Internal comms distribution: getting the right message to the right people, on the right channel
Working in internal comms, you'll know the story all too well.
The policy was published, the update was sent and the CEO mentioned it in the town hall. But three weeks later, colleagues on the frontline are still following the old process from a PDF they downloaded last year, while their manager insists they “never saw anything about an update”.
It's tempting to blame employees for ignoring information, managers for failing to cascade, or the intranet for underperforming. The real issue is usually simpler: publishing and distribution were treated as the same thing — and a page going live was mistaken for a message landing.
In our companion Publishing post, we made the case for a canonical source: one governed place where final content lives, is owned, and can be trusted. That remains true. But it’s only half the job.
Publishing and distribution are two sides of the same coin. One creates the source of truth; the other makes sure that truth has a fighting chance of being seen by the people who need it. In the five-layer model, Distribute is a critical element of the ecosystem, not an optional extra you bolt on after a page is published.
Publishing without distribution is a noticeboard nobody walks past. The content exists. It may even be beautifully written, carefully governed, and perfectly searchable. None of that helps if the people who need the content don’t know it’s there or discover it too late.
That’s why the distribution argument starts with the same principle as publishing: publish once to a single, trusted place, then distribute smartly through channels that point back to it. Not five versions of the truth. One source, many signposts.
Pull alone fails predictably
A surprising amount of internal communication still runs on a fantasy: “We published it on the intranet, so people can go and find it.” They can. But that doesn’t mean they will.
Pull channels are useful, but they fail in predictable ways when left to do the whole job alone. Coverage gaps open. Frontline and deskless colleagues miss what desk-based teams see first. Mandatory updates become optional in practice. Older versions keep circulating because someone downloaded a PDF six months ago and kept sending it around. None of this is random; it’s structural.
In the real world, organisations are fragmented across locations, devices, shifts, contracts, languages and working patterns. That makes “go and look for it” a weak default, especially for anything time-sensitive or operational. Pull alone increases information inequity and creates exposure precisely where organisations can least afford it.
It also wastes effort. Teams spend time crafting content that never reaches the people it was meant to help, simply because it was not surfaced in the right place, at the right moment, in a format that fits people’s working day.
Smart distribution is orchestration, not megaphoning
When pressure rises, the instinct is to push harder: email everyone, post it everywhere, light up every notification you can find. That can feel decisive, but it’s usually just noise.
If pull alone leaves people uninformed, the instinct is to overcorrect and push everything, everywhere. But do that and you’ll quickly overwhelm people with comms, most of which will go unread.
Smart distribution is orchestration, not megaphoning. The discipline is to use the least intrusive channel that still achieves the outcome, to be clear why someone is receiving a message, and to sequence channels deliberately rather than firing them all at once like confetti from a cannon.
That matters because trust in channels is cumulative and fragile. If everything is urgent, nothing is. If every message arrives in every channel, people learn to tune the whole system out. The most powerful distribution decision is often not sending. Sometimes the best move is to let the homepage do the work, or to surface a targeted message rather than sending a reminder to people who already completed the task yesterday. Restraint is not passivity; it’s what protects attention.
The patterns that work are rarely glamorous, but they’re dependable.
Manager-first sequencing gives leaders context before their teams get the all-staff version, which is especially useful for change messages likely to prompt questions. Crisis communication often works best as a sequence: an initial alert, a follow-up nudge in a flow-of-work channel, a central hub with current information, then a summary once the dust settles. Non-urgent content benefits from a digest rhythm rather than a daily drip of interruptions. Underneath all of that sits an attention ladder: start with the least intrusive channel, then escalate only when the message genuinely warrants it.
Right message, right people, right channel
“Right message, right people, right channel” is the slogan. Discipline, data, and design are the job.
Start with the middle part. The right people is a targeting decision before it’s a tool decision. That means using reliable organisational data to decide who actually needs a message: by role, location, business unit, employment type, language, device context, and whatever else is genuinely relevant. Good distribution reduces noise by design. It does not treat relevance as a nice extra.
This is also where teams often muddle two different ideas.
Targeting is who gets the message.
Personalisation is how it appears.
They are not the same thing, although both use data as the enabler. A compliance reminder may need to be targeted only at managers in one region. A version of that message may also be personalised by language or device format so it’s easier to act on.
Both can improve relevance. Both can also go wrong. Over-target and you exclude people who should have seen it. Over-personalise and you drift into something that feels creepy.
The right channel, meanwhile, is not a vendor decision. It’s a sequencing decision across a toolkit. Email is still useful for structured, referenceable updates. Intranet and app surfacing are strong for context and depth. Mobile notifications matter for frontline audiences. Chat nudges can work well for timely prompts. SMS, messaging apps, desktop alerts and pop-up banners are high-intrusion tools and should be treated accordingly. The right channel for an urgent safety alert is not the right channel for a CEO video. The right channel for a local outage is not the right one for a monthly round-up. Build around people’s real work and attention patterns, not the comms team’s publishing calendar.
Governance is essential
Teams that distribute well know who can send what, through which channels, to which audiences, with what approvals, limits and safeguards. Things like:
An Audience Attribute Catalogue so targeting is based on agreed data rather than guesswork or improvised mailing lists
Pre-send checks to confirm audience logic, channel choice, accessibility and links back to the canonical source
Post-send checks to see what happened afterwards: not just who opened something, but who acted, where blind spots appeared, and which segments were underserved.
Prescribe vs subscribe
Giving people more agency over what they receive boosts relevance, and impact. There will always be things people have to receive. Target these to people based on who they are: by role, location, or simply something that must go to all employees. This is what we call prescribe, and it’s what you need for safety alerts, legal changes, mandatory compliance, operational disruption, anything where missing it carries real risk. Other communication should be subscribe: newsletters, interest-based updates, optional communities, the useful but non-essential material that people want on their own terms.
Good governance is what keeps distribution respectful as well as effective. Suppression rules stop people being hit repeatedly through overlapping audiences. Privacy guardrails stop targeting from becoming creepy. Regular audits with colleagues in HR and IT keep the underlying data honest enough to trust. And the measurement question is settled before the send, not invented afterwards. Opens and views may tell you that something appeared on a screen. They do not tell you whether someone completed the task, booked the training, acknowledged the update, or changed what they did next.
The diagnostic test is simple. If you published something critical right now, could you reach the people who actually need it, on a channel they would actually see, without spamming everyone else? And once you sent it, would you know whether they got it and what they did with it?
If either answer is no, the gap is not in your writing. It’s in your distribution layer.
Because publishing was never the same as being read. The job is to help the right people see the right thing at the right moment — without overwhelming everyone else.
Digital Communications at Work goes deeper on the Distribute layer in Chapter 5: from targeting and channel strategy to orchestration, governance and measurement.
