Financial forecasts from companies are usually unreliable and can be as inaccurate as 13 per cent on an average, says a new study from global audit, tax and advisory firm KPMG LLP.

The KPMG study, Forecasting with Confidence, conducted by the Economist Intelligence Unit, was based on surveys of 544 senior executives involved in the forecasting process, of which 35 per cent of those surveyed were from Europe, 30 per cent from the Americas and29 per cent from the Asia Pacific region.

Over 30 per cent of the respondents were CFOs and 59 per cent were from a cross-section of industries and represented organisations with over $1 billion in annual revenues.

The study found that barely one per cent of companies could produce accurate forecasts, while 78 per cent of the companies reported forecasting errors of over 5 per cent.

The underlying theme of the study is that unreliable forecasts cost the company money and revealed that companies with unreliable and inaccurate forecasting suffered an average decline of 6 per cent in their share price over the past three years.

Conversely, the share prices of companies that reined in their forecast fluctuations below 5 per cent rose 46 per cent rise over the same three-year period, compared to a 34 per cent increase among the companies that had more than a 5 per cent margin of error in their forecasts. 

As the preamble to the study noted, ''All organisations use forecasts to predict and manage their future performance. But although organisations invest significant time and effort in this important task, only one in five currently produce a forecast that is reliable.''

Stephen Lis, head of the business performance services practice at KPMG LLP, noted that companies that were serious about the accuracy of their forecasts seriously had measurable business value over the long-term.

Lis added that such companies applied rigour to their forecasting used it as a core management tool, instilling ''forecasting discipline into the culture and day-to-day activity of the organisation."

Organisations with the most accurate forecasting, the study pointed out, held managers accountable for their forecasts and also provided incentives based on the accuracy of their forecasts.

Those who participated in the survey for the Forecasting with Confidence study pointed to three major areas where improvements could be made to improve the forecasting process: 

• Automation through IT systems and tools (42 per cent) 
• Scenario planning (42 per cent), and 
• Rolling forecasts (40 per cent)
• In addition, a third of respondents consider their current technology an impediment to accurate forecasting, with 40 per cent of executives saying they rely solely on spreadsheets to produce their forecasts.

Lis noted, increasing forecasting accuracy should not only improve investor confidence, but can be applied for developing strategy and performance management. ''Executives who can imbed a forecasting discipline into the culture of their organization will see that this can be a key management tool."

In the survey, the most accurate forecasters – those that, over the last three years, had actual results within five per cent of forecast – make up 22 per cent of the total. They differ from the rest in some important ways.

Highlights of the study:
1 They take forecasting more seriously
The most accurate firms: 

• hold managers accountable for agreed forecasts: 87 per cent to 76 per cent 
• incentivise managers for forecast accuracy: 25 per cent to 12 per cent 
• use the forecast for ongoing performance management: 53 per cent to 46 per cent 
• use their forecasts to help with formal earnings guidance: 24 per cent to 16 per cent

2 They look to enhance quality beyond the basics
The most accurate firms:

• are more interested in further scenario planning and sensitivity analysis: 51 per cent to 41 per cent 
• have less need to train further finance staff in forecasting: 11 per cent see it as a priority as compared with 21 per cent

3 They leverage information more effectively
The most accurate firms:

• use external market reports and competitive data more often: 68 per cent to 55 per cent
• have forecasts done more often by operational managers who are closer to where business takes place . 40 per cent to 34 per cent
• give their internal input data a higher rating for reliability, timeliness, relevance and insightfulness

4 They work harder at it
The most accurate firms:

• update forecasts more frequently .58 per cent do so monthly or more often
  compared with 44 per cent of others
• are more likely to review the figures formally: 74 per cent to 64 per cent
• forecast key balance sheet indicators more:  83 per cent to 78 per cent make greater use in the forecasting process of software
  o more advanced than spreadsheets, such as ERP systems: 48 per cent to 37 per cent .
  o off-the-shelf forecasting / planning tools:  47 per cent to 34 per cent .
  o bespoke tools: 34 per cent to 20 per cent.

5 They benefit their shareholders
The most accurate firms:

• attribute lower declines in share price directly to forecasting over the last three years . four per cent to seven per cent;
• saw shares rise faster over the same time period . 46 per cent to 34 per cent.