Company forecasts fall short

Jim Pettigrew, until recently the group finance director of emerging markets
investment firm Ashmore Group, once said he likes to be boring. For anyone that
knows the man, or indeed the firm, this may come as a surprise ­ after all,
investing billions of pounds of client money into unpredictable and potentially
risky markets is anything but dull.

But Pettigrew’s point was different. “I’m looking at making sure we get into
the routine of making it very clear what we’re trying to do and [to] keep, in a
very boring way, delivering on what we said,” he explained when we interviewed
him earlier this year. After all, there’s little the City hates more than a
surprise. Bad news it can handle, unexpected bad news is a different matter
entirely.

The point is illustrated well by a recent research project carried out by the
Economist Intelligence Unit on behalf of KPMG, Forecasting with confidence:
Insights from leading finance functions. The research provides a detailed look
at the forecasting processes of large companies. That process, along with
planning and budgeting, is seen by finance directors as the key area in which
they are most dissatisfied with their current capabilities. It’s also their top
priority for improvement over the next three years.

Looks like trouble ahead

And it’s easy to see why, if the stark picture painted by the research is true.
Only 1% of the 540 companies in the study hit their forecasts exactly, while
just 22% came within 5% either side of it. The executives interviewed, of which
more than 30% were FDs, estimated that such forecasting errors have knocked 6%
off share prices over the same period. And there is further evidence of the
direct impact of poor forecasting. Those companies with forecasts that came
within 5%, saw share prices increase by 46% over a three-year period compared to
34% for others.

In fact, Jim Sutcliffe, chief executive of Old Mutual and one of the research
respondents, believes that if an actual result deviates from predictions, the
share price can “suffer for three or four quarters”. There are a couple of
reasons why this may be true. As mentioned, there’s nothing the City hates more
than a nasty surprise. But a wild contrast between what senior management thinks
will happen and what eventually does happen will also lead to many investors
questioning the competence of the management. Clifford Hurst, FD of listed
company Zotefoams, agrees saying that getting forecasts wildly wrong can be
“career limiting”.
In truth, there is a certain amount of PR, as well as a liberal helping of
psychology, involved in the forecasting process. Sutcliffe, for4 example, says
that analysts “want you to beat forecasts by small amounts that don’t
demonstrate a lack of understanding”.

It is safe to assume that most will adopt a distinctly Scottish approach ­
with a tendency to overstate costs and understate revenues. No surprise there
perhaps. But, as Sutcliffe points out, there are plenty of examples of business
units trying to impress head office by submitting what he refers to as “macho
forecasts”.

Budgeting bravado

Chris Jackson, head of the finance faculty at the ICAEW, is under no illusions
about the nature of the problem. “There’s an awful lot of gamesmanship in
budgeting,” he says. “Different people will approach the psychology of budgeting
in different ways.” He explains how the FD or financial controller are often
called upon to act as negotiator during the forecasting and budgeting process.

But he says that many companies structure their finance teams to make such
procedures ­ and perhaps, the negotiations ­ more transparent and straight
forward. Ensuring that finance has constructive dialogue with individual
business units is very important, and this can be helped by making the reporting
lines as simple as possible.

Zotefoams’ Hurst says that his company uses quarterly forecasts and has built
in a large degree of safety to try and reduce errors. His 12 sales managers will
complete forecasts, which will then be checked and adjusted by the sales
director where necessary. Hurst and the managing director then have final
sign-off and will alter forecasts where they feel necessary.

However, psychology isn’t the only problem that contributes to a poor
forecasting track-record. Bad data is just as much, and often more, of a
contributing factor than over-zealous sales managers. A staggering 47% of
respondents consider the reliability of the financial data they use for
forecasting to be merely adequate or worse. While this is undoubtedly a
technology issue, it also goes beyond IT.

For example, just 40% incorporate externally generated data, such as
government or other economic reports into their forecasts, while just over
one-fifth look at data on non-economic risks that could have an impact on their
market. It is perhaps no surprise that the areas where most forecasting errors
occur are those where external data could prove extremely valuable ­ consumer
demand and economic drivers.

Despite technology not being the only contributing factor to poor quality
data, it’s the area where most companies think improvements would help the
forecasting process the most. Of the respondents, 42% said automation would
increase confidence in their forecasts, the most common response, and 35%
considered the current technology a notable impediment. In fact, two-fifths of
respondents rely solely on spreadsheets to produce forecasts.

Predicting cash flow~

Paul White, director of Microsoft Dynamics UK, agrees that poor forecasting
is a problem for large companies but points to the medium-sized business as
having a similar, if not greater, problem.

“While these organisations might not have to worry so much about share price
movements, cash flow will probably be a much greater concern,” he explains.
“Discussions with any bank are made more challenging if you have no credibility
in terms of forecasting/budgeting.”

It’s an opinion shared by Hurst, although he points out that cash forecasting
is an even more difficult art to get right. “I worked at ICI and cash
forecasting there was a bit of a nightmare, so it’s not just smaller companies,”
he explains.

While Microsoft’s White is bound to say that technology can help, he is under
no illusions that, historically, the budgeting and forecasting technology market
has had a relatively poor track record. “The market for budgeting and
forecasting solutions has been dominated by vendors offering relatively
discrete, specialist solutions that were often expensive and represented yet
another application that people had to learn,” he says. “This limited the
community of people within a business who were engaged by the solution, which
inevitably compromises the success of the project and the quality of the
forecast.”

While KPMG itself is under no illusions of the importance of technology to
the forecasting process, it warns that getting the processes and data right is a
critical first step. “The priority in leveraging the technology investment is to
establish an overall framework for how forecasting fits within the management of
the business,” the accounting firm writes in the report. “Organisations need to
understand strategically how forecasting can benefit them. They need to
determine what they want from it and then ensure the supporting processes are
designed and built in the context of those expectations.”

The problem is, it’s quite a complex process. As Microsoft’s White says:
“Revenue forecasting needs to work up from a set of leading indicators ­ the
economy, industry growth, competitor performance, customer satisfaction,
marketing effectiveness, sales effectiveness and pipeline developments. Cost
forecasting needs to be worked up from the sales forecast, POs issued,
requisitions raised, historical trends and economic data. How many FDs have a
combined 30-, 60- and 90-day forecast in front of them, for both revenue and cos
t? Not many.”

Box: How the best differ from the rest

Organisations with the most accurate forecasts differ from the average
company in five important ways:

• They take forecasting more seriously; holding managers accountable for
  agreed forecasts and incentivising.
• They look to enhance quality beyond the basics; being more interested in
  further scenario planning and sensitivity analysis.
• They leverage information more effectively; using external sources of data
  and involving operational managers in the process.
• They work harder at it; updating forecasts more frequently, reviewing
  figures formally and forecasting key balance sheet indicators.
• They benefit their shareholders; seeing share prices rise faster over the
  same time period.

source: KPMG