Three years ago we began building a customer portal for our contract warehousing clients. Like most of our development work, the time my team and our contractors spent on it was capitalised. Finance treats software we build as an asset, spreads its cost over the years we expect to use it, and does not charge it against profit while it is being built.
By its second year the portal was clearly in trouble. Clients had moved to integrating directly with our systems rather than logging into anything, and usage in the pilot was tiny. I raised it at a steering meeting and suggested we stop. The finance director asked one question: what happens to the balance. About one point four million pounds of capitalised cost sat on the balance sheet. If the project were abandoned, that asset would have to be written off, and the whole amount would hit profit in a single year, the year our chief executive was explaining results to a new owner.
So we did not stop. We reshaped the scope, renamed a phase and carried on for another eleven months, spending a further six hundred thousand pounds, partly so that the asset could be described as still under development. In the end the portal was written off anyway, at two million.
Nobody in that room was being dishonest. The accounting treatment was correct. What none of us had seen was that capitalisation changes the cost of saying no. Every month a project continues, stopping it becomes more expensive in the accounts, even when continuing is plainly more expensive in cash.
We changed how we decide, not how we account. Each project now has stop criteria agreed with finance before work starts, such as the usage a pilot must reach and by when. Capitalisation happens in stages, and costs are only capitalised once the project has passed the gate that shows it will be used. Our finance director now reports the capitalised balance of every live project to the board each quarter beside its progress, so a write off is something the board expects, not something it discovers. Two projects have been stopped under the new rules, both early and both for less than two hundred thousand pounds.
A project cheap to approve can become expensive to cancel. We had not noticed that our own books were making it so.
– Serguey Shinder
Top comments (0)