The same data, typed in three times

An order comes in. Someone enters it in the tracking spreadsheet. Then they type it again into the invoicing software. Then they send it once more, by email, to the warehouse or the accountant.

Nobody decided it should work this way. It built up over time: one program for invoices, one for stock, a spreadsheet for whatever didn’t fit in either. Each one earns its place. It’s just that, between them, a person moves the data.

A few signs the same thing is happening in your business:

1. The same customer exists three times, spelled differently. “Alpha Ltd”, “Alpha Limited” and “ALPHA”. One company, three records, three addresses, one of them out of date. Look up the customer’s history and you find a third of it.

2. Copy and paste is part of someone’s job. There is a person, or a slice of someone’s day, that does nothing but move data from one window to another. It isn’t written down anywhere, but when that person is away for a week, everyone notices.

3. The numbers don’t agree between programs. The stock system says 40, the spreadsheet says 38. Nobody knows which is right, so someone counts the shelf. Usually the mistake is a copy made three weeks ago.

4. One change has to be made in several places. A customer changes their delivery address. It gets fixed in invoicing, but not in the delivery sheet. The goods go to the old address.

5. Nobody knows which copy is the real one. “Where’s the correct version?” gets three answers. Each department trusts its own copy, because that’s the one it checked.

6. Data travels by email and screenshots. The order reaches the warehouse as a phone photo of a monitor. It can’t be searched, it can’t be counted, and two days later it’s lost somewhere in the chat.

What it means

Each program was chosen to do one job well. Nobody chose how they connect, so the connection became a person.

The cost isn’t only time. Every copy is a chance to get something wrong, and mistakes don’t show when they’re made. They show at stocktake, on a disputed invoice, on a returned delivery. Then people look for someone to blame, when the real problem is how the data moves.

There’s one more thing. While the same data sits in three places, no report can be fully trusted. Every figure has to be checked somewhere else before anyone relies on it.

What goes in its place

You rarely need one new system that replaces everything. Most of the time the existing programs stay. What changes is the path the data takes between them.

In practice:

• each kind of information gets one place where it is written: customers in one, stock in another, orders in a third;
• the other programs read from there, automatically, instead of receiving a copy typed by hand;
• an order is entered once, and the invoice, the delivery note and the message to the warehouse are all produced from it;
• duplicate customers are merged into one record, with their whole history together;
• where two programs can’t be connected directly, they’re linked by a scheduled import, not by someone copying;
• any mismatch between them shows up as a warning on the day it happens, not at stocktake.

The person who moved the data stays with the company. They just do something else.

If you’re not sure how many copies of the same data you have, a good starting point is to follow a single order, from the first message to the invoice, and count how many times it was typed. If you’d like us to look at it with you, get in touch.