What Xero's reports tell you about your customers — and what they don't

Xero is very good at telling you what happened. It is not built to tell you what is about to stop happening. Knowing where that line falls saves a lot of hunting through menus.

What Xero does well

Out of the box you can see your income broken down by customer, what each of them has bought, who owes you money and how overdue it is, and how any period compares with another. For invoicing, chasing payment and closing the books, that is most of what a business needs.

All of it answers questions about money that has already moved. Which customers paid, what they bought, what is outstanding.

The question it cannot answer

"Which of my customers are drifting away?"

That question has a shape accounting software is not designed for. To answer it you need to know what is normal for each customer individually, and then notice when the pattern breaks. A customer who orders weekly and one who orders quarterly cannot be judged by the same rule, and a report that sorts by last invoice date treats them as if they can.

Three specific things fall through the gap:

A regular customer who is overdue by their own standard

Sorting customers by last order date puts your least frequent buyers at the top. The fortnightly buyer who has been silent for six weeks — the one who matters — is buried in the middle of the list, looking unremarkable.

A customer who is still ordering, but spending less

They appear in every report as an active customer, because they are. Their order frequency has not changed. They are simply buying half what they did, which usually means somebody else is supplying the other half. No "last invoice" report will ever show this.

A customer who quietly dropped one product

They still order every fortnight and their total spend looks fine, but a line they used to take every single time has vanished. That is a competitor taking a piece of the account, and it is invisible unless you compare line items order by order.

Why the gap exists

It is not an oversight. Accounting software has to be exactly right about what happened — every cent reconciled, every invoice accounted for. Judging whether a customer is drifting is a different kind of question: it requires a view of what is usual for that customer, and a willingness to be approximately right about the future rather than precisely right about the past.

Those are different jobs, and Xero sensibly does the first one.

Covering the gap yourself

You can do it by hand with an invoice export and a few hours in a spreadsheet — we have written up the method step by step. It works, and every business should do it at least once, because the first run is usually a surprise.

The difficulty is repetition. The answer is only useful while it is current, and it goes stale within a fortnight. Most people do it once, find four customers they had lost without noticing, resolve to do it monthly, and never do it again.

What good looks like

Whatever you use, the test is the same. It should:

• judge each customer against their own history, not a single rule for everyone
• rank what it finds by what the customer is worth, not by how late they are
• show its reasoning, so you can tell when it is wrong
• stay quiet when the whole business is quiet — a holiday is not a customer leaving
• give you a handful of names, not a report to work through

A list of thirty customers is a list nobody rings. Three names with a reason beside each is a Monday morning that changes something.

Read next: how to find customers who have stopped ordering, step by step