Talking to People Not in the Room

No affiliate links on this site, and nothing here is ranked.

The category with ten duplicates

The average company runs ten collaboration tools. (Zylo, a SaaS management vendor.)

Nobody chose ten, and unlike most duplication this one is not driven by purchasing — it is driven by whoever started the conversation choosing where.

A customer messages on one platform, a supplier on another, the team uses a third, and somebody's phone number is the fourth. All four are now channels, and none was adopted.

What the channels are actually for

Immediate and short. A message that needs an answer in minutes. Phone, or whatever messaging everybody already has.

Considered and recorded. Anything that decides something. Email, because it is dated, attributed and both parties hold a copy.

Working together at the same time. A call, with screens if needed.

And ambient — knowing what is going on. The category chat tools sell, and the one small firms need least, because a team of four already knows.

The rule that prevents sprawl

One channel per relationship, stated.

"Message me for anything urgent, email for anything we need a record of." Two sentences to a customer at the start of a job, and the four-channel problem does not arise.

Where a customer insists on their platform, use it — and forward anything that decided something into email, so the record exists in one place regardless of where the conversation happened.

That single habit is what keeps a job's history findable when it is spread across three apps.

What to check before adding another

Whether it replaces something. Adding without removing is how a company reaches ten, and every addition should retire something or be justified as genuinely distinct.

Whether the history is retrievable. Chat platforms commonly limit free-tier history to a period, which means the record of a decision expires.

What exports, which for chat is usually poor.

And who can see what. Channels, direct messages and guest access have different visibility, and getting it wrong is how a customer sees an internal exchange.

The meeting question

A call is a synchronous tool and its cost is everybody's time simultaneously.

Four people for thirty minutes is two hours. The comparison is not against a message; it is against two hours of work.

Where the matter is a decision needing discussion, that is frequently worth it. Where it is information transfer, a written note costs one person's ten minutes and everybody reads it when it suits them.

Recording and transcription change this calculation — an asynchronous record of a synchronous event — and are worth having where meetings are unavoidable.

What the free tiers cost here specifically

Message history, and it is the one category where that matters most.

A chat tool retaining ninety days of history is a tool where the decision made in March is gone in July — which is fine for coordination and not fine for anything that settles a question later.

This is why decisions belong in email regardless of where the conversation happened, and it is a habit rather than a purchase.

Where a paid tier is bought for history, check what the export produces before relying on it: chat exports are frequently poor and occasionally unreadable outside the tool.

Notifications, which decide whether it works

Every tool in this category defaults to interrupting.

A team of four with default settings receives an alert for everything, which trains everybody to ignore alerts — after which the urgent message does not get through either.

Set them deliberately, once, and tell people what the convention is: what warrants a notification, and what can wait until somebody looks.

The convention matters more than the tool, which is true of everything in this category and is why adding another one changes nothing.

The channel for people outside

Customers and suppliers should not be in your internal tool unless the tool handles guests properly and somebody has checked what they can see.

The commonest mistake in this category is a guest account with more visibility than intended, discovered when somebody outside comments on something internal.

Where collaboration with outsiders is routine, that is a requirement to check during a trial rather than an assumption.

The tool that is already there

Whatever the customer already uses.

A supplier who insists on their own platform for the convenience of one relationship is asking the customer to adopt something, which is a cost they did not agree to.

Meet them where they are and keep your own record separately. The record is the thing that must be consistent; the channel does not have to be.

Where a real tool earns its place

Three situations, and none is a team of four.

People who do not share a building or a schedule, where ambient awareness has to be manufactured because it does not happen naturally.

Work that must be visible to somebody who was not there — a handover, a shift, an on-call arrangement.

And enough people that a decision made in a conversation does not reach everybody, which is a threshold around a dozen rather than a matter of preference.

Below those, the tools are being bought for a problem the size of the firm has not produced yet, and they add a channel rather than replacing one. For broader context, see Calendly.

The short version

• The average company runs ten collaboration tools, and here the driver is not purchasing but whoever started the conversation
• Four uses: immediate and short, considered and recorded, working together at once, and ambient awareness — the last is what small firms need least
• One channel per relationship, stated in two sentences at the start, prevents the whole problem
• Where a customer insists on their platform, forward anything that decided something into email
• Before adding another, check whether it replaces something, whether history is retrievable, what exports, and who can see what
• A call costs everybody's time simultaneously: four people for thirty minutes is two hours, and the comparison is against that