Mastering Forecasting: Principles and Practices - IBM

Mastering forecasting practice and principles

© 2013 IBM Corporation

Steve Player

Steve Player serves as the program director for the Beyond Budgeting Round Table (BBRT) in North America and works with BBRT member companies to implement continuous planning processes. He has more than 30 years’ experience with improving performance management and implementing strategic planning processes. He is also the managing director of Beyond EPS Advisors, a business consulting firm, and founder of the ABM Advanced Implementation Group (ABM SMART). He is the coauthor of Future Ready: How to Master Business Forecasting and Beyond Performance Management as well as five other books. He writes the blog “The Finance Transformation” for Business Finance Magazine, which features CFO interviews from leading organizations on innovative finance and planning processes.

2

© 2013 IBM Corporation

Pierre Guillaume

Pierre Guillaume has more than 25 years of experience as a management consultant and specializes in cost management, advanced planning and budgeting, and business strategy development. He has been responsible for cost, performance management, planning and budget systems, and strategic information systems implementations for clients in a variety of industries in the North America, Europe and Asia. Pierre is the president of Beyond EPS Advisors, a Dallas-based consulting firm associated with the BBRT, which he coformed after leaving Capital One, where he ran the cost budgeting, forecasting and planning function. Prior to Capital One, Pierre was the northeast region strategic cost and performance management practice leader for Arthur Andersen. Pierre’s work in the field of performance management has been documented and recognized by a number of Harvard Business School case studies. Pierre is a frequent speaker on advanced costing, planning and business strategy. He has also taught accounting and financial management at the Université de Paris and business strategy at the Centre de Perfectionnement aux Affaires in Paris. 3

© 2013 IBM Corporation

Steve Morlidge

Unilever (1978–2006) roles include: •Controller, Unilever Foods UK ($1 billion turnover) •Leader, dynamic performance management change project (part of Unilever’s Finance Academy), 2002–2006

Outside Unilever •Chairman of the BBRT, 2001–2006 •BBRT Associate, 2007 to present •Founder/director, Satori Partners Ltd., 2006 •Ph.D., Hull University (Management Cybernetics), 2005 •Visiting Fellow, Cranfield University, 2007 •Coauthored book Future Ready: How to Master Business Forecasting, 2009 •Editorial Board, Foresight magazine, 2010 •Founder, CatchBull (forecasting performance management software), 2011

4

© 2013 IBM Corporation

Why does it seem like everyone is discussing rolling forecasts?

“Why you should ditch your annual budget”

5

© 2013 IBM Corporation

Question Is forecasting important for managing your business? Are you satisfied with the quality of your forecasts?

6

© 2013 IBM Corporation

Why forecast?

Should we have a barbecue?

Lagging indicator

7

Leading indicator

Forecast

© 2013 IBM Corporation

“ 8

It’s awful. Why did no one see it coming?

” © 2013 IBM Corporation

9

© 2013 IBM Corporation

Other misconceptions

“

It’s just common sense

“ 10

It’s highly specialized

”

” © 2013 IBM Corporation

Is it Safe to Cross the Road? FACTS Speed of cars Change in speed PLUS ASSUMPTIONS Stopping distances Intentions of driver Eyesight of driver Margin of error … all continuously validated creates a STORY which INFORMS DECISIONS … to cross/speed up/slow down 11 Mastering Forecasting — 11

© 2013 IBM Corporation

Changing your business to manage in a volatile world

or

Clean sheet?

12

Cumulative cost structure profile?

© 2013 IBM Corporation

Six key design principles

Mastering purpose

Mastering time

Mastering models

Mastering measurement

Mastering risk

Mastering process

13

© 2013 IBM Corporation

Three time horizons…three types of forecast process

Strategic planning

Business forecasting

Process

Adaptation

Adaptation

How do we structure the business to compete most effectively?

Freedom of action Alternative scenarios of the future environment Broad brush estimates

Future

CREATE THE FUTURE Navigation Navigation Creating options

How do we deploy our resources to their best effect? Decision making

Choice of response limited Best estimate of what will happen (based on current assumptions) Detailed enough (with ranges)

SHAPE THE FUTURE Operational forecasting

Response

Response

How do we service demand efficiently?

Highly constrained Prediction of what will happen Detailed forecasts

INCREASING Unpredictability + choice

Purpose

Execution

PREDICT THE FUTURE

Now

14

© 2013 IBM Corporation

Forecasting is about decision making because …

TARGET ET G D U B

3. ... you first need to work out where you are heading … 4. ... and then what you have to do differently to get back on track.

FORECAST

QUARTER

QUARTER 1. ... while it is necessary to have a plan….

15

2. ... the world often turns out different from the way you expected so …

© 2013 IBM Corporation

Key concepts

A target

A forecast

A set of plans

is what we would like to happen

is what we think will happen

is what we intend to do

| which we achieve by producing |

| based on |

| which we change to achieve | our target

Specification for a forecast T = timely | A = actionable | R = reliable | A = aligned | C = cost-effective

TARAC

16

CARAT

© 2013 IBM Corporation

T imely

Timely

Our need to be able to forecast— to see into the future—is driven by how quickly we can react to what lies ahead.

1m

2m iles

ile ah ea d-

3m ah ea d-

NO TH

CH

ING

iles ah ea d-

AN GE D

ST O

IRE CT ION

100 m ahe eters ad CH DIR ANG E EC TIO N

P

Forecast horizon Q1

Q2

Q3

Q4

Q5

Q6

Adjust existing plans

(Minimum) decisionmaking lead time 17

New actions Major new initiatives © 2013 IBM Corporation

T imely

Timely

VARIABILITY The frequency at which we should forecast is driven by the rate of change in key variables

18

HIGH LOW

CRITICALITY

LOW

HIGH

LESS FREQUENT ROUTINE UPDATE OR BY EXCEPTION

FREQUENT ROUTINE UPDATE

INFREQUENT UPDATE ONLY IF NECESSARY

LESS FREQUENT ROUTINE UPDATE

© 2013 IBM Corporation

T imely

Case study: an airline

VARIABILITY

Variability

Speed of response

Update frequency

Forecast horizon

Revenues

High

High

High

Daily

Quarter

Labor costs

High

Low

Medium

Twice monthly

Six months

Fuel costs

High

High

Medium

Weekly

Quarter

Maintenance spending

Medium

Medium

Medium

Twice monthly

Six months

Advertising spending

Medium

Medium

High

Monthly

Six months

Aircraft ownership costs

Medium

Low

Low

Quarterly

Year

Airport rates and charges

Medium

Medium

Low

Monthly

Six months

Other operating

Medium

Medium

Medium

Twice monthly

Quarter

19

HIGH LOW

Economic relevance

CRITICALITY

LOW

HIGH

LESS FREQUENT ROUTINE UPDATE OR BY EXCEPTION

FREQUENT ROUTINE UPDATE

INFREQUENT UPDATE ONLY IF NECESSARY

LESS FREQUENT ROUTINE UPDATE

© 2013 IBM Corporation

A

Actionable

Every forecast is based on a model—and different approaches have different strengths and weaknesses

ctionable

Types of forecasting model

Statistical

History

Forecast

Input A: Volume

Mathematical

Input B: Price

Cost model

Input A: Usage

A*((B*C*(1+D))

Output: cost forecast

Input B: Wastage

Judgmental

20

© 2013 IBM Corporation

A

Actionable

ctionable

Intervention forecast, based on judgement

Every forecast is based on a model—and different approaches have different strengths and weaknesses

Momentum forecast, based on extrapolation

21

© 2013 IBM Corporation

Question Does your business have a forecast policy that defines what “good” looks like? Do you routinely measure the quality of your forecasts? If not, how do you know whether you can rely on your forecasts?

22

© 2013 IBM Corporation

Forecasting lies at the heart of the planning system….

…and there are multiple eliable potential points of failure

Infrequent, inconsistent, process Managing expectations Multiple adjustments F F

Activity planning bias High volume Noisy demand patterns

Market

Historic Demand

100% judgement Natural optimism Activity

23

Supply Planning

F F F F

Future Demand

Future Results

F F

F F

Operations Future Financial F F Demand Planning Planning

Stakeholders

Lack of trust in the process

Profit Planning

Forecast Uplift Activity F Planning

F F

Resource Allocation

‘Gap’ driven bias Disconnected from operational forecasting

Sales and Operations Planning (S&OP)

Performance Management

© 2013 IBM Corporation

R

Reliable

eliable

Forecasts are generally reliable if they are unbiased and have an acceptable level of variation bias (hook)

24

variation

bias (slice)

© 2013 IBM Corporation

R

How can we tell? The KPI trap

Average net error (bias) Average gross error (variation)

TARGET

ACTUAL

x2 average error—evidence of excessive variation 26

© 2013 IBM Corporation

A

Aligned

ligned

The should be only one set of forecast numbers, but the unpredictable nature of the world needs to be taken into account

Scenario A

Scenario C Scenario B

27

© 2013 IBM Corporation

28

© 2013 IBM Corporation

Cost-effective

C

ost Effective

An effective forecast process is also generally an efficient forecast process 1.Do things in the right order, at the right speed. 2.Follow the process often and in the same way 3.Observe the results, separating bias (hook or slice) from variation. 4.Make adjustments, improve the process

32

© 2013 IBM Corporation

Timely

Actionable

Reliable

Aligned

Cost-effective

In good time to take action

Good forecasts have information about things we can change

Unbiased

Pointing in the same direction

Right the first time, with no rework or waste

33

(+ errors = - errors)

with an acceptable (known) level of variation

© 2013 IBM Corporation

What next?

First steps 1. Engage all levels 2. Measure forecast quality Next steps 1. Design your new forecast process 2. Implement, automate, integrate 3. Educate 4. Measure 5. Learn and improve 34

© 2013 IBM Corporation

Hurricane Ernesto projected path

35

© 2013 IBM Corporation

Hurricane Ernesto actual path

36

© 2013 IBM Corporation

Future Ready: How to Master Business Forecasting

A practical guide to help executives navigate through turbulent times

Steve Player 214-239-0155 [email protected] www.bbrtna.org 37 Mastering Forecasting — 37

© 2013 IBM Corporation

Future Ready: How to Master Business Forecasting

A practical guide to help executives navigate through turbulent times

Pierre Guillaume 617-413-8146 [email protected] www.beyondeps.com 38 Mastering Forecasting — 38

© 2013 IBM Corporation

Future Ready: How to Master Business Forecasting

A practical guide to help executives navigate through turbulent times

Steve Morlidge + 44 1737 822153 [email protected] http://www.satoripartners.co.uk 39 Mastering Forecasting — 39

© 2013 IBM Corporation

40

© 2013 IBM Corporation

Trademarks and notes

IBM Corporation 2013 • IBM, the IBM logo, ibm.com, and the planet icon are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with the appropriate symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml. • Other company, product, and service names may be trademarks or service marks of others. • References in this publication to IBM products or services do not imply that IBM intends to make them available in all countries in which IBM operates.

41

© 2013 IBM Corporation