Search this question and you will get a great deal of advice about subject lines. Use emojis. Keep it under 200 words. Send on a Tuesday. Add a video. Personalise the greeting. Segment your audience. Try a different channel.
Some of it is even true. None of it addresses the reason, which is that in an organisation where people expect to have a say, being told things is a downgrade, and no amount of formatting makes a downgrade feel good.
The uncomfortable diagnosis
Internal announcements have a low open rate for the same reason a notice board has a low readership. They are a one-way channel in an organisation full of people who have opinions about the thing being announced and no route to express them.
Watch what actually happens when internal announcements land. A few people read it. Rather more skim the first line, establish that it does not require anything of them, and move on. And a handful go and discuss it somewhere else, usually in a channel the communications team cannot see, where the real conversation happens: was this a good idea, who decided it, and why were we not asked?
That last group is the interesting one, because their engagement is high. They care a great deal. They are simply not engaging with the announcement, they are engaging with the decision behind it, and the announcement gave them nowhere to do that.
The reframe that changes everything. Low readership is not evidence that people are not interested. It is evidence that the format does not match what they are interested in. People do not want to be informed about decisions. They want to be part of them, and failing that, they want to understand them.
Why this is worse in a flat organisation
If your organisation has deliberately distributed authority, you have raised expectations about participation while often leaving the communications model exactly where it was.
The result is a mismatch people feel keenly. They have been told they are trusted to make decisions in their own domain, and then they receive a company-wide announcement written in the voice of a management team informing staff. The form of the message contradicts the stated culture, and the form is more believable, because form is harder to fake than words.
This is why internal comms engagement is often worse in progressive organisations than in conventional ones. The gap between what people were promised and what the channel delivers is simply wider.
The three things a broadcast is missing
The reasoning
Almost all internal announcements state an outcome and omits the thinking. "From next month, we are changing how we handle expenses." Fine. But the reader's actual question is why, and specifically what problem this solves that the old approach did not, and what was considered instead.
Without that, the reader has two options. Accept it, which requires trust they may or may not have. Or assume the reasoning was weak, which is the default assumption when reasoning is withheld, because people know that good reasoning tends to get shared.
Attaching the decision record to the announcement solves this at essentially no cost. The reasoning already exists, or should. Publishing it converts a directive into an explanation.
A route to respond
Internal announcements that cannot be replied to are telling the reader that their view is not wanted. Most organisations do not intend this and communicate it anyway, purely through channel design.
The fix is not a comments box, which usually attracts either silence or the three most confident people. It is publishing the thing before it is settled, as a proposal with a closing date, so that responding is a normal part of the process rather than an act of dissent.
Evidence that responding has ever mattered
If your organisation has invited input before and visibly changed nothing, that history is doing more damage to your open rates than any formatting decision.
People calibrate quickly. Two rounds of consultation that produced no observable change is enough to teach an entire organisation that participation is theatre, and that lesson survives several years and multiple communications strategies.
What to do instead
Four changes, in rough order of impact.
Move things upstream. Publish the proposal rather than the decision, with a date by which objections need to be raised. This single change converts a broadcast into an invitation, and it does so without any extra communication, because you were going to tell people anyway. You are simply telling them earlier, when their input is still usable.
Attach the reasoning to everything. Internal announcements link to the decision they report, including alternatives considered and the assumption it depends on. Readers who care get the full context, readers who do not are unaffected, and the organisation stops looking arbitrary.
Separate the formats properly. A poll, a proposal and an announcement do different jobs and commit you to different things. Using one where another belongs is the most common cause of the resentment that shows up as disengagement. We have set out how to choose between them.
Close the loop out loud. When input changes something, say so and say whose input it was. When input does not change something, say that too, with a reason. The second case matters more than the first, because it is the one that proves the process is real rather than decorative.
What you send now | What to send instead |
|---|---|
"We have decided to change X" | "We are proposing to change X, objections by Friday" |
Outcome only | Outcome, reasoning, what else was considered |
No reply route | A named place to raise a concern, and who will answer it |
Silence after consultation | What changed, what did not, and why |
Diagram comparing a one-way announcement flow with an upstream proposal flow that includes a response window and a visible outcome
The same information, moved twenty-four hours upstream, stops being a broadcast and starts being a decision people are part of.
What about the things that genuinely are announcements?
Not everything is a decision people should participate in. Legal changes, external news, the office being closed on Monday. These are genuine internal announcements and they should stay that way.
The point is proportion. If everything arriving in the feed is a fait accompli, the channel acquires a reputation as the place where you find out what has already happened, and people read it accordingly. If most items are live and a minority are informational, the informational ones get read too, because the channel has established that it is worth opening.
The ratio matters more than the writing.
The structural version of the fix
You can do all of this with existing tools and enough discipline. Many organisations do, for a while, and then the discipline erodes because nothing in the tooling supports it. Internal announcements are easy to send and proposals are not, so the system drifts back towards broadcast.
Autono's Feed is built the other way round. News, announcements, polls and proposals share one prioritised stream, but they are distinct objects with different mechanics: a proposal has a closing date and a place for objections, a decision carries who decided it and why, and the outcome links back to the input that shaped it. Every decision made through the Feed is logged and auditable by default, so the loop closes because the loop is the product rather than a practice someone has to remember.
If your internal communications are technically working and nobody is reading them, the problem is probably not the writing. Get started with Autono, or request a demo if you would rather be shown what it looks like when people have a reason to open the thing.
Also in this series: how much transparency is too much on why openness alone does not fix reach.
