Strategic Analysis Tools Book
Table of Contents
Tools Book Key Question
1. Strategy Concepts Porter’s 5 Forces SWOT Growth Share (BCG) Matrix Product Life Cycle EOS : ROS ROS – RMS EOS : Learning Curve Typical Output Format ROS / RONA Value Disciplines Happy Line Conversion Waterfall SWOT •
Tool Description • • • • • • • • • •
What are the key market drivers ? What are the top-line factors affecting the issue at hand ? What industries or sectors should we be in ? What stage of the product cycle are we at ? Does size matter ? What are the benefits of scale or experience ? How well are we using our assets ? What type of possible strategies are there in this market ? Where are we over-satisfying our customers ? Where do we lose potential customers ? What are the top-line factors affecting the issue at hand ?
2. Key Visual Tools Pareto Curve (“80:20”) Value Map (“Mekko”) Value Chain Weighted Column Maker Share Gain Line KPC Combs Harvey Balls Quotes Typical Process Lollipops Channel Maps Conceptual Maps Forces at Work Crocodile Chart Parfait • • • • • • • • • • • • • •
What matters ? What are the various segments and sub-segments ?. Where do our core competencies reside ? Which are the most profitable competitors in absolute terms ? Who has been gaining market share ? What are the main purchasing criteria, and how do we fare ? What is the best way way to display “soft” quantitative data ? What is the best way to display qualitative data ? icks &creative Tips data displays of my own ? Can I come upTrwith What is the relative importance of various routes to market ? How can I simply structure a set of data before analysing it ? What are the forces at work ? Can I come up with creative data displays of my own too ? What is the easiest way to display market growth ?
3. Analysis Techniques • • • • • • • • •
Logic Tree Process Maps Sanity Checks Top-down Estimation One-Off Interviews Interview Programmes Focus Groups Clustering Factorial Analysis
How does one organise the research ? What are the key steps ? Does this number make sense ? How can I estimate a sub-set sub-set ? What are the main issues with one-off interviews ? What are the main issues with interview programmes ? What are the main issues with organising focus groups ? How can I group people by preferences ? How can I group people’s preferences in broader sets ?
4. Sources • •
Desk Sources Usual AABC Sources
Where can I find secondary data ? What are the usual AABC sources ?
2
Table of Contents
Tools Book Key Question
1. Strategy Concepts Porter’s 5 Forces SWOT Growth Share (BCG) Matrix Product Life Cycle EOS : ROS ROS – RMS EOS : Learning Curve Typical Output Format ROS / RONA Value Disciplines Happy Line Conversion Waterfall SWOT •
Tool Description • • • • • • • • • •
What are the key market drivers ? What are the top-line factors affecting the issue at hand ? What industries or sectors should we be in ? What stage of the product cycle are we at ? Does size matter ? What are the benefits of scale or experience ? How well are we using our assets ? What type of possible strategies are there in this market ? Where are we over-satisfying our customers ? Where do we lose potential customers ? What are the top-line factors affecting the issue at hand ?
2. Key Visual Tools Pareto Curve (“80:20”) Value Map (“Mekko”) Value Chain Weighted Column Maker Share Gain Line KPC Combs Harvey Balls Quotes Typical Process Lollipops Channel Maps Conceptual Maps Forces at Work Crocodile Chart Parfait • • • • • • • • • • • • • •
What matters ? What are the various segments and sub-segments ?. Where do our core competencies reside ? Which are the most profitable competitors in absolute terms ? Who has been gaining market share ? What are the main purchasing criteria, and how do we fare ? What is the best way way to display “soft” quantitative data ? What is the best way to display qualitative data ? icks &creative Tips data displays of my own ? Can I come upTrwith What is the relative importance of various routes to market ? How can I simply structure a set of data before analysing it ? What are the forces at work ? Can I come up with creative data displays of my own too ? What is the easiest way to display market growth ?
3. Analysis Techniques • • • • • • • • •
Logic Tree Process Maps Sanity Checks Top-down Estimation One-Off Interviews Interview Programmes Focus Groups Clustering Factorial Analysis
How does one organise the research ? What are the key steps ? Does this number make sense ? How can I estimate a sub-set sub-set ? What are the main issues with one-off interviews ? What are the main issues with interview programmes ? What are the main issues with organising focus groups ? How can I group people by preferences ? How can I group people’s preferences in broader sets ?
4. Sources • •
Desk Sources Usual AABC Sources
Where can I find secondary data ? What are the usual AABC sources ?
2
Table of Contents
Tools Book Key Question
Tool Description
Typical Output Format
1. Strategy Concepts
Typical Process
Tricks & Tips
3
Porter’s 5 Forces
Tools Book Key Question
What are the key market drivers ? Tool Description
Porter’s model is a great check list when starting to describe an industry. It helps you organise your research and, more importantly, importantly, provides a good framework framework in which to present your findings. Typical Output Format Porter’s 5 forces map
Threat of New Entrants
Bargaining power of Suppliers
Rivalry among Competitors
Bargaining power of Customers
Threat of Substitution
Typical Process
Tricks & Tips
• Start Start with with top-li top-line ne list list of the the main main players in each of the 5 boxes - key competitor competitorss - key suppliers suppliers - key customers customers - most likely likely entrants entrants - most likely likely substitution substitution •
Carr Carry y out out lit. lit. sear search ch on the name namess compiled above
•
Gather Gather all releva relevant nt arti article cless in in one one bind binder er
•
“lif “lift” t” most most com commo mon n comm commen ents ts int into oa word document
•
Orga Organi nize ze resu result ltss int into o 5 boxe boxess
•
Summarize
Market-I
• Keep Keep a goo good d bala balanc ncee betw betwee een n words and data • “For “Force ce qua quant ntifi ifica cati tion on”” of wor words ds.. For example if you read “customer power is likely to continue growing at broadly the same rate” rate” try and find relevant data and write “customer power is likely to continue growing, keeping price hikes on historical level of 2%p.a.” • Think Think abou aboutt Newt Newton on and and Archimedes (every proactive force creates its own reactive force) and try capturing both the net effect and the 2 opposing forces • Ground Ground model model in the the analyt analytica icall process: what further research would be required to conclude on a given trend?
4
SWOT
Tools Book Key Question
What are the top-line factors affecting the issue at hand ? Tool Description
The SWOT box is a less formal , more intuitive version of Porter’s model . It provides a good framework in which to lay out initial findings - and can be used to start a Logic Tree (see below)
Typical Output Format SWOT analysis of Chelsea FC
Strengths
Opportunities
• Experienced players • Key real estate location • “Cultured” football methods shared across the squad • Midfield play
•
• Attract further key international names • Umbrella branding (internationally recognised name)
•
Typical Process
• • •
• •
Little development opportunities for the youth teams Old players more accident prone Low scoring Inexperienced manager Still backed by private investors MUFC and Arsenal keep getting bigger Leeds, Liverpool, etc back in contention for N° 3 spot
Weaknesses
Threats
Tricks & Tips
• Self-explanatory
• Many issues are both an opportunity and a threat, or a strength and a weakness ! • Do not sit on the fence; you will revisit this analysis in due course but start by choosing a box for every phenomenon observed / anticipated
Competitor-I
5
Growth Share (BCG) Matrix
Tools Book Key Question
Which industries or sectors should we be in? Tool Description
The Share-Growth matrix is a simple matrix devised to visualise multiple investment alternatives. It helps you compares relative attractiveness of different shares vs. growth situations. It requires an Excel sheet and the Bubble function in the Chart Menu. Typical Output Format Fish food service segments, Share & growth, 1998 Segment FORECAST growth (%)
Low 12,0%
High
Services
10,0%
High
Hotels 8,0% Pubs
Restaurants
6,0% 4,0%
Fish & Chips
Health Catering
2,0%
Low
0,0% Clubs
Education
-2,0% 0
10
20
30
40
50
60%
Segment share (% of market)
Typical Process
Tricks & Tips
• Compile competitor or division market shares in their relevant market
• Each quadrant of the matrix corresponds to a generic term, describing the strategic opportunities available to a company or division present there: - bottom-left quadrant, “dog”, small market share of a below par growth industry, little opportunity - top-left , “question mark”, small market share of an above par growth industry, good opportunity - top-right, “star”, big market share of above par growth industry, great opportunity - bottom-right, “cow”, big market share at below par growth, superb position but little opportunity , time to milk !
• Gather market growth rates for each industry or segment • Plot • Split chart in 4 quadrants : Dog; Question Mark; Star; and Milk • Do not worry exceedingly about accuracy of positioning of a given data point in the matrix but focus rather on relative positioning
• Original BCG had the x-axis with RMS (relative market share) and was plotted with leaders on the left rather than the right
Market-I
6
Product Life Cycle
Tools Book Key Question
What stage of the cycle are we at ? Tool Description
The Product Life Cycle is a concept devised to keep in mind that relative competitive advantages can ebb and flow. It helps you position competitors or industries against one another and against a time line. Works better on a theoretical level rather than with lots of data Typical Output Format Product life cycle curve
Measure or phenomenon (ie sales in £, penetration in %, etc)
Time line of phenomenon (in years, or months, or days, etc)
Typical Process
Tricks & Tips
• Find start date of timeline (i.e. first recorded mention of the technology, the product, the industry, etc)
• Tip : it is usually quicker - and as illustrative to plot 3 or 4 milestones than it is plotting the whole line
• Define a vertical axis (i.e. sales, penetration levels, etc) • Create Excel table with measurement records over time • Plot
Market-II
7
EOS : ROS - RMS
Tools Book Key Question
Does size matter ? Tool Description
The ROS-RMS is a simple chart displaying together relative size of competitors and profitability levels. It requires an Excel table and the Bubble function in the Chart Menu. The parallel band lines need be added afterwards Typical Output Format ROS - RMS Industry A
ROS - RMS Industry B
25% ROS 20 (last 3 years, 15 %) 10
ROS (last 3 years, %)
5 0 -5
25% 20 15 10 5
0,01
0,10
1,00
10,00
0 -5
0,01
-10
-10
-15
-15
-20
-20
-25
RMS (log)
Typical Process
• Define list of competitors to be included, and order annual reports • Create Excel table with last year’s sales + last 3 years’ profits • Plot • Annotate competitors with incomplete and / or misguiding data, such as less than 3 years of profits; latest year of sales is not last year; finer data available includes other divisions, not just sales and profits of relevant segments, etc
-25
1,00
10,00
RMS (log)
Tricks & Tips
• Order annual reports in advance : 2 weeks lead-time for foreign companies • Don’t forget 3 years average for ROS • Start plotting data you have : partial is better than nothing • Do not hesitate to call competitor directly - after clearance from case team manager - to gain idea of segment profitability (ie “higher than average of company”, etc)
• Strip down competitor sales to relevant segments • Estimate comparative profitability of remaining segments
Competitor-I & II
0,10
8
EOS : Learning Curve
Tools Book Key Question
What are the benefits of scale or experience ? Tool Description
The Learning Curve is a very simple concept: it is the trend line of a set of data where the Y axis represents the cost/unit of the X axis. It requires an Excel table and the Scatter function in Chart Menu Typical Output Format Manufacturing scale curve, Biscuits industry, 1997 1,2
Cost per Tonne (excl. materials, 1,0 1993, £k/Tonne) 0,8 0,6
A 10 times increase in scale represents a theoretical cost improvement of £275p.T. Some of this, however, would be offset by increased distribution and other costs
Broxburn Linkoping Hatton Durango Lauragais
Ashby X1
Gyor Carlisle
Maastricht
Dortmund
0,4 Jyvas Hyva 0,2 0,0 1 000
10 000
X2
R2 = 65%
Genoa
Tyneside
100 000
Manufacturing Capacity (kTonnes, 1996, Log scale) Typical Process
Tricks & Tips
• Define the exact value chain dimension whose learning curve you want to display (manufacturing, purchasing, etc) •
• Careful! There are 2 versions of the “learning curve”. One is really about learning, the other really about scale • In the first case, the X axis represents the cumulative N° of units produced to date and the Y axis the cost/unit over time (A typical example is plane production. Few units produced per year, and the cost of each unit keeps coming down as assemblers get better at it with experience)
Choose the number of data points you need (5 will result in rather illustrative outcome, 10 - 12 will give more confidence)
• Gather data on cost/unit from client for relevant objects (eg: get exact capacity and cost/tonne for client in 4 plants)
• In the second case, the X axis represents the size of a given process (eg. plant capacity) and the Y axis the cost/unit for this process in various plants.
• Interview competitors’ relevant director for missing data using smart approach (eg: by how much would your cost/unit drop if you doubled size
• Tip: manufacturing directors are not often approached for market research and thus tend to talk a lot
Competitor-II
9
ROS / RONA
Tools Book Key Question
How well are we using our assets ? Tool Description
The ROS-RONA is a simple chart displaying together asset turn and profitability of a set of competitors. It requires an Excel table and the Bubble function in the Chart Menu. The RONA iso-curves need be added afterwards Typical Output Format Major retailers ROS-RONA, UK, 1993-95
20% ROS (last 3 years, %)
1,8% 16% 14%
Body Shop M&S
12%
RONA (%)
Boots
10% 8%
WH Smith
6% 4%
House of Fraser
2% 0%
0
1
2
70% 50% 30% 10%
Kingfisher John Menzies
3
4
Typical Process
5 6 Asset Turn
7
8
9
10
Tricks & Tips
• Check with manager the definition of net assets that will be used • Define list of competitors to be included, and order annual reports • Create Excel table with last year’s sales + last 3 years’ profits + last 3 years’ net assets • Plot, and create RONA iso-curves
• Order annual reports in advance : 2 weeks lead-time for foreign companies
• Annotate competitors with incomplete and / or misguiding data, such as different definition of net assets, less than 3 years of profits; finer data available includes other divisions, not just sales and profits of relevant segments, etc • Strip down competitor sales to relevant segments • Estimate comparative profitability of remaining segments
• Let manager know early if agreed definition of net assets creates data problem - there are many possible definitions and another one might make things easier for you without affecting usefulness of information
Competitor-I & II
• Don’t forget 3 years average for ROS • Start plotting data you have : partial is better than nothing
• Do not hesitate to call competitor directly - after clearance from case team manager - to gain idea of segment profitability (ie “higher than average of company”, etc)
10
Value Disciplines
Tools Book Key Question
What types of possible strategies are there in this market ? Tool Description
The Value Discipline model helps you position competitors or industries against one another according to 3 possible main generic strategies. Works better on a theoretical level rather than with lots of data Typical Output Format Value disciplines, Car industry, UK, 1999
Product Leadership BMW Rolls Lada Vauxhall
Daewoo
Operational Excellence Typical Process
Customer Intimacy Tricks & Tips
• If without data :
• If using data :
• Define list of competitors to be included
• Create Excel table with score of each competitor against each of the 3 dimensions
• Position each competitor on map based on available evidence (lit.search, annual reports, brokers’ reports, gutfeel, etc)
Competitor-I
• Plot data using the Radar function in the Chart Menu
11
Customer “Happy Line”
Tools Book Key Question
Where are we over-satisfying our customers ? Tool Description
The Customer “Happy Line” is a complex, but powerful concept that helps illustrate the areas where we under-deliver on customers’ expectations and those where we over-deliver. It requires an Excel table and the Scatter function in the Chart Menu. The Line needs be added afterwards Typical Output Format « Happy Line », Company X, Industy Y, 1999
10 Good
Our score against each KPC (1 to 10) Average
9 8 7 6
Bad
5 4 9
8,5
8
Critical
7,5
7
6,5
Important
6
5,5
5
4,5
Secondary
KPC Score (i.e. relative importance of criteria to Customers) Typical Process
• Create Excel table with outcome of customer interview programme : average score given to each criteria and score of each competitor against the criteria • Plot client score, with X axis as criteria score • Set X axis so that Min and Max equal the scores of respectively the lowest and highest ranking customer purchasing criteria
Tricks & Tips
• Tip : set Y axis by choosing a relevant set of competitors for which to obtain lowest and highest scores (you do not want to benchmark the client against nohopers in the industry) • Happy Line goes from Top left of chart to the middle, and then horizontally to the right
• Reverse X axis so that Max is to the left • Set Y axis so that Min and Max equal respectively the lowest and highest scores obtained by one of the relevant competitors against any criteria • Add Happy line
Customer-I
12
Conversion Waterfall
Tools Book Key Question
Where do we lose potential customers ? Tool Description
The Conversion Waterfall is a concept that helps you summarise the findings of a customer interview programme along 6 major measures. It requires an Excel table and the Bar function in the Chart Menu. Can work as well on a preliminary, data-less level Typical Output Format Conversion waterfall, Retailer X, France & Germany, 1997
% of population
100% 90%
France Germany
80% 70% 60% 50% 40% 30% 20% 10% 0% Awareness Knowledge
Typical Process
• Create Excel table with outcome of customer interview programme : share of the sample population that meets the successive measures
Liking
Preference
Tricks & Tips
• Self-explanatory
• Plot
Customer-I
Visit
13
Purchase
Table of Contents
Tools Book Key Question
Tool Description
Typical Output Format
2. Key Visual Tools
Typical Process
Tricks & Tips
14
Pareto Curve (“80:20”)
Tools Book Key Question
What matters ? Tool Description
The Pareto Curve is a simple and powerful concept : it posits that quite regularly 20% of a given population will account for 80% of this population’s actions with a third party (ie sales, purchases, etc). It requires an Excel table and the Scatter function in the Chart Menu* Typical Output Format Company X, Customer Pareto, 1998
100% Cumulative % of Sales
80% Pareto Rule
60%
20% of a given population accounts for 80% of this population’s actions (eg. 20% of a customer base account for 80% of sales)
40% 20% 0% 0
10
20
30 40 50 60 70 Customers (Ranked by decreasing Sales)
Typical Process
• Let’s assume we are drawing a Pareto of a client’s customers • Gather client total sales, and sales by customer • Rank customers in column by decreasing sales • Calculate weight of each customer as a % of total client’s sales • Set-up column with cumulative share of customer sales • Plot scatter of cumulative sales column
Customer-I & II
80
90
100
Tricks & Tips
• * : instead of using the scatter function you can use the Bar function, Select Data series, and set Options / Gap to “0” • Tip : do not despair if you only have sales of the Top 50 customers (or of the Top 50 suppliers) . Find out how many customers (or suppliers) make up the total list and posit that each of the smaller ones has the same - tiny - weight. You can now draw a pseudo-Pareto where the top right hand part of the curve will effectively be a nearflat straight line
15
Value Maps (“Mekko”)
Tools Book Key Question
What are the various segments and sub-segments ? Tool Description
A Mekko is a visually simple but analytically complex tool that helps you display in one sole slide 2 successive cuts at a data set (eg market size by product category and by product, industry sales by sector and by competitor, etc). It requires a special input tool; the “Mekko Maker” Typical Output Format General insurance market, UK, 1998
100% Personal vs commercial split for sector
Property £8.2bn
Accident Pecuniary & Health Loss £2.7bn £2.3bn
Motor £6.8bn
Liability £ 1.9bn
Domestic Mortgage
Private Non-Comprehensive
Other
Extended Public/ Warranty General
80
Liability
60
Private Comprehensive
Personal Property
= PI
Personal Payment Protection
40 20
Total Market = £21.9bn
Commercial Property
Commercial Vehicles
Corp orate
0
20
40
Employers Liability
Consequential Loss
Fleet
0
Other
60
80
100%
Sector as per cent of UK market
Typical Process
• Create Excel table formatted as follows : - one segment by column - rank segments by decreasing size - each sub-segment of a given segment in the successive rows of the relevant column - rank sub-segments by decreasing size
Tricks & Tips
• Tip : use only absolute measure of size (ie £), not market shares or other relative measures / ratios • For clarity of reading segments of size inferior to 5% of total market size should be grouped into an “other” segment
• Open “Mekko Maker” • Copy Excel table into “Mekko Maker”
• Likewise, all sub-segments of size inferior to 3% of total size of a given segment should be grouped into an “other” sub-segment • Use shading & patterns
Market-I & II
16
Value Chain
Tools Book Key Question
Where do our core competencies reside ? Tool Description
The Value Chain model helps you highlight for different competitors within a given industries the competencies each one has developed to succeed. Works better on a theoretical level rather than with lots of data Typical Output Format Value chain
Finance HR Technology Logistics
Purchasing
Manuf’ing
Typical Process
Marketing
Sales
Tricks & Tips
• Identify the main competencies in the industry
• Do not hesitate to alter the arrow above
• Create generic industry arrow with support functions and operating functions
• Only principle to follow for visual : - horizontal division separate “support” functions from operating ones - key operating processes are split vertically
• Gather from desk research (annual reports, articles, etc) and one-off interviews the area of the value chain where each competitor chooses to focus (“core competencies”)
• This principle will justify treating different functions differently by industry (e.g. technology is a support competency in most industries, but a key operating one for a software company) • Tip : an organisation chart is often a good starting point to understand emphasis attached to the competency
Competitor-I
17
Weighted Column Maker
Tools Book Key Question
Which are the most profitable competitors in absolute terms ? Tool Description
A Weighted Column chart is a simple tool that helps you display at once both competitor profitability level and overall industry profits. It requires a special input tool; the “Weighted Column Maker” Typical Output Format Industry profitability, Industry X, 1995-98
20% ROS (last 3 15 years,%)
Weighted Average = 10.0%
10 5
Comp 1
Comp 2
Etc
0 0
20
40
60
80
100%
-5 -10% Share of Market, by Competitor
Typical Process
• Define list of competitors to be included, and order annual reports • Create Excel table with last year’s sales + last 3 years’ profits • Plot • Annotate competitors with incomplete and / or misguiding data, such as less than 3 years of profits; latest year of sales is not last year; finer data available includes other divisions, not just sales and profits of relevant segments, etc • Strip down competitor sales to relevant segments • Estimate comparative profitability of remaining segments
Competitor-I & II
Tricks & Tips
• Order annual reports in advance : 2 weeks lead-time for foreign companies • Don’t forget 3 years average for ROS • Start plotting data you have : partial is better than nothing • Do not hesitate to call competitor directly - after clearance from case team manager - to gain idea of segment profitability (ie “higher than average of company”, etc) • Plot Weighted Average industry profitability line based on information you have, but indicate what total % of industry sales you have profitability data for
18
Share Gain Line
Tools Book Key Question
Who has been gaining market share ? Tool Description
The Share Gain Line is a visually simple tool - if a bit difficult to grasp at first - that helps you display relative share performance of competitors in the market context. It requires an Excel table and the Bubble function in the Chart Menu. The “Zero-gain” line needs be added afterwards Typical Output Format Share movements by competitor, Music, Cards & Video retail, UK, 19 95-1999
Competitors Growth (% over 4 years)
60%
Competitors gaining share
50%
Blockbuster Paperchase
40%
WH Smith Disney Store
30% 20%
Disney Store Boots WH Smith Woolworths J. Menzies
Our Price
10%
Woolworths J. Menzies
Competitors losing share
0% -10%
0% -10%
10% WH Smith
Music
20%
30%
Cards
40%
50%
60%
Video
Total Market Growth Typical Process
• Gather historical market data and compile market growth for each market • Define list of competitors to be included by market • Gather historical competitors sales data and compile competitors sales growth
Tricks & Tips
• Always write meaning of bisecting line on the chart: above line = gaining share below line = losing share • Tip : for multi-segments on one graph all the bubbles in a given market should be on the same vertical line
• Create excel table with market size Year Zero (Yo), market size now (Yn), competitors sales Yo and competitor sales Yn • Plot • Draw line bisecting the two axes
Competitor-I & II
19
KPC Combs
Tools Book Key Question
What are the main purchasing criteria, and how do we fare ? Tool Description
The KPC Combs is a visually simple tool - if a bit difficult to grasp at first - that helps you display jointly both the importance of customer purchasing criteria and the ranking of competitors against those criteria. It requires an Excel table and the Bar + Line function in the Chart Menu Typical Output Format Key purchase criteria (KPC) comb, Industry X, Germany, 1999
10 Score (1 to 10)
9 8 7
Criteria Comp 1 Comp 2 Comp 3 Client
6 5 4 3 2 1
Criteria 1
Criteria 2
Criteria 3
Criteria 4
Criteria 5
Criteria 6
Key Purchasing Criteria Typical Process
Tricks & Tips
• Create Excel table with outcome of customer interview programme : average score given to each criteria and score of each competitor against the criteria
• You can plot both the Criteria and competitors score using the same scale : draw chart as a Line chart, select the Criteria line; go into Chart Options and select a Bar
• Plot
• Do not forget to set minimum and maximum of Y axis to values of the range offered (eg from 1 to 10, not 0 to 10) • Tip : for a large interview programme, the dispersion of the average scores will be minimal. You can increase visibility of differences by setting Y values more finely on either side of the average (eg display from 3 to 8, or 4 to 7, instead of from 1 to 10)
Customer-I & II
20
Harvey Balls
Tools Book Key Question
What is the best way to display “soft” quantitative data ? Tool Description
Harvey Balls are a technique that helps you summarise the results of a benchmarking exercise where the item benchmarked obey different metrics. It needs to be hand drawn. Can work as well on a preliminary, data-less level Typical Output Format Competitor benchmark summary table, Mixed retailers, UK, 1999 Competitors KPI
WH Smith
John Menzies
Boots
Woolworths
etc
Supply chain efficiency Customer retention level Staff utilisation level Etc ...
Typical Process
• Select dimensions to be included in table • Write down a code for each dimension (ie what score justifies a full ball, what deserve half, what is a quarter, what is zero)
Tricks & Tips
• Tip : Write code before filling the table, as filling a Harvey Ball table usually throws open at least one implication that wasn’t foreseen with just the base data
• If wanting a “Total” row, give a weighting to each dimension and calculate total for each column • Hand-draw
Competitor-I
21
Quotes
Tools Book Key Question
What is the best way to display qualitative data ? Tool Description
Quotes are a systematically under-used technique for displaying qualitative data in a powerful way.
Typical Output Format
“ There has bee a trend away from sport shoes and towards “work” boots worn as streetwear. This has been going on at the rate of 2-3% p.a. over the last 3 years, is getting stronger now, and is no longer just a teenage fashion” - J. Swartz, CEO, Timberland “Consumer shift towards good value and cheaper shoes, together with the availability of imports from the Far East and economically weaker countries have intensified price pressure and competition in the market” - Euromonitor “Camel is the best selling boot in Germany, with in excess of 5% market share” - Marketing Director, Camel UK
Typical Process
• Create on an on-going basis a database / binder of comments with good quote value • Write comprehensive notes during oneoff interviews • Manage a good filing system when carrying outlit. search
Tricks & Tips
• Format all quotes in a consistent manner, preferably as above • Try and get a number in every other quote, ie : – select the quotes on the basis of their number content (ex-post) – force interviewee to quantify qualitative statements (ex-ante)
• Quote value rests in the nature of comment heard or read and / or in the authority of the source
Customer-I
22
Lollipops
Tools Book Key Question
Can I come up with creative data displays of my own ? Tool Description
Lollipops are a technique that helps you display in one sole slide a large number of detailed information across a large selection of objects (competitors, products, etc). It requires an Excel table and the Bubble function in the Chart Menu Typical Output Format Plant sizes & total capacity by competitor, Storage products industry, Europe, 1998 70k Plant size (k tonnes),
E
60 50
D
40 30 20 10 0 Total Capacity (k tonnes) Competitor
60k tonnes
UK
D
UK B E NL
D S N SF UK NL D
UK
B
D D
l o r e r n x o n t a c o x i t e r n a w e x t a r o s a e r e c a l e t o p i s d m u s i t l T i a p l u d c e r a l c a e f e x i o c a y r u t l M m S A a g v r e e h y r e a b G h o e F P n s t S c D M N E s M r a A M A T C o
Typical Process
• Create Excel table with all the data required : one object / criteria per row (eg competitor) and one measure per box vertically (eg plant capacity of competitors plant) • Plot
Tricks & Tips
• Tip1 : always try creative solution on paper before launching into big Excel manipulation • Tip2 : remember that a nicely laid out Excel table is always a more powerful visual than a cumbersome graph ...
• A Lollipop chart is simply a creative answer to a regular problem : how to display a large number of data points, the detail of which is necessary but less important that the overall picture it paints • A “Mekko” is another creative way to solve the problem described above • Any solution you may come up with that solves your display issue has got to be a good one !
N/A
23
Channel Maps
Tools Book Key Question
What is the relative importance of various routes to market ? Tool Description
Channel Maps are a technique that helps you display the complexity of various routes to market whilst quantifying their relative importance. There isn’t a software available to generate them : they need to be hand drawn Typical Output Format Channel map, Construction materials, UK, 1999 Products
£15.3bn Channels
Additional Products*
Construction Materials Manufacturers/Importers
£22.3bn
Distributors/Stockists
Direct £7.0bn
£4.4bn Specialist Merchants/ Factors Wholesalers
£4.6bn
Builders Merchants
Builders/Contractors
• Gather on one page all the data you have about the industry sales (manufacturer sales, retail sales, wholesale sales), and organise data into 3 tiers : products (or manufacturers), channels, and endcustomers (or retailers) • Firstly, draw the product and endcustomer horizontal bars • Secondly, draw the “direct” flow between these 2 bars (on the left of the above map), and infer from this a MSP to RSP ratio
£0.5bn
£7.3bn
Tricks & Tips
• Be sure to use the same units throughout, £100 of manufacturer sales can be exactly the same as £200 of retailer sales if RSP (retail selling price) = 2xMSP (manufacturer selling price) in thi industry .. • Tip : when possible volumes are a better measure than values (as stay the same throughout the flow normally…)
• Thirdly, split this ratio in two to “body double” the MSP to WSP and WSP to RSP ratios, and “convert” your whole data set in the relevant unit you are using • Finally, draw a first map, and iterate...
Market-I
£7.6bn
Home Customers
£18.2bn
Typical Process
DIY Stores, Superstores and Hardware shops, Garden Centres etc
£6.3bn £0.8bn
EndCustomers
£3.2bn
24
Conceptual Maps
Tools Book Key Question
How can I simply structure a set of data before analysing it ? Tool Description
Conceptual Maps are basic techniques that help you breakdown complex issues into sub-sets so as to better either lay out the thinking, plan the research or map the results. They do not require intricate softwares; just a ruler, a pencil, and an eraser ... Typical Output Format Key conceptual maps “2x2” Matrix
Conceptual Cross
Dimension 1
Dimension 3
High
Dimension 4
Low
Low
High
Dimension 2 Typical Process
• Conceptual Maps come in various shapes. The mother of all conceptual is the “2 x 2” Matrix, where various objects are plotted in 4 boxes : dimensions 1 & 2 with sub-sets “low” and “high” • A Conceptual Cross provides another way to think about how objects can be classified against 2 other dimensions
Customer-I
Tricks & Tips
• The 2x2 Matrix is the most useful consulting tool ever … • The Conceptual Cross is marginally less helpful. It can come handy when you find that the objects you have plotted in a matrix all fall on the diagonal line : a 45° rotation of Matrix dimensions gives you a Cross map
25
Forces at Work
Tools Book Key Question
What are the forces at work ? Tool Description
The Forces at Work visual is simple version of Porter’s model . It highlights complexity rather than simplify / organise it in the way Porter and the SWOT box do.
Typical Output Format
Convergence Convergencein in national nationaltastes tastes across acrossEurope Europe
Clothing Clothingmanufacturing manufacturing shifting shiftingproduction production overseas overseas
Customers Customersdemanding demanding shorter shorterlead leadtimes times
Improvements Improvementsin inthe the speed and cost speed and costof of logistics logisticsacross acrossEurope Europe
Threat Threatof of cheap cheapimports imports from fromAsia Asia Growth Growthof of international international customers, customers,e.g. e.g. Nike Nike& &Benetton Benetton
Typical Process
• Self-explanatory
Clothing Clothingmanufacturers manufacturers shifting shiftingproduction productionfrom from West Westto toEast Eastand andSouth South Europe Europe
Price Priceconvergence convergence between betweenmarkets markets driven drivenby bythe theEuro Euro
Shortening Shorteningproduct productlife life cycles in cycles infashion fashion
Tricks & Tips
• Do it either at the beginning of a case - or at the end to illustrate complexity of forces at work • During case work, try and organise these forces either in Porter’s 5 Forces model, or in SWOT box
N/A
26
Crocodile Chart
Tools Book Key Question
Can I come up with creative data displays of my own too ? Tool Description
Crocodile Charts are a technique that helps you display in one sole slide how the issue you are studying narrows down to your specific area of focus. It requires an Excel table and the Bubble function in the Chart Menu Typical Output Format Crocodile chart, Oil reserves, 1995 100% 90%
Sea
80%
North
70% 60% 50%
Land
UK
40%
South
30%
Europe
20% 10% 0%
Global
Typical Process
• Create Excel table with all the data required • Plot 4 bars
Continental
National
Tricks & Tips
• Tip : it is easier to add lines across all bars and delete unnecessary one upon finalising the chart than the other way round
• Add connecting lines
N/A
Regional
27
Parfait
Tools Book Key Question
What is the easiest way to display market growth ? Tool Description
A Parfait is a visually simple tool that helps you display jointly both the relative size of market segments and their relative growth. It requires an Excel table and the Area function in the Chart Menu Typical Output Format UK Clothing Market 1994-98 3 000 Annual Growth 2 500
Market size (1998, £m)
3%
Childrenswear 2 000
10%
Sportswear
1 500 Menswear
2%
Womenswear
2%
1 000
500
1994
1995
Typical Process
• Create Excel table with all the data required
1996
1998
Tricks & Tips
• Do not forget to add growth data to the right of the chart
• Plot
N/A
1997
28
Table of Contents
Tools Book Key Question
Tool Description
Typical Output Format
3. Analysis Techniques
Typical Process
Tricks & Tips
29
Logic Tree
Tools Book Key Question
How does one organise the research ? Tool Description
One half of Structured Thinking, the Logic Tree turns the initial client brief into a set of discrete questions. Each Logic Tree has a “twin” in the shape of the Research Tree. Once the key questions have been established it is usually clear which research method is the most appropriate Typical Output Format Logic Tree
Research Tree I gather data to inform the applicability of each option
I can find a solution to my sockless state
I can find another sock in the flat
I can trace the other sock
I can trace another pair
I can get an entirely new pair
I can buy a new pair in time
I can have one sent
I can compensate for what I don’t have
I can alter clothing to hide sockless ness
I can justify sockless ness
Typical Process
• Write “top of tree” as a positive statement answering the client question • Create sub-branches by splitting the top of the tree into its natural components (writing a good Logic Tree is an art and a technique that one acquires with experience) • The sub-branches should obey the MECE method (Mutually Exclusive Completely Exhaustive), i.e.all the sub-branches taken together must exhaustively support the branch above, and each subbranch must deal with something different from the other sub-branches • The MECE methods implies that - if one of the sub-branches is false then the top of the tree must be false - if all the sub-branches are true then the top of the tree must be true
N/A
Research inside the flat
Check my room (under the bed, other drawers)
Research outside the flat
Inquire about socklessness
Ask Check Check flatmates there are who is Check sock coming laundry shops Call to room on the get them way to bring one
Check with partner that socks are needed
Ask for a good joke Think of how to turn it into a statement
Tricks & Tips
• A positive statement is more difficult than a negative statement, and thus gives client more confidence on strength of our answer • Tip : some first level “split” of the logic tree occur rather regularly. Typical first level branches are : - Sales, Margins and Costs - Customers, Competitors and Cost • Tip : do the research tree only once finished with logic tree (on 80:20 basis). Indeed some research branches can address several branch of the logic tree at once
30
Process Maps
Tools Book Key Question
What are the key steps ? Tool Description
Process Maps are a technique that helps you group a set of research and/or action points into broad sequential steps
Typical Output Format
Bidding, Negotiation & Closing
Deal Origination
• •
Target identification Assessment of potential postdeal synergies
•
• •
Typical Process
• Self-explanatory
N/A
Market Study / strategic due diligence - External validation - Ammunition for negotiations - ‘Black holes’ check Resolve competition and regulatory issues Valuation analysis
Exit (Trade Sale or Flotation)
Owning
• •
Strategy development Post-acquisition disputes
• • •
Business plan preparation Market study for potential prospectus ‘Outside in’ view for potential buyers’ - Synergy valuation for negotiations
Tricks & Tips
• Try to aim for 4 to 7 steps to keep the process map useful
31
Sanity Checks
Tools Book Key Question
Does this number make sense ? Tool Description
Sanity Checks are basic techniques that help you assess the viability of a data point obtained through desk research or one-off interviews. This do not require anything else than careful consideration ...
Typical Output Format
Main data sanity checks Confidence Ranging Narrowing a variety of N° from different sources down to a range within which you are fairly confident the true answer is to be found (“confidence interval”)
Feel Right Test Connecting data obtained on esoteric issues to more easily understandable dimensions and assessing credibility of implication
Typical Process
• See above & Generic Research
Impact Criticality Test
Materiality Test Putting in perspective the differences between the various estimates obtained typically by expressing biggest delta as % of average value
Establish whether the different estimates obtained lead to the same outcome with regard to the question you are really addressing
Body Doubling Choosing one unique data point as the estimate you will go with pending further information
Tricks & Tips
• Most important test is materiality test, i.e. how different are the various estimates obtained ? • Tip : choose a different color - red for examples - for the body doubles in an Excel worksheet
Generic Research
32
Top - Down Estimation
Tools Book Key Question
How can I estimate a sub-set ? Tool Description
The Top-Down estimation is a method that consists in shaving off parts of something you know (eg. market size) until you are left with the rump (eg. a segment size) that you were trying to estimate in the first place Typical Output Format
Own-label, non-ready meals, frozen fish market top down est imate,1998 Market value (£m, RSP, 1998)
2 000 1 800
Food service
1 600
Canned
1 400 1 200
Fresh
1 000 800
Ready Meals
Chilled
600
Branded
Retail
400
Coated
Frozen
200
Natural
0 Total Fish Market
Share of Retail
Share of nonReady Meals
Share of Frozen
Typical Process
Own Label
Share of Own Label
Our Market
Tricks & Tips
• Gather all the available data on the market you try to estimate as well as data on broader market sets
• The outcome of a top-down estimation is visually complex to illustrate
• Make a list of the “cuts” for which you have data (eg. you have data by channel and by products but not by channel & product)
• The best method is to create a set of 100% vertical bars and one by one turn to “no border & no area” the boxes that you don’t want to see appearing
• Split the starting object into sub-set by applying the cut for which you have least data (eg. you know market share of products in retail but not in wholesale, thus share by channel before you share by product)
Generic Research
• Think Rubik’s cube: you will probably have to try 2 or 3 different sequencing of the cuts before you arrive at a satisfactory outcome
33
One-Off Interviews
Tools Book Key Question
What are the main issues with one-off interviews ? Tool Description
Check-list for One-Off Interviews. Use framework below before proceeding with interviews
Typical Output Format One-off Interviews Check-list
Preparation
Anticipation of Answers
Contact & Persuasion
• Be clear exactly • Define your • Be polite, • especially to which questions objectives for need to be the meeting up secretaries answered front • • Be rigorous in • Choose the • Ensure you using secretaries appropriate come out with where possible method the specific answers you • Write a • • Contact the require professional but correct people deferential fax
Execution
Prioritise questions
Interpretation of Results • Be flexible but focused on the answer
Be prepared for obstructive remarks from interviewees
Wrapping Up
• Be prompt in delivering high quality feedback • Remember who has paid the bill!
Write clear and comprehensive meeting notes
• Do not lie !!
Typical Process
• See above & Generic Research module
Tricks & Tips
• Go through the check-list again & again ! • Tip : always keep in mind the ANSWERS you are expecting and particularly the different SO WHAT that different answers will point to
Generic Research
34
Interview Programmes
Tools Book Key Question
What are the main issues with interview programmes ? Tool Description
Check-list for Interview Programmes. Use framework below before proceeding with interviews
Typical Output Format
Interview Programme Check-List Preparation
• Choose the appropriate method - on the phone - in the street - mix • Select the best sample - suitable proportion of market - hit rate
Anticipation of Answers
• Create a questionnaire which is focused on testing the hypothesis - correct questions - questions asked in the correct way
Contact & Persuasion
• Ensure interviewers are fully briefed
Execution
• Ensure process is managed effectively
• Set-up database • Pilot and maintain it questionnaire throughout and re-draft ineffective • Temp performance questions monitored and data inputed as rapidly as possible
Typical Process
• See above & Customer-I module
Interpretation of Results
• Only appropriate analysis is carried out (80 : 20)
Wrapping Up
• Keep paper copies of questionnaires at hand !
Tricks & Tips
• Go through the check-list again & again ! • Tip : always keep in mind the ANSWERS you are expecting and particularly the different SO WHAT that different answers will point to
Customer-I
35
Focus Groups
Tools Book Key Question
What are the main issues with organising focus groups ? Tool Description
Check-list for Focus Group. Use framework below before proceeding with organising groups
Typical Output Format
• Organising the day • Managing the debate
Focussing debate
Logistics
Getting the right people
• Venue and travel arrangements • Agreeing the plan with the client • Maintaining records • What to take • (equipment, props, • etc...)
• Selecting candidates • Invitations • Hit rate • Incentives Typical Process
• See Tricks & Tips
Customer-I
Tricks & Tips
• If possible, pre-select to ensure representative cross section by asking key questions on the phone • Confirm details in writing (by fax for speed) • Invite up to double the candidates you want unless they are motivated to come - offer cash/other incentive as appropriate as sanctioned by the project manager Think through the timing • Draft issues for debate of key questions beforehand • Stick to the agenda • Use flip charts to document contributions • Keep travel to a minimum • Ensure venue is appropriately comfortable • Keep database as for interview programmes • Be prepared!
36
Clustering
Tools Book Key Question
How can I group people by preferences ? Tool Description
Clusturing is a complex technique that helps you group object (eg customers) according to their similarity of preference between a set of dimensions (eg customer purchasing criteria). It requires an Excel spreadsheet and a conjoint analysis software, such as SPSS Typical Output Format
Cluster analysis, Storage product industry, 1998 Speed of service
A lot
Cluster Y
Average
Cluster Little
X
Average
A lot
All-round support
Z
Average
Cluster
A lot Exclusivity of operations Typical Process
• Create Excel table with detail of customer interview programme : score given to each criteria by each customer (one customer per row, one criteria per column) • Import dataset into conjoint analysis software (SPSS or other) • Run Classify \ Cluster function in Statistics menu of conjoint analysis software • Write down N° of rows that fall in each cluster • Go back to Excel spreadsheet, create Cluster N° column to tag each row according to relevant cluster • Calculate average criteria score by cluster
Customer-I & II
Tricks & Tips
• Spend 3 or 4 hours reading the conjoint analysis software manual, and practicing on fictitious dataset ! • A Cluster analysis will start by giving you X clusters for your X customers (fair enough each one is different …). It will then start grouping the two customers with the most similar preferences into a single cluster, leaving you with X-1 clusters, etc, until all is one big cluster. The real N° of clusters in the population is usually obtained by trial and error, thus … • Tip : Never erase the content of a given Cluster N° column when tagging rows in Excel … a discarded solution usually turns out to be the correct one in the end !
37
Factorial Analysis
Tools Book Key Question
How can I group people’s preferences in broader sets ? Tool Description
Factorial Analysis is a complex technique that helps you group a large number of dimensions of choice (eg 14 customer purchasing criteria) into the relevant, fewer factors that underpin clusters’ preferences. It requires an Excel spreadsheet and a conjoint analysis software, such as SPSS Typical Output Format
Factorial analysis, Storage product industry, 1998 Key Purchasing Criteria
•1. •2. •3. •4. •5. •6. •7. •8. •9. •10. •11. •12. •13. •14.
Speedy quotations turnaround time Distributor brand Choice of payment terms Technical support & design assistance Speedy & reliable delivery Literature with distributor brand/name No competing direct sales force Complete territorial exclusivity Choice of pricing structures Full technical & design digest Free contract management facility Choice of stock holding policy Free & regular training programmes Dedicated account manager
Typical Process
• Create Excel table with detail of customer interview programme : score given to each criteria by each customer (one customer per row, one criteria per column) • Import dataset into conjoint analysis software (SPSS or other) • Run Data Reduction \Factor function in Statistics menu of conjoint analysis software • Check resulting plot, and iterate if need be
Factors Factor 1 Speed of service
Factor 3 All-round support Factor 2 Exclusivity of operations
Factor 1 = KPC 1 & 5 Factor 2 = KPC 2, 7 & 8 Factor 3 = KPC 3, 4, 6, 9, 10, 11, 12, 13 & 14
Tricks & Tips
• Spend 3 or 4 hours reading the conjoint analysis software manual, and practicing on fictitious dataset ! • A good dataset will always give you a reduction of your X criteria into X/3 or X/4 factors. Trying to fathom what these factors represent can be a notoriously difficult exercise. Ask for help • Tip : it is usually easier to understand a factor after having carried out a Cluster analysis (see previous page)
Training Modules Customer-I
38
Table of Contents
Tools Book Key Question
Tool Description
Typical Output Format
4. Sources
Typical Process
Tricks & Tips
39