The problem with taking orders by text message

Text message ordering arrives in a distribution business the way most things do: one customer starts doing it, it is easier than a phone call, and within two years a meaningful share of your revenue comes in through the personal phone of whoever answered first.

It genuinely is better than a phone call. It is asynchronous, it leaves a written record, and it costs nothing. It is also the hardest channel to ever get out of, which is worth understanding before it becomes half your order book.

What it fixes

• The order arrives in writing, so the misheard quantity mostly goes away.
• Nobody has to be free at the same time.
• The customer can send it while standing in the walk-in, which is exactly when they know what they need.

What it does not

Everything else. A text has no pricing, so the customer still does not know what the order costs. It has no stock information. It has no validation, so 3 of the chicken is a message that requires a reply. Somebody still transcribes it into a system, which is the expensive part, and now they are transcribing free-form prose rather than a form.

The new problem: the order lives on a personal phone. When that person is on leave, or leaves, the channel and the relationship go with them. Most distributors discover this exactly once.

The data problem underneath

A year of text orders is a year of order data you cannot query. You cannot see which accounts stopped buying a line, which is the earliest churn signal available to any distributor. You cannot see range penetration. The information exists, in a thread, in a format nothing can read.

This is the same problem the paper form has, which is the tell that text ordering is not the step forward it feels like. It is the paper form with better delivery.

What to do about it

Not ban it. The venues that text you do so because it fits how they work, and taking the channel away without replacing what it does for them will cost you orders.

The realistic path is to make the alternative genuinely faster for the routine case, which means their own list rather than your catalogue, and then let text be what it should be: the exception channel for the venue that has run out of something at 6pm. Some texts will always arrive. The goal is that the weekly restock is not one of them.