Up and Down the Chain

Communication best practices for executive leaders, middle leaders, and the front line.

The premise. Cascading messages down and flagging information up are not two separate skills — they’re one system. If leadership doesn’t get honest information from the middle, the messages they send back down are built on an inaccurate or incomplete picture. Fix one direction and you improve the other.

Why messages break

Almost no one intends to miscommunicate. Messages break for structural reasons: they pass through more hands than anyone accounts for, each handoff strips a little context, and the people asked to carry a message are often the last to understand it themselves. Meanwhile, information moving upward gets filtered through the very real question of what’s safe to say. None of that is a character problem. All of it is fixable by design.

PART ONE

Cascading down: making the message hold

What actually goes wrong

• The telephone game. Each layer paraphrases, and the meaning drifts. By the third retelling it’s a different message.

• The “why” falls off. Executives share their reasoning with middle leaders; middle leaders pass down only the decision. The front line hears an arbitrary directive and fills in a motive — usually a worse one than the truth.

• No time to metabolize it. Middle leaders are asked to deliver a message they haven’t had time to process, let alone question. They read it flatly, or add a spin that contradicts a peer’s version.

• Ragged timing. Some teams hear it in an all-staff, others from a manager a week later, and plenty hear it through the grapevine first. Whoever speaks first sets the meaning.

• No return path. Cascading gets treated as a broadcast. Questions surface at the front line, die there, and never travel back up.

The practices that fix it

Brief the middle before they have to deliver. This is the single highest-leverage practice on this page. Give middle leaders the message early enough to ask real questions and push back privately, before it’s public. A leader who has processed a message delivers it with conviction. A leader hearing it for the first time alongside their team delivers it with a shrug.

Send the why with the what. People can carry hard news. What they can’t carry is unexplained news, because the mind fills the gap with the worst available story. If the real reason is uncomfortable, say the uncomfortable thing — a sanitized reason gets detected and costs you more.

Give a message core, not a script. A script sounds robotic and gets abandoned; free-form drifts. Give the few sentences everyone must convey accurately, then let each leader translate it for their team’s reality. Aligned, but human.

Name what you don’t know. “Here’s what’s decided, here’s what isn’t, and here’s when we’ll know” builds more trust than false certainty. Pretending to have answers you don’t is the fastest way to lose the room.

Decide the sequence on purpose. Who hears what, from whom, in what order, and inside what window. The goal is that no one learns something that affects them from a rumor.

Arm leaders with the hard questions. Anticipate the five questions people will actually ask — including the uncomfortable ones about jobs, workload, and whether this will stick — and give leaders real answers, including “I don’t know yet, and here’s how I’ll find out.”

Say it more than once, in more than one channel. One all-staff email is not communication. People absorb change on a delay and through repetition.

Check for understanding, not agreement. “Any questions?” gets silence. “What are you taking away from this?” and “What will you tell your team tomorrow?” tell you what actually landed.

Close the loop out loud. A cascade isn’t finished when the message is delivered. It’s finished when the questions have traveled back up and been answered where people can see it.

Tool · The Message Core

• What’s changing. In one or two plain sentences, no jargon.

• Why. The real reason — including the uncomfortable part.

• What it means for your team specifically. The translation each leader makes for their own people.

• What is not changing. Often the most reassuring line in the whole message.

• What we don’t know yet. And when we expect to know.

• Where to bring questions. Named person, named channel, named timeframe.

The channels — and the order to use them

Most people hear “use multiple channels” and think “an email and a meeting.” The real list is far longer — and the channels that reach your most distant staff are usually the ones leadership uses least.

What actually counts as a channel

• Live, in person. All-staff meeting or town hall · department and team meetings · shift huddles and start-of-shift briefings · supervisor one-on-ones · skip-level conversations · small-group listening sessions · floor rounds and site visits · open office hours · orientation and training.

• Live, remote. Video all-staff · video team meetings · phone calls · live chat huddles · webinar with live Q&A.

• Written, digital. Email — and it matters whether it comes from the ED or from someone’s own supervisor · staff newsletter or digest · intranet page · chat platform post or pinned message · text message · portal or LMS announcement · a living FAQ document · a note in the project tool.

• Recorded. A short leader-to-camera video · a voice memo · the recording of the live session · a screen-share walkthrough.

• Physical. Break-room and bulletin board postings · a notice at the time clock · printed handouts and mailbox drops · paycheck inserts · signage at the point of work · the team whiteboard.

• Relational. The supervisor cascade — the most powerful channel you have · peer influencers and informal networks · union or staff association representatives · staff committees and employee resource groups · and the grapevine, which runs whether you plan for it or not.

• Return channels. Q&A in meetings · an anonymous form or question box · a named contact person · supervisors collecting questions on the team’s behalf · pulse surveys · office hours.

The order to use them

There is a strategic order, and it follows one principle: proximity to impact determines sequence. The more a person is affected, the earlier and more personally they hear it.

1. Leadership and decision-makers align first — so nobody contradicts anybody in week one.

2. The middle gets the pre-brief, with real room to ask questions and push back before it’s public.

3. Anyone personally affected hears it one-on-one or in a small group, before the general announcement.

4. The full staff hears the general announcement, live where possible.

5. The durable written record goes out, so people can go back and check what was actually said.

6. The catch-up reaches people who weren’t working — next shift, part-time, relief, anyone on leave.

7. External audiences last: board, funders, partners, families, public.

Two rules that hold every time. No one should learn about themselves in a room full of other people. And staff hear it before outsiders do — an employee finding out about their own organization from a funder email or a news story will remember it for years.

Match the channel to the weight of the message

Routine information. Written and asynchronous is fine — a digest, a posting, a chat note.

A change to how people work. Live team conversation first, written follow-up second. People need to ask before they need a record.

Anything that affects a specific person. One-on-one, always, and first.

Crisis or safety. The fastest push channel you have immediately — text, huddle, phone tree — with a live follow-up within hours.

The underlying rule: the harder the message, the more human the channel. Good news survives an email. Hard news delivered by email reads as avoidance, and people remember the choice of channel as part of the message itself.

Who’s furthest from the information?

A cascade designed for the salaried desk worker quietly excludes a lot of people: staff on other shifts, part-time and relief staff, field and outreach workers, people without regular computer access, people whose first language isn’t English, staff at satellite sites, newer employees who don’t know the informal channels, and volunteers. Before you send anything, ask who gets it last, who gets it thinnest, and who has the least power to ask a clarifying question.

And here’s the part that makes this operational rather than merely fair: in most mission-driven organizations, the people furthest from the information are the people closest to the clients. A cascade that fails them doesn’t just feel exclusionary — it degrades service where it’s most visible. The staff member who finds out about a policy change three weeks late is the one explaining it to a family, badly, in real time.

Do this once: map how information actually reaches people

Most communication failures repeat because nobody has ever written down how information physically travels. Build the map once and reuse it for every announcement. For each group of staff, note four things:

• Where do they actually receive information — email, shift huddle, supervisor, break-room posting, text, a note in a mailbox?

• When are they present — which days, which shifts, and how long until the next time they’re on site?

• What language do they work in, and what’s their access — screen reader, reading level, device?

• Who do they trust to tell them things?

Ask the groups themselves rather than guessing; the answers are usually a surprise. Keep the map with your communication plan, and update it when staffing changes.

Channel — meet people where they already are

Stop treating email as the cascade. For anyone who doesn’t sit at a computer during their shift, an email is a message that arrives after the fact. Identify the channel each group truly uses — the shift huddle, the supervisor, the time clock, the break room, a text.

Use two channels for anything that matters. One that pushes (a huddle, a text, a supervisor conversation) and one that stays put (a posting, a printed handout, the portal) so people can go back and check what was actually said.

Put it where they physically stop. A posting at the time clock, the med room, or the break room reaches non-desk staff better than any intranet page ever will.

Consider a two-minute recorded message. For people who won’t read a long email, a short video or voice note from a leader travels well — and caption it so it works for everyone.

Timing — close the gap

Map the schedule before you pick the moment. An 11 a.m. Tuesday all-staff excludes the overnight team, weekend staff, anyone in the field, and most part-time employees. The meeting time is a decision about who counts as present.

Set a no-gap window. Decide the maximum acceptable time between the first person hearing and the last — 24 or 48 hours is a reasonable target — and build the plan backward from that.

When a gap is unavoidable, name it. “The evening team hears this at seven tonight — please hold it until then.” People will generally honor a clear, specific ask. They can’t honor one you never made.

Record the live session. So the people who couldn’t be there get the same words, not a secondhand summary that’s already drifted.

Plan for the ones who aren’t in this week. Relief and per-diem staff may not work for ten days. Decide now how the message reaches them on their next shift, and who’s responsible for it.

Language and access — remove the barriers you didn’t build on purpose

Write plainer than feels necessary. Cut acronyms and internal shorthand. They exclude new staff as efficiently as they exclude anyone else, and they’re invisible to the people who write them.

Translate the message and the questions channel. Translating an announcement while leaving the “bring your questions here” path in English only tells people they may listen but not speak.

Don’t lean on bilingual staff to interpret informally. It’s unpaid labor, it puts them in the middle of a message they didn’t write, and it distorts the content. If interpretation is needed, resource it properly.

Check accessibility before you send. Caption video. Don’t bury critical information in an image or a scanned PDF a screen reader can’t read. Keep formatting simple enough to survive a phone screen.

Strip the assumed context. “As discussed at the leadership retreat” means nothing to someone who wasn’t there — and quietly signals that the message wasn’t written for them.

Messenger — proximity beats seniority

The most trusted messenger is usually the direct supervisor. Which is exactly why the pre-brief matters: equip supervisors first so the message arrives from someone the person already trusts.

For the most distant groups, go to them. Showing up on the overnight shift or at the satellite site is itself information — about whether that team is considered part of the organization.

The return path — make it possible to ask

The people furthest from the information also have the least safe route to ask a question. An open door in the executive suite is not a channel for a part-time overnight worker.

Offer more than one way to ask. A supervisor, a named person, and an anonymous option. Different levels of risk tolerance need different doors.

Ask in small groups. People ask real questions in a group of five. In a group of fifty, they wait to see who goes first — and usually nobody does.

Let supervisors carry questions up on the team’s behalf. So no individual has to be the person who raised it.

Answer back through the same channels. If it went out at the huddle, the answer comes back at the huddle. Otherwise only the email-readers learn what happened.

Check that it landed

Spot-check three people in different roles and shifts. Ask what they took away. If they can’t say it back, the cascade failed regardless of how well it was written.

Notice who never asks anything. Silence from a whole group is a signal, not a comfort. It usually means the question channel doesn’t work for them, or doesn’t feel safe.

Run a short “who did we miss?” review after big announcements. Then fix the map, so the same group isn’t last again next time.

Tool · The pre-send check

• Who hears this last, and how much later?

• Who can’t reach this channel at all?

• Who has to read this in a second language, or on a phone?

• Who can’t safely ask a question about it — and what’s their alternate route?

• What does it cost this group to find out late?

• Who is responsible for reaching the people who aren’t working this week?

What am I supposed to relay?

Leaders routinely assume the middle knows which parts of a message are theirs to pass on. They usually don’t. A manager walks out of a meeting holding a mix of firm decisions, background context, half-formed plans, and a few things said in confidence — with nothing marking which is which. Faced with that ambiguity, some over-share and get ahead of an announcement, some sit on information their team badly needed, and most default to silence, because silence feels like the safer mistake. It isn’t. It just moves the cost onto the front line.

Label everything — never make them guess

One small convention solves most of this. Every piece of information handed to a middle leader should arrive with a label attached.

Tool · The four labels

• Share now. Pass this to your team by [date], in your own words.

• Hold until [date or trigger]. Then share it. Here’s why it’s being held.

• Yours to know. Context to inform your decisions — not for distribution.

• Confidential. Don’t share it — and here is exactly what to say if someone asks.

Then state the default out loud: unless something is labeled otherwise, assume it’s shareable. Without a stated default, the default becomes silence.

What every handoff to the middle should include

• The label — and if it’s shareable, the deadline for relaying it.

• Verbatim or in your own words: say which. Dates, policy language, and legal wording usually need to be exact. Everything else should be translated for the team.

• The channel you expect them to use — “in your team meeting, not by email.”

• The five questions their team will actually ask, with real answers.

• What they’re allowed to say about the things nobody knows yet.

When you can’t share it yet

Say why it’s being held. People honor a restriction they understand, and a reason stops them from inventing a worse one.

Give it an expiry. “Hold until Thursday’s announcement” is honorable. “Hold indefinitely” turns a middle leader into a permanent secret-keeper and quietly corrodes their relationship with their team.

Give them the words. This is the piece almost everyone skips. “I can’t speak to that yet, but I’ll tell you as soon as I can — and I’ll make sure you’re not the last to know” is a competent answer. Without it, people freeze or improvise something worse.

Keep the window short. Asking someone to sit in a meeting about their own team’s future and then act normally for three weeks carries a real human cost. Minimize it deliberately.

What to send back up

The middle also needs to know what’s worth escalating. Left to guess, leaders either escalate everything and become background noise, or escalate nothing and leave leadership blind. Send it up when:

• It affects client care, safety, or rights. Always, and immediately.

• It creates legal, compliance, or regulatory exposure — including something that could become one.

• It’s a pattern, not an incident. One complaint is information; the third identical one is a signal.

• It crosses a boundary you can’t resolve at your level.

• A commitment leadership has already made can’t be met as scoped — flag it while it’s still small.

• Trust or retention is shifting: a key person is at risk of leaving, or a team is losing faith in a decision.

• Something is working that a pending decision could accidentally break (see Part Two).

• Leadership would otherwise be blindsided. The test: would they be upset to learn this first from a funder, a board member, or the news?

Handle it yourself when it sits inside your authority and resources, when it’s a one-off you can resolve, or when it’s venting with no decision attached — until it becomes a pattern. For how to frame an escalation once you’ve decided to send it, see Part Two.

When you disagree with the message you’re carrying

This is the hardest spot in the middle seat, and it deserves a straight answer rather than a platitude.

Raise it before, not after. The pre-brief window exists for exactly this. Disagreement voiced privately beforehand is a contribution. The same disagreement voiced to your team afterward is a fracture.

Represent the decision accurately without faking enthusiasm. “Here’s the decision, here’s the reasoning, I raised these concerns, here’s where we landed, and here’s what I need from you” is honest and undermines nothing.

Don’t say “they’re making us do this.” It buys a moment of rapport at the cost of your own authority — and it teaches your team that you’re not actually part of the organization making decisions about their work.

And if you’re being asked to deliver something you believe is genuinely wrong rather than merely unwelcome, that’s a different conversation. It belongs upward, and it belongs there before delivery.

PART TWO

Communicating up: risk and bright spots

Information moving up fails in two opposite directions. Some middle leaders escalate only problems, and become background noise — leadership learns to discount them. Others shield leadership from bad news until the problem has snowballed beyond what the team can manage — by which point it’s harder and more costly to fix. Both leave executives making decisions from information that’s inaccurate or incomplete.

Why bright spots belong in every update

Most guidance on managing up is entirely about escalating problems. That’s half a practice, and the missing half matters more than people think.

• What works is fragile. What’s working often survives by accident — one person’s discretionary effort, an informal workaround, a scheduling quirk. A well-meant policy change, budget trim, or reorg can kill it without anyone realizing. Leadership can’t protect what it doesn’t know exists.

• What works is scalable. Something working in one team is usually the cheapest available fix for the whole organization. But it can only spread if it’s visible above the team that invented it.

• It buys your risk flags credibility. A leader who only ever brings problems gets filed as a complainer, and their warnings get discounted precisely when one is urgent. A leader who reliably reports both is read as a trustworthy narrator — which is what makes a risk flag land.

Flagging risk well

Flag early, while it’s still small. The value of a warning decays fast. A risk named three weeks out is a planning conversation; the same risk named three days out is a crisis with someone to blame.

Lead with the bottom line. The risk, the impact, the ask — in the first two sentences. Then the detail, for whoever wants it.

Give it size and shape. What could happen, how likely, by when, and who it lands on. “I’m worried about the timeline” is a feeling. “We’ll miss the 31st by about two weeks, which delays 60 client intakes” is a risk.

Bring the tradeoff, not just the alarm. Name what can move — date, scope, or resources — and say which one you’d choose. Arriving with a recommendation is what separates a leader from a reporter.

Say what you need from them. A decision? Air cover? Money? Or just awareness so they’re not surprised later? Executives frequently can’t tell which is being asked, so they do nothing. Name it explicitly.

Don’t sand off the edges. “A slight concern” when you mean “this will fail” is not diplomacy; it’s a failure to be forthright, and it transfers the consequences to someone else.

Flagging bright spots well

Be specific, not warm. “The team is doing great” is unusable. “Since we moved intake reviews to Tuesdays, our turnaround dropped from nine days to four” can be protected and copied.

Say what makes it work. Name the conditions, not just the result — the staffing, the timing, the autonomy, the one person who holds it together. Conditions are what leadership can actually preserve.

Say what would break it. This is the line that converts a nice update into a protection request: “This works because Devon has two uninterrupted mornings a week. If we add another standing meeting, it stops working.”

Name what you’re asking for. Protect it, resource it, spread it to other teams, or simply know about it. Same discipline as a risk flag.

Attribute it to people. Bright spots are almost always somebody’s discretionary effort. Naming them upward is both accurate and one of the cheapest retention acts available to you.

Tool · The Monthly Flag-Up

• What’s working — and what would break it. Two or three bright spots, with the conditions that make them work and the specific ask: protect, resource, spread, or note.

• What’s at risk — and what I need. Impact, likelihood, timeframe, who’s affected, the tradeoff you’d recommend, and the decision or cover you need.

• What I’m watching. Early signals that don’t need action yet. This is how you build a track record of seeing things coming.

Send it on a rhythm — monthly is usually right — so that upward communication isn’t only crisis-driven. A predictable structure trains leadership to read it, and trains you to look for bright spots instead of only firefighting.

PART THREE

What executive leadership owes the middle

Nearly all advice about “managing up” places the burden on the person with less power. That’s incomplete and a little unfair. Whether middle leaders tell the truth upward depends almost entirely on what has happened to the last few people who did.

Never punish the messenger. How you take hard news is a public act, even when the conversation happens privately — it gets retold. One leader treated badly for raising a risk teaches everyone watching to stay quiet, and that lesson travels faster than any encouragement to “speak up.” Watch for the quieter punishments too: handing the flagger the whole fix alone, new scrutiny of their area afterward, or leaving them out of the next decision.

Close the loop where people can see it. When someone flags something, say what you did with it — even if the answer is “we heard it and chose otherwise, here’s why.” Flags sent into silence stop being sent.

Ask for bright spots explicitly. People report what they’re asked about. If every check-in asks only what’s off track, that’s the only picture you’ll ever have.

Give the middle the why before the ask. You cannot expect leaders to carry a message down with conviction if they received it as a directive with no reasoning attached.

Protect what you learn is working. If a bright spot depends on two uninterrupted mornings, and you take those mornings, you taught your organization that reporting good news is how you lose it.

The loop. Down and up are the same circuit. Leaders who are briefed early and given the real why carry messages that hold. Leaders who see their flags taken seriously keep sending them — including the good news worth protecting. Neither half works for long without the other.