Your Budget Is Already Wrong — The Case for Rolling Forecasts in a Tariff Environment

The Budget Was Outdated Before It Was Approved

A manufacturer spends three months building its 2025 budget. Finance consolidates 200+ Excel models from 40 business units. Leadership cycles through four rounds of review. The CFO presents to the board in December. By February, tariff announcements have changed the input cost assumptions by 15%. The budget is wrong, and everyone knows it, and yet the organisation is still managing to it.

This is not a hypothetical. It's a description of what happened to a significant number of manufacturers in Q1 2025. The annual budget — already a compromised document by the time it was approved — became actively misleading within 60 days of sign-off.

What 200 Excel Models Actually Cost

At the automotive manufacturer we worked with — a $40 billion operation — the annual planning cycle consumed 8 weeks of intensive work across 40+ business units. The process involved 200+ Excel models, each maintained by a different person with a slightly different version of the template, linked to each other with fragile cross-workbook references that broke regularly.

The cost of this isn't just time. It's the types of decisions that can't be made while planning is consuming organisational bandwidth. The finance team is so absorbed in consolidation mechanics that it can't do scenario analysis. The business can't ask 'what if material costs increase 10%?' because there's no model architecture that can answer the question quickly. Every variance analysis is retrospective. Nobody is looking forward.

After moving to a connected planning platform, that 8-week cycle compressed to 2 weeks. Not because less work was being done — because the work was happening in a system where data flowed automatically, assumptions were centrally managed, and scenario modelling was built in rather than bolted on.

The Case for Rolling Forecasts in a Tariff Environment

A rolling forecast is not just an annual budget refreshed monthly. It's a fundamentally different planning philosophy. Instead of locking assumptions for a fiscal year, you maintain a 12-18 month forward view that updates with actuals each period. The inputs — headcount, material costs, volume assumptions, pricing — are driver-based, so changing one assumption cascades through the model automatically.

In a tariff environment, this matters for a specific reason: the cost of being wrong is not symmetrical. If your budget assumes 8% steel costs and the actual is 12%, the impact on your operating margin is immediate and cumulative. The longer you manage to the wrong assumption, the larger the gap grows. A rolling forecast that incorporates updated cost curves monthly limits the gap to a single period.

The political argument for annual budgets is usually about accountability — 'if we're not managing to a fixed number, how do we hold people accountable?' The answer is that you hold people accountable to the latest forecast, which reflects current reality. Holding a plant manager accountable to a January assumption in October, when the market has moved materially, is not accountability. It's theatre.

What the Transition Requires

The shift from annual budgeting to rolling forecasts requires three things: a connected planning tool (Anaplan, Planful, Adaptive, or similar), driver-based model architecture rather than line-by-line manual inputs, and a change management process that resets what finance means to the business.

The last one is hardest. Finance teams that have spent years building budget packs and variance commentary need to pivot to building forecast models and scenario analysis. The skill set is adjacent but not identical. The mindset shift — from recording history to shaping the future — takes longer than the technical implementation.