Six Places to Look. No Place to Trust.

Picture a community center getting ready for a Saturday event.

The coordinator changes the opening time on Thursday. The calendar gets updated. A text goes to the volunteers. Someone posts a reminder in the group chat. The printed sheet by the door still has the old time. An email from Tuesday is sitting in everybody’s inbox.

Saturday morning, half the volunteers arrive early. The person with the keys arrives at the new time. Nobody is trying to cause a problem. Everybody is following a message they were given.

Now the coordinator has a decision to make. Blame the people who missed the update, or admit that the group was never told which message counted.

This is a constructed example, but the problem is ordinary. A message can be sent six times without anybody knowing which version is true.

You do not have a communication problem just because people did not read your message. You may have a system that forces them to solve contradictions you created.

THE JOB SHOULD NOT BE A SCAVENGER HUNT

It sounds like a small distinction until you watch what people do when two instructions disagree.

They ask a coworker. They look through old messages. They call the person who wrote the last version. They wait for someone with a title to break the tie. Or they pick the version that seems safest and hope it was right.

You can call that checking. You can call it diligence. But if the same question comes up every week, your people are doing unpaid detective work before they can do their actual work.

The mistake begins when a leader treats delivery as proof of understanding.

I sent the email.

I told the team.

It was on the board.

We covered it in the meeting.

Those sentences prove that something left your end of the line. They do not prove that the next person could find it, recognize it as current, or act on it.

A handoff has two ends. The message has to survive the trip.

A RECENT CLUE, NOT A UNIVERSAL VERDICT

On October 27, 2026, Staffbase published findings from a Forrester Consulting survey commissioned by Staffbase. The underlying August survey covered 189 frontline workers and managers at US enterprises.

More than half reported getting the information they needed through six or more channels. Thirty-five percent reported encountering conflicting, outdated or unclear information at least weekly. The study also recorded self-reported effects on productivity, errors and decisions.

That is a warning worth examining. It is not proof that six channels cause mistakes, or that buying one new app fixes a company. The sample is small, from one surveyed population, and the sponsor sells employee communication software.

The useful question is not how many apps your organization has.

It is this: When two answers disagree, how does the person doing the work know which one wins?

If the answer is ‘ask me,’ you have made yourself a necessary part of every routine decision. If the answer is ‘check the chat,’ you may have made yesterday’s conversation compete with today’s standard.

Neither is a reliable way to run a shop, a family, a school or a volunteer team.

THE PERSON WHO KNOWS WHERE EVERYTHING IS

Every group has someone who seems to know everything.

The shop has the technician who remembers which version of a procedure is current. A school has the teacher who knows which schedule was changed. A family has the adult who remembers appointments, passwords, permission slips and who promised to pick someone up.

That person looks like a gift.

They often are. They have paid attention. They have built relationships. They know how things actually work.

But a group can become dependent on that person’s memory without noticing. When they are out, the work stops. When they leave, everybody discovers that the instructions were never really shared. They were stored in a human being.

There is a real psychological idea nearby. Daniel Wegner and colleagues studied how close partners divide up what they remember. One person knows one set of things, the other remembers another, and the pair can work as a memory system. That research does not mean a workplace should write nothing down. It helps explain why asking the one person who knows can feel easier than making knowledge usable for everyone.

The weakness is not that people specialize.

The weakness is building essential work around one person’s availability, then calling the failure a communication mistake when that person is absent.

A good system protects expertise without turning the expert into a search engine.

ONE HOME. MANY DOORS.

The fix is not to ban chat, email, whiteboards or conversation. They do different jobs.

A chat can tell someone that a decision was made. A calendar can show when it applies. A printed instruction can help someone working without a device. An email can reach a person off shift.

Those are doors.

The authoritative instruction needs a home.

When the underlying answer changes, update the home first. Then use the doors to point people to it. If you must print or copy it, mark the copy so people can tell whether it is current.

That is different from putting the same paragraph in six places and hoping all six stay correct.

The principle works without an expensive platform. A clearly dated binder can be better than a beautiful app nobody can open. A shared calendar can beat a chain of family texts. A simple page with an owner and a last-reviewed date can beat a folder full of files named FINAL, FINAL2 and REALLY FINAL.

The right tool is the one people can reach at the moment they need the answer.

ASK WHO CAN ACTUALLY USE IT

A leader who works at a desk may love a digital link. The person who needs it might be standing outside in gloves, working a late shift, sharing a device, or reading a second language.

One authoritative home does not mean one rigid way of reaching it.

Make the current information reachable by the people who actually use it. Check that a printed copy points to the current instruction. Check access from the floor, from home, from the back room, or wherever the decision gets made.

When the cost of an error is high, do not rely on a casual message as a replacement for required controls, approved procedures or legal obligations.

The point is not to make every instruction informal.

The point is to make the correct instruction unmistakable.

A LEADER’S TEN-MINUTE TEST

Pick one instruction that has been corrected or explained more than once this month. Not the biggest problem. One ordinary piece of work that keeps generating the same question.

Spend two minutes writing what the current answer is.

Spend two minutes asking two people who use it: ‘Where would you look for the current version?’ Do not tell them where to go. Watch what they actually do.

Spend two minutes comparing what they find. If the answers differ, you have found the problem. No need to accuse anybody of missing the memo.

Spend two minutes choosing the one place that will hold the answer from now on. Put an owner and a date on it. Make the older version clearly obsolete. Do not quietly erase records that must be kept.

Spend two minutes telling people where the current answer lives, then have one person find it without your help.

That last minute matters. The test is not whether you can explain your system. The test is whether it works when you are not standing there.

The next time the answer changes, update that home before sending another message.

THREE DIFFERENT PLACES TO SEE IT

At home, the problem may be an appointment changed in one text thread while the family calendar stays wrong. One person has to keep reminding everyone. The fix is not another reminder. The fix is that the calendar gets changed when the plan changes, and messages point back to it.

In a classroom, the problem may be an assignment due date on the board, a different date in an online system, and a third date mentioned aloud. Students get called careless for believing the wrong one. The teacher can still make a change. The teacher also has to make clear which version governs.

On a sales team, the problem may be a customer promise repeated from an old email. A salesperson needs to know which price or service terms are authorized today. A current policy and a clear decision owner are stronger than ten messages saying ‘check with me first.’ Financial and contractual commitments still require the proper authorization.

And in a small shop, the person who can do the work should not spend the morning calling around to find out which drawing or instruction is current. The owner can demand the correct result. The owner also has to make the correct input available.

These are built examples. They are not stories from my career.

THE THINKERS ARE PARALLELS, NOT EXCUSES

Mary Parker Follett argued that people should study the actual situation together rather than make every order a contest of rank. Her point was about responsibility and authority, not software or document control. But the parallel holds: the correct next move should come from a situation people can see, not from guessing which boss’s message wins.

Wegner’s work on shared memory offers another lens. People naturally learn who knows what. That can make a group capable. It becomes fragile when the answer to every recurring question is a person’s name instead of something the group can find.

A hypothetical shop owner would make it plainer: ‘If I go fishing tomorrow, will anybody know which version to use?’

All three questions aim at the same test. Does the work still move when the person who last explained it is gone?

THE OBJECTION

Someone will say, ‘We cannot write down everything.’

Correct.

Do not write everything down. Write down what people need to do repeatedly, what changes often enough to confuse them, and what becomes costly or dangerous when they guess. Keep the current answer short. Archive the old one clearly. Give someone ownership of the update.

Someone else will say, ‘People should pay attention.’

Also correct. Adults have a responsibility to listen and follow the instructions they have been given.

But attention cannot settle two official answers that disagree. A leader cannot send competing instructions and then measure commitment by who guessed correctly.

One more objection: ‘We already have one system.’

That is a software answer to a work question. Go to the point where the answer is needed. Ask someone to find it. A platform is not a standard until people can use it.

WHAT CHANGES TOMORROW

Stop asking whether the memo went out.

Ask whether the next person can find the answer and tell that it is current.

Pick one recurring question. Give its answer a home, an owner and a date. Point every other channel there. Then leave the room and see whether somebody else can find it.

When that works, the team has gained something more useful than another message.

It has one less reason to wait for you.