28 October 2026 · 13 min read

Working in silos: why more meetings are not the fix

Nobody builds a silo. Each team gets very good at its own week, and the stretch between two teams ends up belonging to nobody.

That is why you rarely find one inside a team's own work. You find it at the handoff.

The promise that did not cross

Priya runs onboarding at a software company of about two hundred people. On Tuesday morning, a few minutes into a kickoff call with a new customer, their finance director says, "And the data migration is included, as Wen agreed." The handoff note from sales says "standard package". Priya says, "Let me come back to you on that," which the finance director hears, correctly, as a no.

After the call she messages Wen. He answers in under a minute, with the whole story: what he promised, why, and the date he gave. He had known all of it the entire time. He mentioned it in the deal channel three weeks ago, in a thread Priya was never added to, and the handoff form has no box for "things I promised that are not in the package".

Nobody was careless. Wen's team is measured on deals closed and Priya's on customers live on time. Each did the job as it is written. The one thing that needed to cross was the thing neither job described.

How much information fails to arrive?

Wen is invented. This survey is not.

In the CIPD's 2026 survey, 26% of employees rated their manager poor or very poor at keeping employees informed of management discussions or decisions, and 41% rated them good or very good (CIPD, Good Work Index 2026: Survey report, July 2026, Figure 41).

That is about information coming down from a manager, not two-way communication or two teams, so it cannot measure Priya's problem. It does show that information someone is meant to pass on does not always arrive.

Why do teams end up working in silos?

A team gets better by narrowing its attention: its own measures, its own words, its own heroes. The side effect is that the neighbouring team slowly stops being a group of people and becomes a label, and a label is easy to blame and hard to help. Priya's team has a word for what sales does to a handoff. Wen's has one for what onboarding does to a date. Neither has been tested on the people it describes.

None of the research below counts silos. It measures sharing, and what sharing is linked to.

Finding is easy and carrying is hard

Across 120 new-product projects in 41 divisions of one large electronics company, weak ties between divisions helped project teams find useful knowledge elsewhere in the company, but slowed projects down when the knowledge that had to be transferred was highly complex (Hansen, Administrative Science Quarterly, 1999). A weak tie is a thin, infrequent relationship rather than a close one. When the knowledge was simple, the same ties speeded projects up.

Priya could find the answer in a minute, which is what weak ties are good at. What failed was the carrying. That is my reading across from product projects to a sales handoff, not Hansen's finding.

The more different the people, the more the sharing is linked to results

In a field study of 182 work groups in a Fortune 500 telecommunications firm, sharing knowledge with people outside the group was more strongly associated with performance, as rated by senior executives, in groups whose members came from more varied locations, functions, managers and business units (Cummings, Management Science, 2004). It is one firm, performance means executives' ratings rather than output, and it does not say that sharing with outsiders pays off everywhere.

Inside a team the evidence is broader

A meta-analysis of 72 independent studies (4,795 groups, 17,279 people) found that information sharing within a team was positively related to team performance, and to cohesion, decision satisfaction and knowledge integration (Mesmer-Magnus and DeChurch, Journal of Applied Psychology, 2009). That is sharing inside one team, not across departments, so it says nothing about silos alone. It sits here because the other two studies are about the line between teams.

Together they support a narrower claim than "silos are bad". Sharing is linked to better performance, inside a team and across the line, and across it the link is stronger where the people involved are more different. Crossing is not free, though: finding what is over there is easy, and carrying something complicated across is not.

How to tell a silo mentality from a busy week

Every team is busy. These are the signs that the busyness has a wall in it.

• The next team over is "them". Listen for it in the first five minutes of any planning meeting.
• A handoff feels like a border crossing. There is paperwork, and nobody on the far side is expecting you.
• The same dispute keeps coming back. It lands wherever the two teams' work meets, under a new name each quarter.
• Jargon goes untranslated. "It's blocked on the freeze" arrives, and some of the listeners quietly hear "excuse".
• Nobody answers when the other team is run down. The silence is taken as agreement.
• One or two people carry the link by hand. Everybody knows who to ask, and nobody has written that down.

What working in silos costs, and who carries it

The customer pays first, because a customer meets one company. A gap between two teams reaches them as a delay or a question asked twice, and neither team may be watching for it. Priya's finance director has been a customer for a few minutes and has already learned that two parts of the company do not talk.

Then there is the cost nobody books. Every company has a Joy. She has been in support for six years and knows who to message in every team, so when a customer is stuck between sales and onboarding somebody says "ask Joy". That is on no job description and no scoreboard, and people doing it tend to wear out. Her review says she is helpful. If she left, the company would find the line the way you find a load-bearing wall.

Should you break down silos at all?

Breaking down silos is the phrase, and it asks for slightly the wrong thing.

Nobody needs Wen's team and Priya's to become one team. Sales should think about deals and onboarding should think about go-live dates, and a company full of people who all think about everything has a different problem. Cummings's finding points the same way: the link between outside sharing and performance was stronger where groups were more varied, so difference may be where crossing pays best. I read that as an argument against flattening the walls, and Hansen's as an argument against thin contact: a tie good enough to find something may not be good enough to move it.

So the job is smaller and more specific than the phrase suggests. Find the few places where something complicated has to cross, and put a door there with a person on each side. The rest of the wall can stay. Understanding the other team's view does not require agreeing with it, and it does not require a reorganisation.

Why another meeting does not fix it

The two easy answers both raise the amount of contact: a standing cross-team meeting, and a shared channel.

The leads attend the meeting. They already know each other and already know the answer to "are we aligned?". Wen's promise was made by one person to one customer on a Tuesday, and it was never going to reach a lead's agenda. The channel carries so much that people mute it, and the one message that mattered sits in a thread nobody added Priya to.

The research on communication points the same way. A meta-analysis of team communication and performance found that communication quality had a significantly stronger relationship with team performance than communication frequency (Marlow, Lacerenza, Paoletti, Burke and Salas, Organizational Behavior and Human Decision Processes, 2018). It is a study of teams, not departments, and cannot say how much contact is right. It is a reason to doubt that more is the answer.

At a boundary, quality is specific: what has to cross, who holds it, and where the person receiving it will look.

How to improve cross-team collaboration tomorrow

None of this needs a product. Three things for this week.

Walk one failed handoff backwards

Take the last piece of work that crossed between two teams and went wrong. Put the person who received it and the person who sent it together, and ask the receiver: "What did you need on the first day that you found out later?" Then ask the sender where they said it. Wherever the answer is "in a thread" or "in my head", the handoff is missing a field. Add it, written by the person who holds the knowledge, in the place the receiver already looks.

Give the line two names, not a channel

Name one person on each side as the other team's first call, and tell both teams. A channel is a place to post; a pair is two people who will answer. Ask them to talk briefly before any piece of work crosses the line. This is my reading of Hansen, not something he tested: a tie that can find an answer is not yet one that can carry it.

Ask what number the other team is measured on, and tell them yours

Disputes at a boundary are often two scoreboards argued over one piece of work. Wen's promise made sense against deals closed; the standard package made sense against customers live on time. Write both numbers on one page. You may not agree which matters more, but you can read each other's behaviour against a number instead of a motive. If the numbers truly pull in opposite directions, that is not a communication problem, and it goes to whoever sets them.

All three fail the same way. They rely on somebody remembering in the busiest week.

Three questions about silo mentality

What is a silo mentality?

A silo mentality is a team working as if it were sealed off from the teams around it, usually without anybody deciding it should. The neighbouring team becomes "them", and every unclear move gets a motive attached. It is rarely ill will. Mostly it comes from never having watched how the other team really works, so a caricature fills the space.

Do silos only happen in big companies?

No, because a silo needs two teams with two scoreboards, not a big headcount. Priya's company has about two hundred people, and a firm of forty with sales and delivery measured differently has the same line. Size changes how long a problem can sit there unnoticed, and a company that has just doubled can grow one in a year.

Who should own a handoff between two teams?

The person who holds the knowledge writes it, and the person receiving it says what it must contain. Leaving either half to the other side is how a form ends up with no box for the thing that mattered. Then name a pair to own the gap itself, because anything owned by both teams together tends to be owned by neither.

Where MeaningFuel helps with working in silos, and where it cannot

What reaches both sides of the line every morning

A company chooses the values and skills it wants to work on, and each working day everyone gets one short insight on one of them. A silo is a problem, so it cannot be chosen for the daily culture, but Collaboration, Communication and Teamwork can. In an ordinary week a reader might meet a line on what listening costs on Tuesday and one on asking early on Thursday. Sales and onboarding do not need to agree on anything to have one shared thing to have noticed, which is more than many neighbouring teams have.

Two modules, one for each side of the problem

MeaningFuel has two modules for this problem, covering different halves of it. A module is a short course of daily lessons for a team's named people: booked for a team, it invites them, and those who join get one short lesson each working day.

Working Well Together is for a team whose own members work side by side rather than together. It borrows an idea from CREW (Civility, Respect and Engagement in the Workplace), a public programme from the US Veterans Health Administration: norms the group chooses itself and practises daily. In the lesson on asking, each person sets a struggle budget, the time they will struggle alone before asking, and when it is spent they ask with their working shown: "stuck on X, tried A and B, need a second pair of eyes." That is a habit inside one team, the level the meta-analysis above covers, and it does not cross a line between teams.

Working Across Teams is for the line itself, and is booked for the team that wants to change its side of it. It draws on research by Adam Galinsky, who found that deliberately imagining how someone else sees things reduces how far people rely on stereotypes about other groups. So in the second lesson each person writes the neighbouring team's week as that team would tell it - their Monday, the first thing that goes wrong, the number they are measured on - and sends it to a counterpart with one line: "is this close?" Their corrections replace a guess with what the other team actually says. Later comes a joint planning session with the other team, working out how the shared work could fail while that is still cheap to change, starting from the customer's side, because a customer meets one company. Where the two teams' incentives really do pull in opposite directions, the module says so: it asks for one sentence to go up to leadership with both numbers attached, because practice cannot fix a structure.

What Insights reads, and the line it cannot see

Insights reads teams one at a time: how much of each takes part and how the insights land, pooled, with nobody named. A silo lives between two teams, so it is close to invisible there. Wen's team and Priya's could both look healthy on the morning of the kickoff. It would show you whether Working Across Teams is being done, and it would never show you whether a promise crossed. For that you still need the customer, or Priya.

What it does not do

Neither promises an outcome, and neither changes what anyone is measured on. They put the habits in front of the people at the line every working day, so none depends on whoever remembered this week.

Back to the kickoff

Wen had the answer on Tuesday. What the company lacked was a place on the handoff for it to land, and a person on the other side whose job was to ask. Neither needs another standing meeting.

• Working in Silos
• Silo Mentality
• Breaking Down Silos
• Collaboration
• Workplace Culture