missed.
What should an internal communication dashboard show?
One row per measure, split by group: reach, comprehension, 7-day retention or action, relevance, and messages per employee per week. Keep opens, clicks and views in a separate delivery panel below them, so nobody reads delivery as understanding.
What is a good internal email open rate?
There is no neutral benchmark: published averages come from email-tool vendors' own customers, and privacy features and image blocking distort pixel-based opens. Treat any open rate as a delivery figure, and compare a channel with itself over time. A high open rate with a low comprehension score means the message arrived and failed.
How often should you measure internal communication?
Check key messages every time: understanding within two to three days, recall about seven days after sending. Review channel load and reach each quarter. Run a full audit once a year, or before a major change, whichever comes first.
Can you measure the business impact of internal communication?
Yes, when the message asks for a behaviour that shows up in operational data — fewer errors in a process, faster adoption of a system, fewer safety incidents. The Barcelona Principles 4.0 (AMEC, 2025) ask communicators to measure their contribution by outcome and impact, and to stop using invalid measures such as advertising value equivalents.
What survey questions should you ask employees about internal communication?
Ask questions with a checkable answer or a real choice, not agreement ratings. For example: Where did you first hear about the new attendance rule? What is the one thing you need to do differently because of it? The last time you asked about a company announcement, did you get an answer? Which is more true for you: "I get more messages than I can read" or "I miss information I need for my work"?
What is the difference between an internal communication survey and an audit?
A survey tells you what employees say about communication. An audit adds what the survey cannot see: a review of every channel and sender, a content analysis, interviews with leaders and managers, and the governance behind it. Start with a survey when you need employees' view of communication; start with an audit when you cannot say which groups your messages actually reach.
Sources
- Gallagher, Employee Communications Report 2026, Global Edition (State of the Sector 2025/26; more than 1,300 communication and HR professionals in 40 countries, September–November 2025; most respondents in the US and UK). ajg.com
- Microsoft, Work Trend Index 2024 (31,000 knowledge workers in 31 markets, plus Microsoft 365 usage data). microsoft.com
- Microsoft, Technology Unlocks a New Future for Frontline Workers (9,600 frontline workers in 8 markets, October–November 2021). microsoft.com
- AMEC, Integrated Evaluation Framework and taxonomy. amecorg.com
- AMEC, Barcelona Principles 4.0 (June 2025).
- Harzing, A.W. (2006), "Response styles in cross-national survey research: a 26-country study", International Journal of Cross Cultural Management 6(2). harzing.com