SlideShare una empresa de Scribd logo
1 de 19
Descargar para leer sin conexión
perspective JUL 2013

DEMYSTIFYING
FORECAST
ACCURACY
ACHIEVING HIGH
SALES FORECAST
ACCURACY IS
A MATTER
OF DISCIPLINE
Alberto Calvo, Alessandro Barmettler, Alberto Oteri
Demystifying forecast accuracy:
achieving high sales forecast
accuracy is a matter of discipline
Published by
Value Partners Management Consulting
via Vespri Siciliani 9
20146 Milan, Italy
July 2013
Written and edited by:
Alberto Calvo, Alessandro Barmettler,
Alberto Oteri
If you would like an electronic copy
please write to:
cristina.goddi@valuepartners.com
For more information on the issues
raised in the report please contact:
alberto.calvo@valuepartners.com
alessandro.barmettler@ valuepartners.com
alberto.oteri@valuepartners.com
If you would like to subscribe
or to be removed from our mailing list
please write to:
subscription@valuepartners.com
valuepartners.com
Copyright © Value Partners
Management Consulting
All rights reserved
3

executive summary

My wife and I walked into a Milan patisserie to buy
a small cake. We walked out with a tray full
of mouthwatering cannoli, in a quantity suitable
for a small army. For free.
How did that happen?
Something with the shop’s forecasting had gone
terribly wrong that day. Whilst thanking the chef
for this unexpected tray-full of bliss we found out
that more or less each evening he has to somehow
‘dispose’ of unsold cakes, either supplying them
to soup kitchens or handing them out for free to
unsuspecting clients.
If an apparently simple operation such as a
patisserie – employing 10 skilled workers who have
crafted the same products for decades – isn’t able
to plan its production accurately, then how can
a multinational company successfully manage its
own forecasting?
Like our patisserie, multinationals struggle
with forecasting though this is not altogether
surprising – demand planning is one the most
challenging jobs around.

perspective DEMYSTIFYING FORECAST ACCURACY
4

Annex 1
Europe, Quarterly variation of Sales, Percentage, 1Q 2000 – 1Q 2013

2

1,5

1
31%
0,5
14%
57%

0

31%

-0,5

-1

-1,5

-2
2000

2001

2002

2003

2004

2005

2006

2007

Standard deviation

Source: Eurostat.

perspective DEMYSTIFYING FORECAST ACCURACY

2008

2009

Automotive

2010

2011

2012

2013

Food, beverages, tobacco
5

have you to cope
with increasing
volatility ?
Over the last few years a number of
firms have suffered the effects of
increased market demand volatility,
especially those with a varied product
offering and a high number of SKUs.
(see annex 1)
As the opening example showed, the
process of demand forecasting is not
only crucial to a company’s success, but
is also inherently challenging. Within the
context of larger firms, this process may
become further complicated through
the interplay of many factors such as
investment decisions, supply chain management and strategic planning.

From our experiences, we believe that
the process of company forecasting
seeks to address four key objectives:
1.	

To guide and support commercial
planning – helping to align objectives regarding volume, channels,
client, price and strategy

2.	 To calibrate the supply chain and
operations – enabling an increase in
service levels through the optimisation of logistics and manufacturing
planning in the short term
3.	 To decide which products to manufacture in which plant – optimising
the investment decisions in the
medium-long term
4.	 To quantify the expected economic
and financial results – helping anticipate and forecast future financial incomes, maintain control over
strategic directives, manage cash
flows, etc.

However when addressing the
issue, many firms mistakenly try
to find solutions through the use However, despite the clarity of these
objectives, the process of accurate
forecasting is becoming increasingly
of IT, investing a considerable
complicated and having correspondingamount of resources in software- ly detrimental effects on the accuracy
of forecasting models (see annex 2).
based programs such as data
Typically, the complexity of a forecastanalysis and statistical software, ing model for an individual product
(which may include multiple SKUs)
will be determined by four key facwithout developing a more
tors: the time-frame of the model, the
geographical coverage, the product’s
comprehensive and wider view channel presence and the total number
of clients served.
of the issue.
perspective DEMYSTIFYING FORECAST ACCURACY
6

Annex 2
MAPE %, 2000-2012

120%

100%

80%

60%

40%

20%

0%
2000

2001

2002

2003

2004

	

2005

2006

Automotive company

Source: Value Partners.

perspective DEMYSTIFYING FORECAST ACCURACY

2007

	
	

2008

2009

Consumer goods (food)
company

2010

	
	

2011

2012

Consumer goods (white)
company
7

By considering each of these four factors in turn, one can deduce the origins
of increasing forecasting complexity.
For instance, with respect to the
geographic coverage, the internationalisation of many firms means that they
are now operating in multiple localities
with a corresponding increase in the
number of inputs to be considered in
the forecasting process. Additionally,
with respect to channel presence, the
emerging trend of ‘omni-channel’ behaviour by businesses where companies
are selling products online, in-stores
and over-the-phone further complicates
their forecasting processes.
These complications increase significantly for firms with a wide-ranging
product portfolio as each individual
product is likely to have a unique set of
drivers and delivery channels and the
sales of one product may have a direct
effect upon the sales of another.
Indeed, these ‘internal’ complications
are mirrored by additional complications
emerging from ‘external’ factors. With
wide-ranging supply chains and target
markets, many companies are becoming
increasingly exposed to a wide range
of market volatilities. Furthermore, it is
important to consider the general health
and direction of the economy overall.
For instance, in a fast and booming
economy, forecasts are likely to have to
take account of growing demand levels
and hence supply volumes will necessarily need to increase as well.

perspective DEMYSTIFYING FORECAST ACCURACY

In a recessive economy, those firms
which ignore the general economic
trends may see a corresponding overestimation of overall volumes.
Despite the complexity of forecasting
increasing, the costs of low levels of
accuracy have remained as high as ever.
Fundamentally, if the aforementioned
complexities are not correctly managed, the consequentially low forecasting accuracy will generate inefficiencies
for the firm, negatively impacting upon
the company’s cash-flow and P&L. More
specifically, these inaccuracies can lead
to inefficiencies in a range of different
ways such as mis-matched levels of
stock, poor levels of service and lower
levels of production flexibility through
poor system planning.
However, when addressing the issue,
many firms mistakenly try to find solutions through the use of IT, investing a
considerable amount of resources in
software-based programs such as data
analytics and statistical software, without developing a more comprehensive
and wider view of the issue.
8

Annex 3

• High level KPls

• ABC error

• Operative KPls

• ...

• ...

1
7

Define key
metrics

• IT systems evolution

Analyze error
determinants

2

• Tools simplicity
• ...

Optimize
tools

FCST

6
Design
the new
process

• Roles

Qualify
the FCST
model
Identify
information
ownership

5

• Activities work flows

• Responsibility
• Info work flows
• ...

Source: Value Partners.

perspective DEMYSTIFYING FORECAST ACCURACY

• Rules & definitions
• ...

3

• Market estimates
• Demand segmentation

4

• ...

• Objectives

Restate
the “basics”

• ...
9

coming to grips with
forecast accuracy:
the power of discipline
is always underrated
It is within this context, and based upon
our extensive experience, that Value
Partners would like to present a range
of ways in which we feel that companies
are able to refine and improve their
forecasting abilities. (see annex 3)
1. Understanding the major causes
of forecasting errors. It is essential to
analyse the deviation between forecast
predictions and recorded data with
respect to two main variables timeframe
and granularity. Firstly, it is necessary
to assess the ability of a firm to develop
estimates in both the short and mediumlong term and to establish whether any
differences emerge between the two.

Annex 4
MAPE %
Forecast Timeframe

+79%
201%
164%
+38%
112%

+45%
81%
56%

1m

68%

2m

3m

Production frozen 60-80%

6m

12m
CAPEX allocation

Source: Value Partners.

perspective DEMYSTIFYING FORECAST ACCURACY

24m

Secondly, it is important to ensure that
the correct level of detail is captured
such that changing trends in total
volume and SKU mix are correctly built
in to the forecast.
Through our previous work, Value
Partners have noticed that the ability
for firms to forecast over different
time periods may vary. For instance,
when analysing different forecast
timeframes during an advisory project
for a major global automotive player,
it emerged that the firm was entirely
focussed on the short term (i.e. 1-2
months), resulting in a considerable
elevation of MAPE (Mean Average
Percentage Error; one of the main
KPIs adopted by enterprises to assess
forecast accuracy) for longer term
forecasts (i.e. over 3 months).
(see annex 4)
Another example can be found in the
case of an international company operating in the white goods space that was
experiencing a limited ability to forecast
their sales mix. This was due to a low
accuracy in estimating total volumes as
a result of a budget that was misaligned
from wider market trends.
Causes for high deviation are numerous, usually hard to identify and driven
by erroneous activities that have often
become common practice over time.
For these reasons it is necessary to
analyse every process and sub-process,
the consistency of the timeframe,
information quality and responsibility
allocation, both at central and local
levels. (see annex 5)
10

Annex 5

Unpredictable demand

Orders to be received
difficult to foreseen

Volumes

+

Almost certain demand

Part of demand to be
segmented with specific
rules

+

Certain demand

Contracted orders
to be delivered

short-term

medium-term
FCST timeframe

Source: Value Partners.

perspective DEMYSTIFYING FORECAST ACCURACY

long-term
11

An in-depth evaluation of forecasting
would enable the tracing of each component of the errors that lead to a high
deviation in forecast accuracy. The overview and identification of every cause
of error is the first step towards the
development of a new forecast model,
which must include every strategic and
operational aspect of the firm.
2. Clarification of the basic rules and
meaning of forecasting. The meaning of
the term ‘forecast’ is vague and could
therefore be interpreted differently by
different firms. In addition, the dimensions upon which forecasts are built are
numerous (orders, shipments, turnover,
etc.) and may have correspondingly
different implications for the forecasting
process. The understanding and interpretation of this term should therefore
be clarified for all involved.
Fundamentally, forecasting accomplishes three main tasks and becomes the input of three different business functions:
Sales (commercial planning), Operations
& Supply-Chain (logistics optimisation
and investment allocation) Corporate
Finance and Control (impact appraisal
and economic/financial results).
An understanding of the desired outputs from a forecasting model should
also be understood in order to ensure
accurate outputs and avoid subsequent
complications.

perspective DEMYSTIFYING FORECAST ACCURACY

For instance, many firms use order
forecasting as a proxy for estimating
turnover, however turnover forecasts
can result in a number of paybacks
(e.g. more coherent financial projections with the firm’s real potential) especially when sales forecasts are based
on complex operating models.
3. Definition of the firm’s specific
forecast model. First and foremost
sales estimates should be based upon
realistic market projections.
The developments of recent years have
proven that the market and the general
economic context can rapidly alter.
For this reason, forecasts (e.g. budget,
strategic plan, etc.) with 6-12 month
timeframes can swiftly prove to be mistaken in their structure and content.
In many cases, it is more appropriate
to define a composite model to estimate the total volumes for each market
and each strategic segment before
improving the capability of forecasting
which is a combination of these separate elements. Such a model should
necessarily include competitive intelligence, which incorporates the actions
and strategies of a company’s main
competitors.
Secondly, clients should be constantly
monitored in order to positively refine
the accuracy of forecasts. Depending
on the business of the firm, segmenting the customer base can facilitate the
assessment of the market at the same
time as helping to capture all of the
relevant phenomena.
12

Annex 6

Commercial planning

(by client)

Service level

(e.g. less Lost
Sales)

2

3

MIX
detail
by sku

Optimization

of product
allocations
and investments
(e.g. moulds)

dimensions

Stock level

rationalization

1
Commercial planning

(by client)
Volumes
by macrO
segment

Factory planning

and investments

short-term

medium-term
FCST timeframe

Source: Value Partners.

perspective DEMYSTIFYING FORECAST ACCURACY

long-term
13

For example, some firms have been
implementing joint-forecast models together with their clients, even integrating their forecast and order management systems to gain a real-time view
over clients’ orders.

4. Identification of information owners.
It is important to assign or identify a
point of contact within a firm with responsibility for data gathering, and for
this individual or group to be appraised
on a performance basis.

Lastly, to guarantee accuracy throughout the processes of market and client
base assessment it is necessary to rigorously define the criteria and practices
that underpin effective demand segmentation.

In addition to a point of contact within a
firm, it is equally important throughout
the forecast process to review the exact
timing of sharing of information on different external/ internal processes.

The forecast should be built upon two
types of demand (with minor differences depending on the business): ‘certain’
demand, relating to received orders;
and ‘open’ demand, referring to potential upcoming orders. (see annex 6)
Of the two, the second demand type is
the most challenging to assess. In order
to gain full understanding it is therefore
crucial to define and employ appropriate methodologies (e.g. the use of
statistics).

The overview and identification
of all the causes is the first
step towards the development
of a new forecast model,
which will have to include
every strategic and operational
aspect of the firm.
perspective DEMYSTIFYING FORECAST ACCURACY

Furthermore, if the quality of the overall
process is contingent on the demand
Planning function, it is also necessary to
develop forecasting in conjunction with
further business functions and integrate
the different information streams to create a holistic picture. Demand Planning,
Sales, Logistics and Pricing should work
together to ensure that all the relevant
data is captured, shared and included in
the forecast.
14

Annex 7

+

<50%

<20%

<30%

fragmentation
(# clients / channels)

<40%

–

–

+
Granularity
(# sku)

Source: Value Partners.

perspective DEMYSTIFYING FORECAST ACCURACY
15

5. Design of the new forecasting process. Once the objectives and nature of
the forecast under development are
established and all required information (including the identification of an
individual responsible for information
gathering), it is then possible to design
a forecasting process that is capable of
improving overall efficiency.
The design of the forecasting process should start from the definition of
strategic objectives, identifying the right
balance of main operational constraints;
a delayed forecast with fewer operational constraints would likely be more
accurate, however the level of service
and the ability of the supply chain to
react promptly would be impaired.
(see annex 7)
Firms with a substantial geographical
heterogeneity should also include the
sharing of information from central and
local sources in the process, as information could be stored at a local level (e.g.
local regulations, client data, local commercial strategy, etc.) as well as at central level (e.g. long-term trends, regional
level trends, product strategy, etc.).
Like all business processes it is always
necessary to ensure substantive commitment from all the parties involved.
Coherently empowering all of the actors
involved in the process and identifying
specific KPIs to appraise every actor’s
contribution to the overall quality of the
forecast facilitates a more productive
and coordinated focus aligned with the
importance of the process.

perspective DEMYSTIFYING FORECAST ACCURACY

Based on our experience we would
argue that many firms suffer low commitment to sales forecasting due to a
deficiency in the clarity of objectives,
absence of rules and a lack of consensus on the value of projects.
6. Support from IT systems to activities. As previously mentioned, the
employment of IT resources does not
directly augment forecasting accuracy,
and may even be an obstacle to the
correct integration of processes and
information in some instances. However,
it remains of primary importance within
the overall process.
Statistical tools must be considered
differently. Several firms have achieved
positive results thanks to the employment of statistical forecasting; many
of these firms are in the spare parts sector which is predominantly influenced
by historical sales rather than market
trends.
When employing statistical tools it becomes critical to ensure that algorithms
are not exclusively based upon historical data though take into consideration
future trends, market trends, macroeconomic metrics (GDP, consumer
index, etc.), market characteristics and
exogenous factors (e.g. new regulations, climate changes, supply chain
stock, etc.).
16

7. Identification of KPIs to monitor.
In order for an organisation’s forecasting performance to be evaluated
successfully it is necessary to identify a
simple and exhaustive set of KPIs along
two main dimensions:
•	 Forecast timeframe – both in the
short and medium-long term

However the MAPE of a single month is
not sufficient to ensure understanding
of the nature of main forecast errors; to
guarantee comprehensive understanding it must be integrated with other
metrics measuring the main forecast
dimensions (e.g. granularity, timeframe,
etc.).

•	 Forecast granularity – measuring the
forecast accuracy on macro figures,
product mix and SKU
Amongst the metrics identified and analysed, MAPE (Mean Average Percentage
Error) is the most valid and accurate
when measuring the difference between
actual and forecast data, as it does not
have the problem concerning averaging positive and negative errors which
afflicts other metrics.

Indeed the forecast accomplishes
three main tasks and becomes the
input of three different business
functions: Sales (commercial
planning), Operations & Supply-Chain
(logistics optimisation and investment
allocation) Corporate Finance and
Control (impact appraisal and
economic/financial results).
perspective DEMYSTIFYING FORECAST ACCURACY
17

Conclusions

In summary, the adoption of an easy
and efficient forecast model based
on sales predictions derived from an
analysis of market trends and the firm’s
commercial plans produces more accurate and realistic forecasts.
The ability to assess the market, understand the main trends and drivers and
react responsively to a market slowdown as well as expansion is not just a
matter of being internally efficient but
also having a competitive advantage
that is hard to replicate.

Based on our extensive experience in
demand forecasting for large industrial
players, we have been able to record
and benchmark errors to develop a
framework for acceptable forecast
errors which should be used by all
organisations.
Such values (which vary depending on
the business) are based upon two main
directives: number of SKUs and market fragmentation, depending on the
number of channels and clients served.
(see annex 8)

Annex 8
Q&A

•	 Demand Planning is one of the roles that people try always
to avoid, I wonder why.

•	 It seems impossible that to have a decent service level,
you must set so high stock targets levels.

•	 the Demand Planner is often alone, neither Logistics nor Sales
are supporting him. Why this lack of inter-functionality?

•	 When Forecast error is high: Demand Planning blames
Logistics and Logistics blame Sales, and so on. So who should
be responsible for MAPE?

perspective DEMYSTIFYING FORECAST ACCURACY
18

AUTHORS

Alberto Calvo
Partner, Milan Office
alberto.calvo@valuepartners.com

alessandro barmettler
Senior Engagement Manager, Milan Office
alessandro.barmettler@valuepartners.com

alberto oteri
Associate, Milan Office
alberto.oteri@valuepartners.com

perspective DEMYSTIFYING FORECAST ACCURACY
19

About
Value Partners

Value Partners is a global management consulting firm that works
with multinational corporations
and high-potential entrepreneurial
businesses to identify and pursue
value enhancement initiatives across
innovation, international expansion,
and operational effectiveness.

In 2007 Value Partners acquired
Spectrum Strategy Consultants – a
leading UK company specialized in
publishing, broadcasting, entertainment, IPTV and mobile – thus further
strengthening its international presence. Today Value Partners is a leading advisor in the telecom, media
and technology sectors worldwide.

Founded in Milan in 1993, Value Partners’ rapid growth testifies to the
value it has created for clients over
time. Today it draws on 25 partners
and 280 professionals from 23 nations, working out of offices in Milan,
London, Istanbul, São Paulo, Buenos
Aires, Beijing, Shanghai, Hong Kong
and Singapore.

For more information on the issues
raised in this note please contact
the authors.
Find all the contact details on
valuepartners.com
Milan
London
Istanbul
São Paulo
Buenos Aires
Beijing
Shanghai
Hong Kong
Singapore

Value Partners has built a portfolio
of more than 350 international
clients from the original 10 in 1993
with a worldwide revenue mix.
Value Partners combines methodological approaches and analytical
frameworks with hands-on attitude
and practical industry experience
developed in an executive capacity
within each sector: telecommunications, new media, financial services,
energy, manufacturing and hi-tech.

Copyright © Value Partners
Management Consulting Limited
All rights reserved
perspective DEMYSTIFYING FORECAST ACCURACY

Más contenido relacionado

La actualidad más candente

La actualidad más candente (20)

Demand forcasting
Demand forcastingDemand forcasting
Demand forcasting
 
6 Demand forecasting
6 Demand forecasting6 Demand forecasting
6 Demand forecasting
 
Demand Forecasting
Demand ForecastingDemand Forecasting
Demand Forecasting
 
Economic forecasting Techniques
Economic forecasting TechniquesEconomic forecasting Techniques
Economic forecasting Techniques
 
3 demand-forecasting
3 demand-forecasting3 demand-forecasting
3 demand-forecasting
 
Demand forcasting
Demand forcastingDemand forcasting
Demand forcasting
 
Demand forecasting.
Demand forecasting.Demand forecasting.
Demand forecasting.
 
Demand forecasting
Demand forecastingDemand forecasting
Demand forecasting
 
Presentation on forecasting
Presentation on forecasting Presentation on forecasting
Presentation on forecasting
 
demand forecasting
demand forecastingdemand forecasting
demand forecasting
 
Fundamental analysis
Fundamental analysisFundamental analysis
Fundamental analysis
 
Demand forecasting
Demand forecastingDemand forecasting
Demand forecasting
 
Demand forecasting
Demand forecastingDemand forecasting
Demand forecasting
 
Fundamental Analysis of securities
Fundamental Analysis of securitiesFundamental Analysis of securities
Fundamental Analysis of securities
 
Demand Forecasting
Demand ForecastingDemand Forecasting
Demand Forecasting
 
Demand forecasting
Demand forecastingDemand forecasting
Demand forecasting
 
Demand forecasting
Demand forecastingDemand forecasting
Demand forecasting
 
Demand forecasting
Demand forecastingDemand forecasting
Demand forecasting
 
Curriculum vitae_Vinit
Curriculum vitae_VinitCurriculum vitae_Vinit
Curriculum vitae_Vinit
 
Demand Forecasting and Market planning
Demand Forecasting and Market planningDemand Forecasting and Market planning
Demand Forecasting and Market planning
 

Similar a Forecast 072013-digiversion

Margin Analysis Example
Margin Analysis ExampleMargin Analysis Example
Margin Analysis ExampleLewis Lin 🦊
 
Digitizing the Supply Chain, from Planning and Procurement to Execution
Digitizing the Supply Chain, from Planning and Procurement to ExecutionDigitizing the Supply Chain, from Planning and Procurement to Execution
Digitizing the Supply Chain, from Planning and Procurement to ExecutionCognizant
 
Property & Casualty Commercial Lines Underwriting: The New Playbook
Property & Casualty Commercial Lines Underwriting: The New PlaybookProperty & Casualty Commercial Lines Underwriting: The New Playbook
Property & Casualty Commercial Lines Underwriting: The New PlaybookCognizant
 
8 Steps to Sustainability Reporting
8 Steps to Sustainability Reporting8 Steps to Sustainability Reporting
8 Steps to Sustainability ReportingJackson Seng
 
Strategic assignment v1
Strategic assignment v1Strategic assignment v1
Strategic assignment v1Tabish Ahmad
 
Role of Analytics in Consumer Packaged Goods Industry
Role of Analytics in Consumer Packaged Goods IndustryRole of Analytics in Consumer Packaged Goods Industry
Role of Analytics in Consumer Packaged Goods IndustryPerceptive Analytics
 
Charting a Course for the Future: A Report on Firm Preparedness
Charting a Course for the Future: A Report on Firm PreparednessCharting a Course for the Future: A Report on Firm Preparedness
Charting a Course for the Future: A Report on Firm PreparednessWolters Kluwer Tax & Accounting US
 
Multidimensional-Performance-Measurement
Multidimensional-Performance-MeasurementMultidimensional-Performance-Measurement
Multidimensional-Performance-MeasurementAlessia Corriero
 
Report_Intern_210.docx
Report_Intern_210.docxReport_Intern_210.docx
Report_Intern_210.docxBandiYashwant
 
Global trend report for CPG Retail 2016_Final
Global trend report for CPG  Retail 2016_FinalGlobal trend report for CPG  Retail 2016_Final
Global trend report for CPG Retail 2016_FinalAbhinav Verma
 
The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...
The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...
The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...Balaji Venkat Chellam Iyer
 
Making Metrics Matter: An investigation of sales-aligned marketing reporting...
Making Metrics Matter: An investigation of sales-aligned marketing  reporting...Making Metrics Matter: An investigation of sales-aligned marketing  reporting...
Making Metrics Matter: An investigation of sales-aligned marketing reporting...Remen Okoruwa
 
Competing to Win in the Media & Entertainment Industry
Competing to Win in the Media & Entertainment IndustryCompeting to Win in the Media & Entertainment Industry
Competing to Win in the Media & Entertainment IndustryCognizant
 
Next Generation S&OP
Next Generation S&OPNext Generation S&OP
Next Generation S&OPPeter Murray
 
Csod investor deck first quarter fina lv3
Csod investor deck first quarter fina lv3Csod investor deck first quarter fina lv3
Csod investor deck first quarter fina lv3ircornerstone
 

Similar a Forecast 072013-digiversion (20)

Margin Analysis Example
Margin Analysis ExampleMargin Analysis Example
Margin Analysis Example
 
Digitizing the Supply Chain, from Planning and Procurement to Execution
Digitizing the Supply Chain, from Planning and Procurement to ExecutionDigitizing the Supply Chain, from Planning and Procurement to Execution
Digitizing the Supply Chain, from Planning and Procurement to Execution
 
Property & Casualty Commercial Lines Underwriting: The New Playbook
Property & Casualty Commercial Lines Underwriting: The New PlaybookProperty & Casualty Commercial Lines Underwriting: The New Playbook
Property & Casualty Commercial Lines Underwriting: The New Playbook
 
A Roadmap To World Class Forecasting Accuracy
A Roadmap To World Class Forecasting AccuracyA Roadmap To World Class Forecasting Accuracy
A Roadmap To World Class Forecasting Accuracy
 
8 Steps to Sustainability Reporting
8 Steps to Sustainability Reporting8 Steps to Sustainability Reporting
8 Steps to Sustainability Reporting
 
Strategic assignment v1
Strategic assignment v1Strategic assignment v1
Strategic assignment v1
 
Role of Analytics in Consumer Packaged Goods Industry
Role of Analytics in Consumer Packaged Goods IndustryRole of Analytics in Consumer Packaged Goods Industry
Role of Analytics in Consumer Packaged Goods Industry
 
Charting a Course for the Future: A Report on Firm Preparedness
Charting a Course for the Future: A Report on Firm PreparednessCharting a Course for the Future: A Report on Firm Preparedness
Charting a Course for the Future: A Report on Firm Preparedness
 
Performance in-the-crosshairs-v3
Performance in-the-crosshairs-v3Performance in-the-crosshairs-v3
Performance in-the-crosshairs-v3
 
Vantage point 2012_issue2
Vantage point 2012_issue2Vantage point 2012_issue2
Vantage point 2012_issue2
 
Multidimensional-Performance-Measurement
Multidimensional-Performance-MeasurementMultidimensional-Performance-Measurement
Multidimensional-Performance-Measurement
 
Investor Day 2013: Global Strategy in Analytics
Investor Day 2013: Global Strategy in AnalyticsInvestor Day 2013: Global Strategy in Analytics
Investor Day 2013: Global Strategy in Analytics
 
Report_Intern_210.docx
Report_Intern_210.docxReport_Intern_210.docx
Report_Intern_210.docx
 
Global trend report for CPG Retail 2016_Final
Global trend report for CPG  Retail 2016_FinalGlobal trend report for CPG  Retail 2016_Final
Global trend report for CPG Retail 2016_Final
 
The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...
The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...
The new ‘A and B’ of the Finance Function: Analytics and Big Data - -Evolutio...
 
Making Metrics Matter: An investigation of sales-aligned marketing reporting...
Making Metrics Matter: An investigation of sales-aligned marketing  reporting...Making Metrics Matter: An investigation of sales-aligned marketing  reporting...
Making Metrics Matter: An investigation of sales-aligned marketing reporting...
 
Competing to Win in the Media & Entertainment Industry
Competing to Win in the Media & Entertainment IndustryCompeting to Win in the Media & Entertainment Industry
Competing to Win in the Media & Entertainment Industry
 
SCM
SCMSCM
SCM
 
Next Generation S&OP
Next Generation S&OPNext Generation S&OP
Next Generation S&OP
 
Csod investor deck first quarter fina lv3
Csod investor deck first quarter fina lv3Csod investor deck first quarter fina lv3
Csod investor deck first quarter fina lv3
 

Más de Value Partners

Credit cards regulation evolution
Credit cards regulation evolutionCredit cards regulation evolution
Credit cards regulation evolutionValue Partners
 
The policy and prospects of China’s fixed broadband Market liberalization
The policy and prospects of China’s fixed broadband Market liberalizationThe policy and prospects of China’s fixed broadband Market liberalization
The policy and prospects of China’s fixed broadband Market liberalizationValue Partners
 
Dynamic ticket pricing. Squeezing more juice from half time oranges
Dynamic ticket pricing. Squeezing more juice from half time oranges  Dynamic ticket pricing. Squeezing more juice from half time oranges
Dynamic ticket pricing. Squeezing more juice from half time oranges Value Partners
 
The Future of Mobile Roaming Helping mobile operators remain competitive in t...
The Future of Mobile Roaming Helping mobile operators remain competitive in t...The Future of Mobile Roaming Helping mobile operators remain competitive in t...
The Future of Mobile Roaming Helping mobile operators remain competitive in t...Value Partners
 
What does it take for brands to go digital. Same but different
What does it take for brands to go digital. Same but different  What does it take for brands to go digital. Same but different
What does it take for brands to go digital. Same but different Value Partners
 
Online to offline. What is in for traditional retailers?
Online to offline. What is in for traditional retailers?  Online to offline. What is in for traditional retailers?
Online to offline. What is in for traditional retailers? Value Partners
 
Connected as it never was. The launch of China's MVNOs
Connected as it never was. The launch of China's MVNOs  Connected as it never was. The launch of China's MVNOs
Connected as it never was. The launch of China's MVNOs Value Partners
 
Customer Service: Achieving excellence through a company-wide approach
Customer Service: Achieving excellence through a company-wide approachCustomer Service: Achieving excellence through a company-wide approach
Customer Service: Achieving excellence through a company-wide approachValue Partners
 
ATM Benchmarking Study 2014 and Industry Report
ATM Benchmarking Study 2014 and Industry Report  ATM Benchmarking Study 2014 and Industry Report
ATM Benchmarking Study 2014 and Industry Report Value Partners
 
Magazine Publishers' Transformation: The Time to Act is Now!
Magazine Publishers' Transformation: The Time to Act is Now!  Magazine Publishers' Transformation: The Time to Act is Now!
Magazine Publishers' Transformation: The Time to Act is Now! Value Partners
 
Transparency health sector 122013
Transparency health sector 122013Transparency health sector 122013
Transparency health sector 122013Value Partners
 
Perspective online-luxury-012013-digiversion
Perspective online-luxury-012013-digiversionPerspective online-luxury-012013-digiversion
Perspective online-luxury-012013-digiversionValue Partners
 
Mobile spectrum-032013-digiversion
Mobile spectrum-032013-digiversionMobile spectrum-032013-digiversion
Mobile spectrum-032013-digiversionValue Partners
 
Broadband data-052013-digiversion
Broadband data-052013-digiversionBroadband data-052013-digiversion
Broadband data-052013-digiversionValue Partners
 
Private equity 072013-digiversion
Private equity 072013-digiversionPrivate equity 072013-digiversion
Private equity 072013-digiversionValue Partners
 
Mechanization of farms 112013-digiversion
Mechanization of farms 112013-digiversionMechanization of farms 112013-digiversion
Mechanization of farms 112013-digiversionValue Partners
 
Vp financial-fraud-report-digiversion
Vp financial-fraud-report-digiversionVp financial-fraud-report-digiversion
Vp financial-fraud-report-digiversionValue Partners
 
Broadband data-052013-digiversion
Broadband data-052013-digiversionBroadband data-052013-digiversion
Broadband data-052013-digiversionValue Partners
 
Beating the Offside Trap
Beating the Offside TrapBeating the Offside Trap
Beating the Offside TrapValue Partners
 

Más de Value Partners (20)

Credit cards regulation evolution
Credit cards regulation evolutionCredit cards regulation evolution
Credit cards regulation evolution
 
The policy and prospects of China’s fixed broadband Market liberalization
The policy and prospects of China’s fixed broadband Market liberalizationThe policy and prospects of China’s fixed broadband Market liberalization
The policy and prospects of China’s fixed broadband Market liberalization
 
Dynamic ticket pricing. Squeezing more juice from half time oranges
Dynamic ticket pricing. Squeezing more juice from half time oranges  Dynamic ticket pricing. Squeezing more juice from half time oranges
Dynamic ticket pricing. Squeezing more juice from half time oranges
 
The Future of Mobile Roaming Helping mobile operators remain competitive in t...
The Future of Mobile Roaming Helping mobile operators remain competitive in t...The Future of Mobile Roaming Helping mobile operators remain competitive in t...
The Future of Mobile Roaming Helping mobile operators remain competitive in t...
 
What does it take for brands to go digital. Same but different
What does it take for brands to go digital. Same but different  What does it take for brands to go digital. Same but different
What does it take for brands to go digital. Same but different
 
Online to offline. What is in for traditional retailers?
Online to offline. What is in for traditional retailers?  Online to offline. What is in for traditional retailers?
Online to offline. What is in for traditional retailers?
 
Connected as it never was. The launch of China's MVNOs
Connected as it never was. The launch of China's MVNOs  Connected as it never was. The launch of China's MVNOs
Connected as it never was. The launch of China's MVNOs
 
Customer Service: Achieving excellence through a company-wide approach
Customer Service: Achieving excellence through a company-wide approachCustomer Service: Achieving excellence through a company-wide approach
Customer Service: Achieving excellence through a company-wide approach
 
ATM Benchmarking Study 2014 and Industry Report
ATM Benchmarking Study 2014 and Industry Report  ATM Benchmarking Study 2014 and Industry Report
ATM Benchmarking Study 2014 and Industry Report
 
Magazine Publishers' Transformation: The Time to Act is Now!
Magazine Publishers' Transformation: The Time to Act is Now!  Magazine Publishers' Transformation: The Time to Act is Now!
Magazine Publishers' Transformation: The Time to Act is Now!
 
Transparency health sector 122013
Transparency health sector 122013Transparency health sector 122013
Transparency health sector 122013
 
Perspective online-luxury-012013-digiversion
Perspective online-luxury-012013-digiversionPerspective online-luxury-012013-digiversion
Perspective online-luxury-012013-digiversion
 
Mobile spectrum-032013-digiversion
Mobile spectrum-032013-digiversionMobile spectrum-032013-digiversion
Mobile spectrum-032013-digiversion
 
Broadband data-052013-digiversion
Broadband data-052013-digiversionBroadband data-052013-digiversion
Broadband data-052013-digiversion
 
Private equity 072013-digiversion
Private equity 072013-digiversionPrivate equity 072013-digiversion
Private equity 072013-digiversion
 
Mechanization of farms 112013-digiversion
Mechanization of farms 112013-digiversionMechanization of farms 112013-digiversion
Mechanization of farms 112013-digiversion
 
Vp financial-fraud-report-digiversion
Vp financial-fraud-report-digiversionVp financial-fraud-report-digiversion
Vp financial-fraud-report-digiversion
 
Broadband data-052013-digiversion
Broadband data-052013-digiversionBroadband data-052013-digiversion
Broadband data-052013-digiversion
 
Pharma china-042013-1
Pharma china-042013-1Pharma china-042013-1
Pharma china-042013-1
 
Beating the Offside Trap
Beating the Offside TrapBeating the Offside Trap
Beating the Offside Trap
 

Último

Pitch Deck Teardown: Geodesic.Life's $500k Pre-seed deck
Pitch Deck Teardown: Geodesic.Life's $500k Pre-seed deckPitch Deck Teardown: Geodesic.Life's $500k Pre-seed deck
Pitch Deck Teardown: Geodesic.Life's $500k Pre-seed deckHajeJanKamps
 
Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Seta Wicaksana
 
Church Building Grants To Assist With New Construction, Additions, And Restor...
Church Building Grants To Assist With New Construction, Additions, And Restor...Church Building Grants To Assist With New Construction, Additions, And Restor...
Church Building Grants To Assist With New Construction, Additions, And Restor...Americas Got Grants
 
Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.Anamaria Contreras
 
The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...
The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...
The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...Operational Excellence Consulting
 
Welding Electrode Making Machine By Deccan Dynamics
Welding Electrode Making Machine By Deccan DynamicsWelding Electrode Making Machine By Deccan Dynamics
Welding Electrode Making Machine By Deccan DynamicsIndiaMART InterMESH Limited
 
Buy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail AccountsBuy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail AccountsBuy Verified Accounts
 
business environment micro environment macro environment.pptx
business environment micro environment macro environment.pptxbusiness environment micro environment macro environment.pptx
business environment micro environment macro environment.pptxShruti Mittal
 
Unveiling the Soundscape Music for Psychedelic Experiences
Unveiling the Soundscape Music for Psychedelic ExperiencesUnveiling the Soundscape Music for Psychedelic Experiences
Unveiling the Soundscape Music for Psychedelic ExperiencesDoe Paoro
 
Excvation Safety for safety officers reference
Excvation Safety for safety officers referenceExcvation Safety for safety officers reference
Excvation Safety for safety officers referencessuser2c065e
 
Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Peter Ward
 
Entrepreneurship lessons in Philippines
Entrepreneurship lessons in  PhilippinesEntrepreneurship lessons in  Philippines
Entrepreneurship lessons in PhilippinesDavidSamuel525586
 
20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdf
20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdf20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdf
20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdfChris Skinner
 
Onemonitar Android Spy App Features: Explore Advanced Monitoring Capabilities
Onemonitar Android Spy App Features: Explore Advanced Monitoring CapabilitiesOnemonitar Android Spy App Features: Explore Advanced Monitoring Capabilities
Onemonitar Android Spy App Features: Explore Advanced Monitoring CapabilitiesOne Monitar
 
MAHA Global and IPR: Do Actions Speak Louder Than Words?
MAHA Global and IPR: Do Actions Speak Louder Than Words?MAHA Global and IPR: Do Actions Speak Louder Than Words?
MAHA Global and IPR: Do Actions Speak Louder Than Words?Olivia Kresic
 
WSMM Technology February.March Newsletter_vF.pdf
WSMM Technology February.March Newsletter_vF.pdfWSMM Technology February.March Newsletter_vF.pdf
WSMM Technology February.March Newsletter_vF.pdfJamesConcepcion7
 
Memorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQMMemorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQMVoces Mineras
 

Último (20)

Pitch Deck Teardown: Geodesic.Life's $500k Pre-seed deck
Pitch Deck Teardown: Geodesic.Life's $500k Pre-seed deckPitch Deck Teardown: Geodesic.Life's $500k Pre-seed deck
Pitch Deck Teardown: Geodesic.Life's $500k Pre-seed deck
 
Japan IT Week 2024 Brochure by 47Billion (English)
Japan IT Week 2024 Brochure by 47Billion (English)Japan IT Week 2024 Brochure by 47Billion (English)
Japan IT Week 2024 Brochure by 47Billion (English)
 
Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...
 
Church Building Grants To Assist With New Construction, Additions, And Restor...
Church Building Grants To Assist With New Construction, Additions, And Restor...Church Building Grants To Assist With New Construction, Additions, And Restor...
Church Building Grants To Assist With New Construction, Additions, And Restor...
 
WAM Corporate Presentation April 12 2024.pdf
WAM Corporate Presentation April 12 2024.pdfWAM Corporate Presentation April 12 2024.pdf
WAM Corporate Presentation April 12 2024.pdf
 
Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.
 
The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...
The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...
The McKinsey 7S Framework: A Holistic Approach to Harmonizing All Parts of th...
 
Welding Electrode Making Machine By Deccan Dynamics
Welding Electrode Making Machine By Deccan DynamicsWelding Electrode Making Machine By Deccan Dynamics
Welding Electrode Making Machine By Deccan Dynamics
 
Buy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail AccountsBuy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail Accounts
 
business environment micro environment macro environment.pptx
business environment micro environment macro environment.pptxbusiness environment micro environment macro environment.pptx
business environment micro environment macro environment.pptx
 
Unveiling the Soundscape Music for Psychedelic Experiences
Unveiling the Soundscape Music for Psychedelic ExperiencesUnveiling the Soundscape Music for Psychedelic Experiences
Unveiling the Soundscape Music for Psychedelic Experiences
 
Excvation Safety for safety officers reference
Excvation Safety for safety officers referenceExcvation Safety for safety officers reference
Excvation Safety for safety officers reference
 
Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...
 
Entrepreneurship lessons in Philippines
Entrepreneurship lessons in  PhilippinesEntrepreneurship lessons in  Philippines
Entrepreneurship lessons in Philippines
 
20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdf
20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdf20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdf
20220816-EthicsGrade_Scorecard-JP_Morgan_Chase-Q2-63_57.pdf
 
Onemonitar Android Spy App Features: Explore Advanced Monitoring Capabilities
Onemonitar Android Spy App Features: Explore Advanced Monitoring CapabilitiesOnemonitar Android Spy App Features: Explore Advanced Monitoring Capabilities
Onemonitar Android Spy App Features: Explore Advanced Monitoring Capabilities
 
Corporate Profile 47Billion Information Technology
Corporate Profile 47Billion Information TechnologyCorporate Profile 47Billion Information Technology
Corporate Profile 47Billion Information Technology
 
MAHA Global and IPR: Do Actions Speak Louder Than Words?
MAHA Global and IPR: Do Actions Speak Louder Than Words?MAHA Global and IPR: Do Actions Speak Louder Than Words?
MAHA Global and IPR: Do Actions Speak Louder Than Words?
 
WSMM Technology February.March Newsletter_vF.pdf
WSMM Technology February.March Newsletter_vF.pdfWSMM Technology February.March Newsletter_vF.pdf
WSMM Technology February.March Newsletter_vF.pdf
 
Memorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQMMemorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQM
 

Forecast 072013-digiversion

  • 1. perspective JUL 2013 DEMYSTIFYING FORECAST ACCURACY ACHIEVING HIGH SALES FORECAST ACCURACY IS A MATTER OF DISCIPLINE Alberto Calvo, Alessandro Barmettler, Alberto Oteri
  • 2. Demystifying forecast accuracy: achieving high sales forecast accuracy is a matter of discipline Published by Value Partners Management Consulting via Vespri Siciliani 9 20146 Milan, Italy July 2013 Written and edited by: Alberto Calvo, Alessandro Barmettler, Alberto Oteri If you would like an electronic copy please write to: cristina.goddi@valuepartners.com For more information on the issues raised in the report please contact: alberto.calvo@valuepartners.com alessandro.barmettler@ valuepartners.com alberto.oteri@valuepartners.com If you would like to subscribe or to be removed from our mailing list please write to: subscription@valuepartners.com valuepartners.com Copyright © Value Partners Management Consulting All rights reserved
  • 3. 3 executive summary My wife and I walked into a Milan patisserie to buy a small cake. We walked out with a tray full of mouthwatering cannoli, in a quantity suitable for a small army. For free. How did that happen? Something with the shop’s forecasting had gone terribly wrong that day. Whilst thanking the chef for this unexpected tray-full of bliss we found out that more or less each evening he has to somehow ‘dispose’ of unsold cakes, either supplying them to soup kitchens or handing them out for free to unsuspecting clients. If an apparently simple operation such as a patisserie – employing 10 skilled workers who have crafted the same products for decades – isn’t able to plan its production accurately, then how can a multinational company successfully manage its own forecasting? Like our patisserie, multinationals struggle with forecasting though this is not altogether surprising – demand planning is one the most challenging jobs around. perspective DEMYSTIFYING FORECAST ACCURACY
  • 4. 4 Annex 1 Europe, Quarterly variation of Sales, Percentage, 1Q 2000 – 1Q 2013 2 1,5 1 31% 0,5 14% 57% 0 31% -0,5 -1 -1,5 -2 2000 2001 2002 2003 2004 2005 2006 2007 Standard deviation Source: Eurostat. perspective DEMYSTIFYING FORECAST ACCURACY 2008 2009 Automotive 2010 2011 2012 2013 Food, beverages, tobacco
  • 5. 5 have you to cope with increasing volatility ? Over the last few years a number of firms have suffered the effects of increased market demand volatility, especially those with a varied product offering and a high number of SKUs. (see annex 1) As the opening example showed, the process of demand forecasting is not only crucial to a company’s success, but is also inherently challenging. Within the context of larger firms, this process may become further complicated through the interplay of many factors such as investment decisions, supply chain management and strategic planning. From our experiences, we believe that the process of company forecasting seeks to address four key objectives: 1. To guide and support commercial planning – helping to align objectives regarding volume, channels, client, price and strategy 2. To calibrate the supply chain and operations – enabling an increase in service levels through the optimisation of logistics and manufacturing planning in the short term 3. To decide which products to manufacture in which plant – optimising the investment decisions in the medium-long term 4. To quantify the expected economic and financial results – helping anticipate and forecast future financial incomes, maintain control over strategic directives, manage cash flows, etc. However when addressing the issue, many firms mistakenly try to find solutions through the use However, despite the clarity of these objectives, the process of accurate forecasting is becoming increasingly of IT, investing a considerable complicated and having correspondingamount of resources in software- ly detrimental effects on the accuracy of forecasting models (see annex 2). based programs such as data Typically, the complexity of a forecastanalysis and statistical software, ing model for an individual product (which may include multiple SKUs) will be determined by four key facwithout developing a more tors: the time-frame of the model, the geographical coverage, the product’s comprehensive and wider view channel presence and the total number of clients served. of the issue. perspective DEMYSTIFYING FORECAST ACCURACY
  • 6. 6 Annex 2 MAPE %, 2000-2012 120% 100% 80% 60% 40% 20% 0% 2000 2001 2002 2003 2004 2005 2006 Automotive company Source: Value Partners. perspective DEMYSTIFYING FORECAST ACCURACY 2007 2008 2009 Consumer goods (food) company 2010 2011 2012 Consumer goods (white) company
  • 7. 7 By considering each of these four factors in turn, one can deduce the origins of increasing forecasting complexity. For instance, with respect to the geographic coverage, the internationalisation of many firms means that they are now operating in multiple localities with a corresponding increase in the number of inputs to be considered in the forecasting process. Additionally, with respect to channel presence, the emerging trend of ‘omni-channel’ behaviour by businesses where companies are selling products online, in-stores and over-the-phone further complicates their forecasting processes. These complications increase significantly for firms with a wide-ranging product portfolio as each individual product is likely to have a unique set of drivers and delivery channels and the sales of one product may have a direct effect upon the sales of another. Indeed, these ‘internal’ complications are mirrored by additional complications emerging from ‘external’ factors. With wide-ranging supply chains and target markets, many companies are becoming increasingly exposed to a wide range of market volatilities. Furthermore, it is important to consider the general health and direction of the economy overall. For instance, in a fast and booming economy, forecasts are likely to have to take account of growing demand levels and hence supply volumes will necessarily need to increase as well. perspective DEMYSTIFYING FORECAST ACCURACY In a recessive economy, those firms which ignore the general economic trends may see a corresponding overestimation of overall volumes. Despite the complexity of forecasting increasing, the costs of low levels of accuracy have remained as high as ever. Fundamentally, if the aforementioned complexities are not correctly managed, the consequentially low forecasting accuracy will generate inefficiencies for the firm, negatively impacting upon the company’s cash-flow and P&L. More specifically, these inaccuracies can lead to inefficiencies in a range of different ways such as mis-matched levels of stock, poor levels of service and lower levels of production flexibility through poor system planning. However, when addressing the issue, many firms mistakenly try to find solutions through the use of IT, investing a considerable amount of resources in software-based programs such as data analytics and statistical software, without developing a more comprehensive and wider view of the issue.
  • 8. 8 Annex 3 • High level KPls • ABC error • Operative KPls • ... • ... 1 7 Define key metrics • IT systems evolution Analyze error determinants 2 • Tools simplicity • ... Optimize tools FCST 6 Design the new process • Roles Qualify the FCST model Identify information ownership 5 • Activities work flows • Responsibility • Info work flows • ... Source: Value Partners. perspective DEMYSTIFYING FORECAST ACCURACY • Rules & definitions • ... 3 • Market estimates • Demand segmentation 4 • ... • Objectives Restate the “basics” • ...
  • 9. 9 coming to grips with forecast accuracy: the power of discipline is always underrated It is within this context, and based upon our extensive experience, that Value Partners would like to present a range of ways in which we feel that companies are able to refine and improve their forecasting abilities. (see annex 3) 1. Understanding the major causes of forecasting errors. It is essential to analyse the deviation between forecast predictions and recorded data with respect to two main variables timeframe and granularity. Firstly, it is necessary to assess the ability of a firm to develop estimates in both the short and mediumlong term and to establish whether any differences emerge between the two. Annex 4 MAPE % Forecast Timeframe +79% 201% 164% +38% 112% +45% 81% 56% 1m 68% 2m 3m Production frozen 60-80% 6m 12m CAPEX allocation Source: Value Partners. perspective DEMYSTIFYING FORECAST ACCURACY 24m Secondly, it is important to ensure that the correct level of detail is captured such that changing trends in total volume and SKU mix are correctly built in to the forecast. Through our previous work, Value Partners have noticed that the ability for firms to forecast over different time periods may vary. For instance, when analysing different forecast timeframes during an advisory project for a major global automotive player, it emerged that the firm was entirely focussed on the short term (i.e. 1-2 months), resulting in a considerable elevation of MAPE (Mean Average Percentage Error; one of the main KPIs adopted by enterprises to assess forecast accuracy) for longer term forecasts (i.e. over 3 months). (see annex 4) Another example can be found in the case of an international company operating in the white goods space that was experiencing a limited ability to forecast their sales mix. This was due to a low accuracy in estimating total volumes as a result of a budget that was misaligned from wider market trends. Causes for high deviation are numerous, usually hard to identify and driven by erroneous activities that have often become common practice over time. For these reasons it is necessary to analyse every process and sub-process, the consistency of the timeframe, information quality and responsibility allocation, both at central and local levels. (see annex 5)
  • 10. 10 Annex 5 Unpredictable demand Orders to be received difficult to foreseen Volumes + Almost certain demand Part of demand to be segmented with specific rules + Certain demand Contracted orders to be delivered short-term medium-term FCST timeframe Source: Value Partners. perspective DEMYSTIFYING FORECAST ACCURACY long-term
  • 11. 11 An in-depth evaluation of forecasting would enable the tracing of each component of the errors that lead to a high deviation in forecast accuracy. The overview and identification of every cause of error is the first step towards the development of a new forecast model, which must include every strategic and operational aspect of the firm. 2. Clarification of the basic rules and meaning of forecasting. The meaning of the term ‘forecast’ is vague and could therefore be interpreted differently by different firms. In addition, the dimensions upon which forecasts are built are numerous (orders, shipments, turnover, etc.) and may have correspondingly different implications for the forecasting process. The understanding and interpretation of this term should therefore be clarified for all involved. Fundamentally, forecasting accomplishes three main tasks and becomes the input of three different business functions: Sales (commercial planning), Operations & Supply-Chain (logistics optimisation and investment allocation) Corporate Finance and Control (impact appraisal and economic/financial results). An understanding of the desired outputs from a forecasting model should also be understood in order to ensure accurate outputs and avoid subsequent complications. perspective DEMYSTIFYING FORECAST ACCURACY For instance, many firms use order forecasting as a proxy for estimating turnover, however turnover forecasts can result in a number of paybacks (e.g. more coherent financial projections with the firm’s real potential) especially when sales forecasts are based on complex operating models. 3. Definition of the firm’s specific forecast model. First and foremost sales estimates should be based upon realistic market projections. The developments of recent years have proven that the market and the general economic context can rapidly alter. For this reason, forecasts (e.g. budget, strategic plan, etc.) with 6-12 month timeframes can swiftly prove to be mistaken in their structure and content. In many cases, it is more appropriate to define a composite model to estimate the total volumes for each market and each strategic segment before improving the capability of forecasting which is a combination of these separate elements. Such a model should necessarily include competitive intelligence, which incorporates the actions and strategies of a company’s main competitors. Secondly, clients should be constantly monitored in order to positively refine the accuracy of forecasts. Depending on the business of the firm, segmenting the customer base can facilitate the assessment of the market at the same time as helping to capture all of the relevant phenomena.
  • 12. 12 Annex 6 Commercial planning (by client) Service level (e.g. less Lost Sales) 2 3 MIX detail by sku Optimization of product allocations and investments (e.g. moulds) dimensions Stock level rationalization 1 Commercial planning (by client) Volumes by macrO segment Factory planning and investments short-term medium-term FCST timeframe Source: Value Partners. perspective DEMYSTIFYING FORECAST ACCURACY long-term
  • 13. 13 For example, some firms have been implementing joint-forecast models together with their clients, even integrating their forecast and order management systems to gain a real-time view over clients’ orders. 4. Identification of information owners. It is important to assign or identify a point of contact within a firm with responsibility for data gathering, and for this individual or group to be appraised on a performance basis. Lastly, to guarantee accuracy throughout the processes of market and client base assessment it is necessary to rigorously define the criteria and practices that underpin effective demand segmentation. In addition to a point of contact within a firm, it is equally important throughout the forecast process to review the exact timing of sharing of information on different external/ internal processes. The forecast should be built upon two types of demand (with minor differences depending on the business): ‘certain’ demand, relating to received orders; and ‘open’ demand, referring to potential upcoming orders. (see annex 6) Of the two, the second demand type is the most challenging to assess. In order to gain full understanding it is therefore crucial to define and employ appropriate methodologies (e.g. the use of statistics). The overview and identification of all the causes is the first step towards the development of a new forecast model, which will have to include every strategic and operational aspect of the firm. perspective DEMYSTIFYING FORECAST ACCURACY Furthermore, if the quality of the overall process is contingent on the demand Planning function, it is also necessary to develop forecasting in conjunction with further business functions and integrate the different information streams to create a holistic picture. Demand Planning, Sales, Logistics and Pricing should work together to ensure that all the relevant data is captured, shared and included in the forecast.
  • 14. 14 Annex 7 + <50% <20% <30% fragmentation (# clients / channels) <40% – – + Granularity (# sku) Source: Value Partners. perspective DEMYSTIFYING FORECAST ACCURACY
  • 15. 15 5. Design of the new forecasting process. Once the objectives and nature of the forecast under development are established and all required information (including the identification of an individual responsible for information gathering), it is then possible to design a forecasting process that is capable of improving overall efficiency. The design of the forecasting process should start from the definition of strategic objectives, identifying the right balance of main operational constraints; a delayed forecast with fewer operational constraints would likely be more accurate, however the level of service and the ability of the supply chain to react promptly would be impaired. (see annex 7) Firms with a substantial geographical heterogeneity should also include the sharing of information from central and local sources in the process, as information could be stored at a local level (e.g. local regulations, client data, local commercial strategy, etc.) as well as at central level (e.g. long-term trends, regional level trends, product strategy, etc.). Like all business processes it is always necessary to ensure substantive commitment from all the parties involved. Coherently empowering all of the actors involved in the process and identifying specific KPIs to appraise every actor’s contribution to the overall quality of the forecast facilitates a more productive and coordinated focus aligned with the importance of the process. perspective DEMYSTIFYING FORECAST ACCURACY Based on our experience we would argue that many firms suffer low commitment to sales forecasting due to a deficiency in the clarity of objectives, absence of rules and a lack of consensus on the value of projects. 6. Support from IT systems to activities. As previously mentioned, the employment of IT resources does not directly augment forecasting accuracy, and may even be an obstacle to the correct integration of processes and information in some instances. However, it remains of primary importance within the overall process. Statistical tools must be considered differently. Several firms have achieved positive results thanks to the employment of statistical forecasting; many of these firms are in the spare parts sector which is predominantly influenced by historical sales rather than market trends. When employing statistical tools it becomes critical to ensure that algorithms are not exclusively based upon historical data though take into consideration future trends, market trends, macroeconomic metrics (GDP, consumer index, etc.), market characteristics and exogenous factors (e.g. new regulations, climate changes, supply chain stock, etc.).
  • 16. 16 7. Identification of KPIs to monitor. In order for an organisation’s forecasting performance to be evaluated successfully it is necessary to identify a simple and exhaustive set of KPIs along two main dimensions: • Forecast timeframe – both in the short and medium-long term However the MAPE of a single month is not sufficient to ensure understanding of the nature of main forecast errors; to guarantee comprehensive understanding it must be integrated with other metrics measuring the main forecast dimensions (e.g. granularity, timeframe, etc.). • Forecast granularity – measuring the forecast accuracy on macro figures, product mix and SKU Amongst the metrics identified and analysed, MAPE (Mean Average Percentage Error) is the most valid and accurate when measuring the difference between actual and forecast data, as it does not have the problem concerning averaging positive and negative errors which afflicts other metrics. Indeed the forecast accomplishes three main tasks and becomes the input of three different business functions: Sales (commercial planning), Operations & Supply-Chain (logistics optimisation and investment allocation) Corporate Finance and Control (impact appraisal and economic/financial results). perspective DEMYSTIFYING FORECAST ACCURACY
  • 17. 17 Conclusions In summary, the adoption of an easy and efficient forecast model based on sales predictions derived from an analysis of market trends and the firm’s commercial plans produces more accurate and realistic forecasts. The ability to assess the market, understand the main trends and drivers and react responsively to a market slowdown as well as expansion is not just a matter of being internally efficient but also having a competitive advantage that is hard to replicate. Based on our extensive experience in demand forecasting for large industrial players, we have been able to record and benchmark errors to develop a framework for acceptable forecast errors which should be used by all organisations. Such values (which vary depending on the business) are based upon two main directives: number of SKUs and market fragmentation, depending on the number of channels and clients served. (see annex 8) Annex 8 Q&A • Demand Planning is one of the roles that people try always to avoid, I wonder why. • It seems impossible that to have a decent service level, you must set so high stock targets levels. • the Demand Planner is often alone, neither Logistics nor Sales are supporting him. Why this lack of inter-functionality? • When Forecast error is high: Demand Planning blames Logistics and Logistics blame Sales, and so on. So who should be responsible for MAPE? perspective DEMYSTIFYING FORECAST ACCURACY
  • 18. 18 AUTHORS Alberto Calvo Partner, Milan Office alberto.calvo@valuepartners.com alessandro barmettler Senior Engagement Manager, Milan Office alessandro.barmettler@valuepartners.com alberto oteri Associate, Milan Office alberto.oteri@valuepartners.com perspective DEMYSTIFYING FORECAST ACCURACY
  • 19. 19 About Value Partners Value Partners is a global management consulting firm that works with multinational corporations and high-potential entrepreneurial businesses to identify and pursue value enhancement initiatives across innovation, international expansion, and operational effectiveness. In 2007 Value Partners acquired Spectrum Strategy Consultants – a leading UK company specialized in publishing, broadcasting, entertainment, IPTV and mobile – thus further strengthening its international presence. Today Value Partners is a leading advisor in the telecom, media and technology sectors worldwide. Founded in Milan in 1993, Value Partners’ rapid growth testifies to the value it has created for clients over time. Today it draws on 25 partners and 280 professionals from 23 nations, working out of offices in Milan, London, Istanbul, São Paulo, Buenos Aires, Beijing, Shanghai, Hong Kong and Singapore. For more information on the issues raised in this note please contact the authors. Find all the contact details on valuepartners.com Milan London Istanbul São Paulo Buenos Aires Beijing Shanghai Hong Kong Singapore Value Partners has built a portfolio of more than 350 international clients from the original 10 in 1993 with a worldwide revenue mix. Value Partners combines methodological approaches and analytical frameworks with hands-on attitude and practical industry experience developed in an executive capacity within each sector: telecommunications, new media, financial services, energy, manufacturing and hi-tech. Copyright © Value Partners Management Consulting Limited All rights reserved perspective DEMYSTIFYING FORECAST ACCURACY