Why you've got financial forecasting back to front

Basing plans on last year's revenue is a recipe for disaster.

by Geoff Tuff and Steven Goldbach
Last Updated: 30 May 2018

Geoff Tuff and Steven Goldbach are principals at Deloitte Consulting LLP, serving as senior leader at the Innovation and Applied Design practice and chief strategy officer respectively.

Their book DETONATE: Why—And How—Corporations Must Blow Up Best Practices (And Bring a Beginner’s Mind) to Survive (Wiley; May 2018), discusses how many of the most closely held business orthodoxies are actually destroying value.

In this adapted extract, they look at financial forecasting. It all began in 1995, when they were evaluating the potential of up-and-coming Canadian doughnut chain Tim Hortons, which their client was interested in acquiring. The chain had publicly declared that its goal was to have 2,000 stores by the year 2000, and it was faith in Tim Hortons’ ability to meet this goal that underpinned the price any buyer would pay for it.

So like any good consultants, Tuff and Goldbach decided to test it with a financial forecast...


Was 2,000 by 2000 anywhere close to an achievable goal? Canada had already the most doughnut stores per capita of any country in the world. We assumed that if markets added stores, and Tim Hortons kept or grew its share modestly, then we could pressure test its growth assumptions.

Naturally, our conclusion was that it was highly unlikely that Tim Hortons could make its store target and hence its revenue target in such a saturated market. We completed our financial forecast by assuming costs would continue to behave consistently with the proportion of revenue, as they had in the past, and delivered a very defensible conclusion.

It was a moot point. Before our project was complete, Tim Hortons was sold to Wendy’s. But even though it didn’t matter, we got it wrong. Tim Hortons did achieve its 2,000 stores by 2000 and now has roughly 3,500 locations in Canada and more than 4,000 globally. So where did we go wrong?

For starters, Tim Hortons challenged the notion that stores had to be stand-alone, freestanding locations. They started to install kiosks in gas stations, which became a strong growth driver as consumers loved filling up their cars and themselves for a drive. They disproportionately took share from their competition by focusing on great customer service, branding that tugged at the heartstrings of Canadians, and an expanded menu.

We had assumed they would continue to have a space in a world that looked like it did the day before, yet Tim Hortons caused the world to be different. Why did we get it wrong? Because it would have been really hard for us to defend a forecast that had no basis in previously demonstrable facts.

The problem with revenue-first forecasting

Our forecasting process was fairly typical. When organisations ask what they think revenue will be next year, it’s usually some function of looking at what they’ve done in the past, comparing it to ‘expert forecasts’, thinking about how they’ll increase their share (it’s always increase, isn’t it?), and then landing on a revenue goal.

Once they have that goal set in stone (because once it’s spoken in a meeting, it’s not getting changed), they ask what profit they want to make. That’s normally a function of what they’ve promised the capital markets or other important groups, or what they need to make obligations such as debt servicing. Only then do they turn to costs: ‘knowing’ their revenue, what can they afford to spend to make their profit targets?

Then the budgeting process starts in earnest. The ‘gap’ is established between costs that organisations think are necessary against the targets for revenue and profitability. Rinse, repeat, year after year.

Just to check that these weren’t our own observations, we asked a large sample of established businesses. Nearly 78% of respondents reported that they start forecasting with a target financial outcome, and 68% forecast costs based on a percent of revenue target.

The core assumption is that the future will be the same as the past. We largely presume that spending will continue to follow a predictable pattern of the percentage of revenue that it occupied in prior years. This reflects no reality that we’ve ever inhabited. In fact, reality literally works in the opposite way. Costs create revenue, not the other way around.

Forecasting behaviour

By not asking ‘why’ revenue ought to behave consistent with the past, you’re missing the underlying cause of your revenue – your customers’ behaviour. Each year, you have to give customers a reason to behave favourably toward you, whether that’s simply continuing to buy your product or service, switching from a competitor, or agreeing to pay more. By thinking about what behaviour needs to happen to cause revenue, you then naturally have to think about what you needs to spend to cause that behaviour. Sometimes this might look like it did in the past. Many times, it won’t.

Businesses also frequently underestimate the impact of competitive moves because they aren’t knowable at the time of the forecast. The implicit assumption is that competitors will act as they did in the past, which, too, is obviously flawed. In our survey, around 75% of respondents’ companies budgeted cost by starting with what they spent last year, something that zero-based budgeting – which resets budgets go to zero each year and requires every dollar of spend needs to be justified – attempts to address.

A better way of planning

Financial forecasting, as practiced today, focuses management attention on understanding why the plan was different from what transpired in the real world, rather than on what actually matters: improving performance. Once a plan is complete, the organisation spends untold time and energy understanding why the real world is different than what is on paper – at the expense of spending that time on making performance in the real world better.

Unfortunately, focusing on meeting or beating a financial plan creates incentives to make decisions that may not be best for the business in the long run. Comparing to plan, meanwhile, doesn’t actually tell us what the cause of performance might be.

Does that mean that we should just forego financial planning altogether? If the choice were between planning the way most organisations do now versus doing no planning at all, it would be about a push. The time and resources saved would be astounding, and if those resources were ploughed back into better customer insights or things that delight the customer, the trade-off might be worth it. Having said that, we think there is a better way to plan that would give organisations much greater insight on what really matters.

The core principle is to focus on behaviour. Financial results are the outcome of underlying behavioural changes. If you consistently stray from the behaviour that causes financial results, you implicitly assume that the behaviours that caused past results are simply continuing in the same ratio or patterns into the future.

This assumption is increasingly under pressure in a more discontinuous world. As a result, grounding financial plans in customer behaviour makes financial forecasts more valid – and therefore useful.

Image credit: Everett Collection/Shutterstock

Tags:

Geoff Tuff and Steven Goldbach recommends

The secret to success? There is no secret

Read more

Find this article useful?

Get more great articles like this in your inbox every lunchtime