Management science concepts for better decision making.
Product Portfolio
Analysis: BCG Share/Growth Matrix, GE/McKinsey
Array
PPA is used for assessing the competitiveness of businesses in
one company's portfolio. Certain companies look at product mix
decisions as portfolio decisions. Each product requires
investment and promises a certain return. The role of
management is to determine the products that comprise the
portfolio and the funds to allocate to them. In this sense, PPA
becomes useful twice a year, or more frequently if structural
changes take place.
Brand Management Tools:
Brand Switch, Profile Manager, Brand Mapping
Brand positioning, brand loyalty, and market share behavior
are issues of strategic relevance to marketers. The tools in
this collection tackle these issues. They may give managers a
competitive edge, and they will certainly prove useful in
stimulating strategic thinking within the team.
Forecast Manager
Forecasts are probabilistic statements about future events,
and there are many models to select from. Forecast Manager
works with time series for short-term projections. After
choosing the 'right' model it selects the 'right' method and
sets the 'correct' parameters for an optimized fit. The bulk of
the work is hidden behind a few mouse-clicks.
Decision Tree
Decision Tree helps in setting up frameworks for dealing with
decision problems involving risk and uncertainty. For instance,
your company calls for incremental funding and you are required
to estimate potential sales for several projects and to focus
on those projects that best satisfy the growth goal. There are
two aspects to be considered when approaching this
challenge:
a. the estimation of each project outcome
b. the selection of the most appealing project(s)
The former can be handled with simulation models. The latter
can be facilitated using decision trees.
Quality Analyst
Statistical quality control helps companies to increase their
ability to compete effectively by improving the quality of the
output they pour in the market. To do so, the characteristics
of a sample of products or one or more processes are measured
in order to make decisions regarding their quality. MM4XL
software makes available in a brilliant package all tools
needed to perform accurate, fast and visually effective
statistical quality control directly in MS Excel.
To build well-balanced product portfolios requires monitoring tools to look at the present financial and competitive status of the SBU's, in addition to keeping track of changes that might take place over time among products. In this framework the Boston Consulting Group (BCG) Share/Growth Matrix is a widely used tool. It is based on the common belief that success in business is strongly linked to cash flow (the cash flow of one product is shown as Sales minus Costs. From the financial point of view this computation can be extremely complex, yet the general meaning is simply the subtraction of costs from revenues), which is a function of market share and market growth. While the former generates cash, the latter uses it. The following diagram synthesizes this concept.
 |
Tip: It's easy to get up to speed with MM4XL in a matter of seconds:
|
It is very easy to run the Product Portfolio Analysis
with Marketing Manager. Select Portfolio Analysis from
the MM4XL menu, or click the first button from the left in the
floating toolbar, and the following window appears.
Input the number of products you want to analyze. If your
company markets 20 products, write 20 in the box beside
Number of products and a predefined report will be
automatically created. This window lets us set two important
options to enhance the readability of maps. The Max
Logarithmic Market Share = 1000
sets the horizontal axis of the grid at a maximum length of
1000 (equal to four times the size of the direct competitor).
All leader products with sales higher than four times that of
their direct competitor will be placed
at 1000. There is no point in displaying broader measures, and
in some cases, leaders being seven, eight or ten times larger
than competitors can adversely affect the whole map. The same
concept is applied to the vertical axis. Tick Max market
growth = 100% to set the maximum height at 100%. If you
have already stored the data in the correct format for
analysis, you could click Next and run the analysis
now.
Below is an example of the predefined input data range of
the Product Portfolio Analysis built in Marketing
Manager.
In column A you can input any sign or value, the function is
case sensitive. This is a very useful option when running
dynamic or comparative analysis. For instance, with code 1999
and code 2000 you can analyze the same portfolio at two
different moments in time. Alternatively, you can code products
belonging to different categories, like in our example below,
or to different departments. All products sharing the same code
will be displayed on the grid using the same color.
Then, press Click here for the next steps and the analysis runs.
The Product Portfolio Analysis function produces a grid and a summary report as shown below.
The summary report shows sales, market shares and market growth of products grouped in the four classes: question marks, stars, cash cows and dogs.
In column B the number of products belonging to each category are summed up. The sales of each class are expressed in value form (column C) and as a percentage (column D) computed on total sales of the products portfolio.
The portfolio matrix is built on two axes. The horizontal axis (X) tracks the Logarithmic Relative Market Share (the Relative Market Share is computed dividing the sales of one product in the portfolio by the sales of its direct competitor. The Logarithmic Relative Market Share is simply the Relative Market Share expressed on a logarithmic scale. This helps to highlight the decreasing effect of competitors power as the product's market share increases)and the vertical one (Y) shows the Market Growth as a percentage. The picture below summarizes the major aspects to consider when interpreting the matrix.
Notice that logarithms to base 10 are used to scale the X axis. Full details of the logarithmic scale computation is given in the Technicalities chapter. At this time, for the sake of simplicity, a logarithmic relative market share of 1 is obtained when the sales of our product exactly match those of our direct competitor. A logarithmic relative market share of 10 equates to our sales being twice those of our competitor, 100 means three times and so on. When our logarithmic relative market share equals 0,1 we are 10 times smaller than our direct competitor.
The Y axis intersects the X axis on the value 1, all bubbles (products) placed on the right side of this value are not market leaders (market leader is the product collecting the highest revenue in one marketplace, in either value or units).
The Y-axis, market growth, is usually set at the average level of all markets in the portfolio. However this is not always a convenient way of computing it, so the Product Portfolio tool offers two alternatives: the median of all markets or a manually input value. The analyst will use the median to get rid of very high or low growth values that can produce unreliable average figures. It is important to define a coherent crossing value for the Y axis, for it splits the products in fast or slow growing markets, and this affects the way the position of single products will be interpreted and evaluated.
Both leaders and non-leaders can compete in a market with
high, slow or negative growth rates. In all cases there are
implications concerning cash flow and
resources allocation.
There are four different quadrants, or product profiles,
displayed on a Portfolio Matrix and, according to BCG, each
should be managed observing, at least, the following rules:
These products live in an uncertain situation. The market is dynamic but their market share is low. The management carrying on these products must plan high investments to keep them competitive and cannot expect returns, least not in the short term. This can be the case with newly launched products. With established products, the divestment option can be evaluated, given the fairly unattractive position of the question marks in the product portfolio.
The stars are leaders in dynamic markets. They need large amounts of money to keep investing, yet they also produce cash to finance themselves. This way stars result in a slightly negative or positive saldo. Their strategic orientation must be aggressive to constantly gain market share. These are the products, that will probably be feeding the company's portfolio in the future, and it is important to retain a certain number of them in the current assets.
The cows, leaders in mature markets, mainly sustain the cash need of the portfolio . Cows do not have a high cash requirement and generate a large positive saldo. One can expect these products to be under the attack of smaller competitors. The reasonable strategic path for these products is to maintain their market share.
Products, which are not leaders and compete in slow growing or recessing markets, are termed dogs. These either need or generate cash. Dogs can definitely be unattractive and it is suggested to harvest them. However, not all dogs are unattractive. In the grid above, the dotted line splitting the low quadrant on the right side of the map in two triangles, divides the very bad products (lower triangle) from the less bad ones (upper triangle). Usually, dogs tend to gather as much cash as possible before being divested, although it can be hard sometimes to keep their saldo positive.
BCG Interpreter, as the name implies, reads and evaluates the output of the Portfolio Analysis built into MM4XL. It looks in an objective manner at the product portfolio and highlights strengths and weaknesses. This helps managers to take corrective measures where needed, and ensures consistency in pursuing the envisaged product portfolio policy.
Interpreter saves you time and offers solid support for users who have not yet mastered the way Portfolio Analysis works. Managers called upon to shape the strategic route of companies will find it a valuable support tool, also taking investment levels into account and enlarging the overall strategic picture.
Warning:
BCG Interpreter alone, does not work with dynamic portfolio analyses, and consequently does not provide useful data.
Choose BCG Portfolio Analysis from the MM4XL menu in Excel and enter the number of products to be analyzed, as shown in the chapter How to Run the Product Portfolio Analysis. Hit OK and the window below displays.
Tick the checkbox Interpret my portfolio analysis and input the two required values (see the chapter Input Values for details):
Hit OK and the predefined input data range of the BCG Portfolio Analysis appears. This differs in the last column (Investment), from the input range of the BCG analysis without Interpreter. Fill in all columns. Investment values must be in the same units as sales, for both yours and those of your competitors. If sales are expressed in millions, investment should also be in millions.
This is all you input to run BCG Interpreter:
BCG Interpreter returns a numerical summary report, three charts, and a verbal report that highlights the most relevant aspects of the analysis.
|
Tip: Divide figures in millions by 1,000 or, even better, 1,000,000. Charts and reports will be neater. |
Before you run Interpreter, define the format of
the sales data and make your present portfolio policy
clear.
Using cost price for sales rather than retail price has a
profound impact on both investment and cash
flow levels. This is especially true for investment levels,
which should also be defined as accurately as possible, in
order for the analysis tool to produce reliable results.
Choose one of two alternatives for portfolio policy: Profit or
Competitiveness maximization. The latter is a long-term view
and the former is short-term, and as many suggest, the first
should be favored. In order to maximize competitiveness,
investments in growing markets should be high. This, however,
implies the company has attractive products to invest on and
has the willingness and skills to establish a brand in a
leading position. Maximizing profit implies lower investments,
in order to show as much cash as possible. In general,
short-term policies may make sense for some products in a
mature market, such as some stars and dogs. Yet, not
necessarily all stars and dogs should follow a short-term
approach.
|
Tip: Portfolios made up of many old products may be split into products that require a profit maximization approach, and products that require competitiveness maximization. Two analyses should be run and the results interpreted conjointly. |
Investment level is a standard that changes from industry to industry and from market to market. A value equal to 1,2 means that, in order to be balanced investments in growing markets (stars and question marks), must be at least 20% higher than investments in slow growing or recessive markets (cash cows and dogs). The same concept applies to the Cash flow level. Cash flow produced by leader products must be higher than that produced by non-leader products. A value equal 2 means that leaders must produce twice the cash flow of non-leaders. Both cash flow and investments cannot have a value lower than 1. At least 1,2 is recommended for investments and 2,0 for cash flow.
|
Tip: Try the default values, read the analysis, and then change the values if required. |
The picture below is a legend that summarizes the concept of portfolio evaluation built using BCG Interpreter.
In order to set the coefficients at reasonable levels, it must be clear which investments are included in the analysis. In general, cost of sales force, communication (advertising, promotions and public relations) and variable costs may cover a reasonable share of the whole expenditure managed in a marketing department.
BCG Interpreter produces three charts, one table
of values and a verbal report.
The bubble chart plots investments and cash flows, both in
value. The size of the bubbles is proportional to the sales of
all products belonging to one market segment (product group).
The axes of the chart cross at the average value of both
dimensions.
The remaining two charts are self-explanatory.
Investment and cash flow values computed for the whole portfolio are displayed just above the verbal report.
In our example, investments in growing markets
are 1,75 times the level of investment in non-growing markets.
This is consistent with the portfolio management theory
outlined above. Cash flow of leader
products is also higher than that of non-leader ones. This is
also a reasonable condition. As a double check, run the
analysis again using values slightly higher than the values
computed by the software, say 1,9 and 2,3, and compare the two
analyses. This may prove useful to prevent surprises related to
financial and competitiveness matters.
The report of BCG Interpreter follows the BCG Summary
Table.
The verbal report offers an interpretation of the portfolio as a whole, of investments, cash flow, and of each quadrant in the share/growth grid. The illustration below shows the structure of the report.
The portfolio is evaluated by means of investments and cash flow. MM4XL compares the user input to the values produced by the portfolio analysis. A positive judgment is expressed if the latter are higher than the former, otherwise, MM4XL reports on the risk. Beside each evaluation, a solution is always offered. The same concept, evaluation and suggestion apply to both investments and cash flow.
The quadrants of the share/growth grid are evaluated in
terms of number of products, market growth, and strength of
leadership. Market growth is not evaluated for non-leader
quadrants (question marks and dogs).
MM4XL warns when a portfolio has few cash generators and too
many cash absorbers. It also flags the presence of slow or
dynamic markets. In all cases the software elaborates and
offers suggestions of how to deal with the situation.
The evaluation of strength of leadership is based on the concept that leaders are powerful when they make at least 1,5 times the sales of the direct competitor. MM4XL also applies the concept the other way around, to non-leaders. In other words, to have a strong position, leaders must have a logarithmic relative market share of at least 1,5. On the other side, non-leaders with a logarithmic relative market share of less than 0,5 are considered very weak.
The Market Segments table follows the verbal report. It shows statistics of the product groups (as defined in the column Product Groups of the predefined data range. See the chapter How to Run the Product Portfolio Analysis ). An example is given below.
In the above illustration, the product group
Canned Food, hence market segment, comprises 2 products, which
made sales for 1779,0 and the direct competitors for 1684,0. We
are leader, as shown by the relative market
share, but with very little margin of safety, the strength
of leadership is low. The segment grows 5.3%. We have invested
1237,9 and 619,0 on average by segment. The segment has a cash flow of 541,0, on average 270,5 for
each product. Investments are about 27% of sales. The data
above is given for each segment that the analysis accounts
for.
Users can take full advantage of Excel flexibility. Any parts
of the analysis can be cut and pasted elsewhere.
|
Tip: Run the BCG analysis twice, once with real data and once with investment, market growth and sales values as products of educated guesses. Use the different scenarios to speculate about the future. |
The output of the BCG Portfolio Analysis combined with the output of BCG Interpreter present an objective view of a portfolio and can be used to evaluate strategic decisions.
There are two major assumptions behind the Product Portfolio Analysis, which help to allocate resources whilst reducing risk:
The following map depicts the four discussed product typologies.
If we translate the two rules above as equations and also
include the equilibrium and the opposite of each rule, we have
six different conditions:
1. Investment(Stars + Question marks) > Investment(Cows +
Dogs)
2. Investment(Stars + Question marks) = Investment(Cows +
Dogs)
3. Investment(Stars + Question marks) < Investment(Cows +
Dogs)
4. Cash(Stars + Cows) > Cash(Question marks + Dogs)
5. Cash(Stars + Cows) = Cash(Question marks + Dogs)
6. Cash(Stars + Cows) < Cash(Question marks + Dogs)
Combining the six outcomes gives nine possibilities when defining a portfolio, representing five major classes of product portfolios, respectively:
Each of the five positions implies certain peculiarities.
A product portfolio in equilibrium is not necessarily a good one. The financial resources are very limited and, should it also be the result of a mix of products competing in growing markets, the competitive power can be very weak and perhaps not worth the risk.
A profit-maximizing portfolio can generate cash in the short term, yet this is hardly a position to be held long term. Investments are low and can be insufficient to maintain the current market share level, however, it can be an interesting position to be held for a short time. A useful stop-gap while building resources to be used against a specific target, such as a competitor or market niche .
This is the result of an unbalanced portfolio, which will normally survive for a very short time only, unless action is taken to reduce the investments or enlarge the cash flow.
The optimal condition is given by a portfolio whose products lie in fast growing markets, investing more than those in stagnant or recessive markets and the cash flow of leaders is larger than for non leader products.
The portfolio optimization is a topic mainly handled by financial analysts. Markowitz and Sharpe have developed models to make empirical analyses of the portfolio performance and to seek the best balance of assets in order to maximize return and minimize risk. An interesting discussion on the components of the economic evaluation can also be found in Wind. Although written a long time ago, Alexander and Francis present a well-detailed review of models to optimize assets portfolio, which can be easily adapted to product portfolios.
The logarithmic scale of the X axis is justified by the fact that the competitive power of one product does not grow linearly with its sales, but rather in a logarithmic manner. For example, the following table shows the sales of our product, held constant over time and equal to 100, the sales of our major competitor, increased in a linear way, the relative market share and the logarithmic market share. These values have been used to draw the graph below and show the importance of working with logarithms.
The slope of the curve of the relative market share of our product (called Linear in the graph) slows down almost linearly as the competitor wins share. This means that no reference is made to the increasing power of the competitor, given its fast growing absolute dimension. This is wrong. Indeed, we know that well-managed best selling products get better prices for row materials, large productions sink variable costs, large market shares attract consumption faster than low selling products. These and other factors can boost one company's competitive power, yet decreasingly so.
A logarithmic curve drawn on an equally scaled axis is pictured below:
The BCG was aware of this and encapsulated the concept in the logarithmic scale of the relative market share. Short term, the larger one product's sales, the stronger its competitive power, yet at a slower increasing degree. The logarithmic relative market share is computed with the formula:
The diameter of the circles displayed on the Product Portfolio Analysis grid is proportional to the product sales. The bigger one product's circle is, the larger its share of sales in the whole products' portfolio. The sum of all shares of sales must equal 1. The size of a product's circle is given by the formula:
Joram Y. Wind
Product Policy: Concepts, Methods and Strategy
Addison Wesley, 1981
Arnoldo C. Hax and Nicolas S. Majluf
Direzione Strategica
IPSOA, 1987
Lilien, Gary L., & Rangaswamy, Arvind
Marketing Engineering
Addison Wesley, 1997
Philip Kotler
Marketing Management: Analisi, Pianificazione e
Controllo
ISEDI, 1984
Mourray Bourne
Interactive Mathematics: Exponential and Logarithmic
Functions
http://www.np.ac.sg/~bms/Index4.htm
Markowitz
Mean-Variance Analysis in Portfolio Choice and Capital
Markets
Basil Blackwell Ltd, 1987
Sharpe
Portfolio Theory and Capital Markets
McGraw-Hill, 1970
Alexander, Gordon J. and Francis, Jack Clark
Portfolio Analysis
Prentice-Hall, 1986, 3rd edition
Barry Hedley
Strategy and the Business Portfolio
Long Range Planning, vol. 10, Feb 1977, pgg 9-15.