Double data entry

Somebody in your business types the same information into two places every day. They have done it so long that nobody experiences it as a problem. That is exactly why it is expensive.

Why it hides

Loud problems get fixed. Double entry is quiet: it never fails, never throws an error, never causes an argument. It just consumes ten minutes at a time, several times a day, forever. Ask your team what is broken and nobody will mention it. Ask them to narrate one job end to end and you will watch it happen three times.

How to find every instance

Follow one unit of work — one booking, one quote, one order — from first contact to paid, and write down every place a human puts information into a system. Then mark every field that appeared more than once.

The common chains, in roughly the order we find them:

• Enquiry arrives by email or phone → typed into a spreadsheet → typed into a CRM
• Quote built in a document → re-keyed into the invoice tool when accepted
• Booking taken by phone → written in a diary → copied into a calendar → copied into a job sheet
• Supplier invoice → typed into accounting → typed again into a job-costing sheet
• Payment taken on a card terminal → reconciled by hand against the accounting system
• New customer details → entered into billing, into the mailing list, and into the delivery system separately

What it costs

Use loaded cost, not salary: hourly rate plus roughly 30%. Then touches per week × minutes per touch ÷ 60 × loaded rate × number of people × 48 weeks. The calculator does this if you would rather not.

Two admin staff at $28/hour, re-keying for 25 minutes a day each, is about $8,700 a year. That is one chain. Most operations have four or five.

The error cost nobody counts

Every re-key is a chance to transpose a number. The direct labour cost is the visible half. The invisible half is the wrong address, the invoice for the wrong amount, the booking on the wrong day — and the hours spent finding and fixing each one, usually by someone senior and usually in front of a customer.

What fixing it actually looks like

Ranked by what we most often recommend:

1. Delete the second system. The cheapest automation is the one you do not build. Often the spreadsheet exists because of one report nobody reads any more.
2. Use the native integration. Most tools already talk to each other and nobody turned it on. This is free and takes an afternoon.
3. One automation tool. Zapier, Make or n8n moving records between two systems on a trigger. Which one depends on volume and shape.
4. Custom build. Last resort. Correct when the logic is genuinely yours and no off-the-shelf tool models it.

Most double entry is solved at steps 1 and 2, for nothing. Anyone who leads with step 4 is selling you step 4.