The budget is not the forecast
A budget is a commitment, a forecast is a best estimate and a target is an ambition. Most planning processes blur the three into one number, and then wonder why nobody believes it.
Insights
Writing from the Corporate Business Alliance on the practice of finance, accounting, management and technology.
A budget is a commitment, a forecast is a best estimate and a target is an ambition. Most planning processes blur the three into one number, and then wonder why nobody believes it.
Flexing every input by ten per cent produces a table, not an insight. Useful sensitivity analysis starts from the decision, chooses ranges from evidence and finds the assumptions that could change the answer.
A business can forecast a profit and run out of cash in the same quarter. The gap is usually working capital, and it is usually forecast as an afterthought.
Closing the books in fewer days rarely comes from working faster. It comes from moving work out of the close, reconciling continuously and deciding in advance what is material enough to wait for.
A reconciliation prepared every month and never challenged provides comfort but little assurance. What makes it a control is an independent review that asks the right questions of the reconciling items.
Transactions that flow through a system are checked by the system. Manual journals are not, which is why they carry a disproportionate share of both errors and fraud risk — and why they deserve a review designed for them.
Negotiation gets the attention, but the requirement, the specification and the evaluation decide most of what a contract will cost. By the time suppliers are bidding, much of the value has already been fixed.
Every process has one step that limits what the whole can produce. Improving anything else feels productive and changes nothing, which is why the first job in operations is finding the constraint.
Most organisations keep a risk register. Many are updated once a year, describe risks too vaguely to act on and rate them in ways nobody can reproduce. A few changes make a register something that drives decisions.
Governance starts with an inventory: which AI systems an organisation uses, who owns each one and what decision each informs. The principles in the policy come after.
Generative AI can draft in seconds what took an hour. The skill that matters is no longer producing the first draft but specifying the task well and checking the result before anyone relies on it.
Analysts spend much of their time cleaning data that was entered wrongly at source, and they clean the same errors every month. Lasting data quality comes from fixing capture, ownership and definitions where data is created.
Announcements about the Corporate Business Alliance are in News; longer guidance on applying the standards is in the publications library.