Develop a decision framework for project portfolio management at XYZ:
• What are the objectives?
• What are the constraints?
• What are the risks involved?
• What are your alternatives?
• What information is required for project portfolio management at XYZ and how can it be collected?
Part 2 – Project Valuation
Before thinking about appropriate portfolio decisions, the value of each project in the portfolio needs to be determined. How would you determine the value of the following project (‘Project 1’) in XYZ’s portfolio, a project in the pre-clinical phase, part of the Oncology therapeutic area? What additional information would you collect? Which quantitative tool(s) might help you in determining the value of the project?
Part 3 – Project Risk
When implementing project 1, you face technical and market risk. How would you assess the risks embedded in Project 1? What additional information would you collect? Which quantitative tool(s) might help you in determining the project risk?
Part 4 – Project Portfolio Decisions
Suppose that next year’s R&D budget for the oncology area has been reduced to $50 million. How would you decide which projects to continue, and which to put on hold? What additional information would you collect? Which quantitative tool(s) might help you in determining the best portfolio?
Project Portfolio Management
at XYZ Pharma
Early morning, Monday 29th August 2005. John Smith, head of portfolio management and strategic
planning, was paging through the slides he had prepared for the Portfolio Management Board (PMB)
meeting which would start at 9 am, and which was scheduled to last until Friday. We have been
preparing this meeting for weeks, he thought, and it seems the PMB has some tough decisions to
The PMB of XYZ Pharma, the pharmaceutical division of XYZ, one of the worlds leading companies
in the life science sector, convenes yearly in August to review the composition of the research and
development (R&D) project portfolio. It also meets on a monthly basis to monitor the project portfolio
and make decisions regarding new developments. According to John Smith, The PMB is an
important decision making body because it shapes the future of the company by determining its
The PMB members include the CEO of XYZ, the CEO of XYZ Pharma, the heads of the different
business units, the heads of Development, Research, Global Marketing and Strategic Planning, the
regional heads for the US, Europe and Japan and the functional managers for Regulation, Clinical,
Licensing, Technical Research and Development, and Patents.
The portfolio group, led by John Smith, had analysed the project portfolio carefully and had
highlighted several potential threats that required action. According to John, There will be an indepth discussion of which projects will be allocated additional resources, and at expense of which
other projects this will be.
The Pharmaceutical Industry
The lion share of the pharmaceutical market is captured by approximately hundred manufacturers,
which account for more than 90% of global sales. Exhibit 1 contains the top twenty pharmaceutical
companies ranked by sales.
Since the mid-1980s, the pharmaceutical industry has been
characterized by large and frequent mergers and acquisitions (see Exhibit 2), which have had a
dramatic impact on the pharmaceutical landscape. Nevertheless, each pharmaceutical giant only
holds a relatively small share of the total drug market.
This case is written by Bert De Reyck, Associate Professor of Decision Sciences at London Business School, Zeger Degraeve,
Professor of Decision Sciences, and Pascale Crama, doctoral candidate. The case is prepared solely as the basis for class
discussion, and is not intended to serve as a source of primary data, or as an illustration of effective or ineffective management.
Copyright 2005 by London Business School. No part of this publication may be reproduced, stored in a retrieval system,
used in a spreadsheet, or transmitted in any form or by any means electronic, mechanical, photocopying, recording, or
otherwise without the permission of London Business School.
Project Portfolio Management at XYZ Pharma
The pharmaceutical market is characterized by increasing competition between brand-name drugs,
illustrated by the shrinking time span in which a drug is the sole drug for a specific therapeutic class.
Also, the profitable lifetime for drugs has substantially decreased over the last decade, largely due to
quick approvals of generic copies of brand-name drugs, virtually eliminating the time lag between
patent expiration and entry of generic competitors into the market.
The pharmaceutical industry is increasingly multinational in scope, with most research-based
companies marketing products globally. Approximately 47% of R&D is performed in the United
States, followed by Japan with 13%, the United Kingdom with 9%, France with 8% and Germany with
7% (see Exhibit 3). Approximately 45% of drugs developed are from U.S. origin, 14% originated from
the U.K., 9% from Switzerland, 7% from Germany and Japan and 5% from Belgium (see Exhibit 4).
The US is by far the largest market, accounting for almost half of global sales, which totalled $550
billion in 2004 (see Exhibit 5).1
The Drug Development Process
R&D Expenditure as % of Sales
Drug discovery and development is an extremely risky, time-consuming and expensive process. The
average time from compound to market has grown from 8.1 years in the 1960s, to 11.6 years in the
1970s, to 14.2 years in the 1980s and 1990s.2 Lengthening development times also increase
development costs. Recent estimates indicate that the cost of developing a medicine is around $800
million3, significantly higher when compared to 1990, due to a substantial increase in important cost
drivers such as the number of required clinical trials and patients per trial. This has resulted in a
doubling of development costs since 1991, and a threefold increase since 1980. In contrast, the cost
of demonstrating bio-equivalence of a generic product, the key requirement for approval of a generic
Global R&D expenditures by research-based
drug, is currently estimated at $1 million4.
pharmaceutical companies is estimated at around $40 billion in 2001, increasing at around 15% per
year (Figure 1). As a percentage of sales, R&D expenditures have risen from around 11% in the
1970s to approximately 16% in 2004.5
Figure 1. Worldwide R&D Expenditures ($ millions, inflation adjusted)
The Pharmaceutical Market Outlook to 2015: Implementing innovative, long-term strategies for sustainable future growth,
Business Insights (citing IMS), May 2005.
Joseph A. DiMasi, Director of Economic Analysis, Tufts Center for the Study of Drug Development, Tufts University, Boston,
Massachusetts, testimony before the House Committee on Commerce, Subcommittee on health and the Environment, 105th
Congress, 1st Session (April 23, 1997).
A Methodology for Counting Costs for Pharmaceutical R&Ds, Tufts Center for the Study of Drug Development, Nov. 2001.
Barfield, C.E. and C. Beltz, Balancing and Rebalancing the National Interest in the Patent System, American Enterprise
Institute, Oct. 1995.
Pharmaceutical Industry Profile 2004, PhRMA, Pharmaceutical Research and Manufacturers of America.
Project Portfolio Management at XYZ Pharma
Newly developed drugs are protected by patents, providing pharmaceutical companies with the
opportunity to recuperate their investments and create profits during a period of market exclusivity.
Typically, 20-year patents are granted, although in general this results in a post-approval patent life of
approximately 12 years. After a patent has expired, generic drugs identical to the newly developed
drug can be freely sold without the need for extensive clinical trials.6
The drug registration process is heavily regulated. Stringent scientific procedures have to be followed
to ensure patient safety in distinct stages, including pre-clinical and clinical tests, before a medicine
can be approved for production and marketing. The drug development process in the United States is
monitored by the U.S. Food and Drug Administration (FDA). Comparable institutions exist in other
countries around the world. The EU created a pan-European equivalent, the European Agency for
the Evaluation of Medicinal Products (EMEA) which grants marketing authorisation for the whole EU.
The US drug development and review process is typically as follows (a similar process is followed in
Basic Research (approximately 2 years) In this phase, numerous compounds are synthesized,
extracted and tested in a combinatorial and iterative manner in order to discover new substances with
beneficial effects. This stage lasts for about two years, costs around $30 to $50 million, and on
average only 40 out of an initial 10,000 compounds are taken to the next stage of pre-clinical testing.
Pre-Clinical Testing (approximately 3 years) In this phase, drug safety and toxicology is
established through animal testing, while data is also gathered on the biological effects. The
development of a drug is terminated when tests suggest that it poses a significant risk for humans,
especially in the areas of organ damage, genetic defects, birth defects or cancer. On average, only
one in four drugs passes this phase.
Human Clinical Trials (approximately 6 years) Drugs for which the pre-clinical animal data does
not show an unacceptable safety risk for humans, termed Investigational New Drugs (INDs), are
then subjected to human clinical trials, the most stringent and time-consuming process, in which
people are observed for adverse effects. All harmful reactions result in termination of the drug, or are
incorporated in the drugs package labelling if the adverse effect is deemed acceptable. On average,
one in four drugs passes this stage to move on to the FDA review. This phase entails approximately
70 clinical trials involving 4,000 volunteers, with total costs often exceeding $200 million. It is
composed of three sub-phases:
Phase I Safety Trials (1 year)
This phase involves testing highest tolerated doses and toxicity, typically done with a few
dozen healthy volunteers (50-100).
Phase II Safety & Efficacy Trials (2 years)
In phase II, efficacy and long-term safety of the drug are tested with hundreds (200-300) of
volunteer patients with a control group receiving placebos.
Phase III Long-Term Safety & Efficacy Trials (3 years)
Phase III is the longest and most expensive phase, where the drug is tested on thousands
(more than 3,000) of volunteer patients (including elderly people, patients with multiple
diseases and patients with impaired organs) for long-term safety, optimum dosage levels and
more subtle adverse effects.
FDA Review (approximately 1-2 years) In this phase, a New Drug Application (NDA) document is
submitted to the FDA with data on each treated patient, and with production plans. An NDA
documents typically contains thousands of pages, and takes up to two years to review by the FDA.
The FDA continues to monitor the process after approval is granted for production and marketing. On
average, eight out of ten drugs make it through this phase.
In some countries, including Argentina, Brazil, Mexico, India, Egypt and South Africa, patent piracy, where protected drugs
are copied without compensation, has sometimes been a major problem. However, many of these countries have recently
tightened their patent protection to international standards.
Based on data from the Center for the Study of Drug Development, Tufts University, 1995.
Project Portfolio Management at XYZ Pharma
to next Stage
Table 1. Probability of passing each of the drug development phases
Table 1 illustrates the high risks that are inherent to the pharmaceutical industry. On average, only
one in five drugs entering clinical trials is launched on the market8. Overall, only one in five thousand
developed compounds in the research phase makes it to the market. As a consequence, a large
portion of all development costs are spent on drugs that never reach the market, illustrating the high
technical risks involved. Even of the drugs that reach the market, only 30% achieve the commercial
success necessary to recover the (after-tax) development costs to yield a healthy return, illustrating
the additional commercial risks involved. Generally, the top 20% of the products with the highest
revenues generate 70% of the returns. Thus, companies must rely on a limited number of highly
successful products to finance their continuing R&D.9 Nevertheless, pharmaceutical companies in
recent years have been able to report healthy profits, of about 20% on gross revenues. Roughly,
production costs account for 10%-15% of total manufacturing costs, R&D for 20%, taxes for 15%,
30% for advertising and marketing, leaving approximately 20% profit.10
Pipeline and Portfolio Management
The top management of XYZ is committed to a vigorous growth in total sales and the creation of
shareholder value. Because the global pharmaceutical industry is increasingly competitive, a
constant stream of product introductions has to be maintained. A well-managed product pipeline is
essential to support sales and profits, making product or project portfolio management a crucial
success factor, says John Smith. And because of the long R&D lead time, a good performance
today is actually determined to a large extent by which decisions have been made 10 years ago.
The XYZ Pharmaceutical product pipeline is one of the broadest in the industry and currently
comprises a total of 69 projects in clinical development, and 106 projects from the pre-clinical stage
onwards. According to the CEO: XYZs pipeline is already one of the strongest in the industry. He
added: We have a number of strong pipeline compounds as well as limited patent expiry exposure.
The projects in the pipeline include both new molecular entities (NMEs) and additional indications or
formulations for marketed products. Overall, there are 27 projects in late-stage development (Phase
III or FDA review), to sustain mid-term growth, and 32 projects in Phase II. XYZ expects to be able to
launch one or two NMEs per year and plans to introduce new products at a sustained pace.
XYZ Pharma Research is working in a wide range of therapeutic areas, in research centres all over
the world. Each therapeutic area is a separate business unit, responsible for its own performance.
Each of the business units is allocated a research fund from corporate headquarters, based on a
commitment to contribute a certain profit to the Pharma division. Additional profits beyond the agreed
value can be re-used to fund research, or can be transferred to headquarters, resulting in bonuses for
the units employees. Unlike some of its more focused competitors, XYZ Pharmas products span a
wide range of therapeutic areas, including immunology, inflammatory diseases, central nervous
system disorders, cardiovascular, endocrine and metabolic diseases, oncology, dermatology and
asthma. In recent years, however, XYZ has been focusing on both cardiovascular diseases and
cancer. This strategy has paid off, confirmed John Smith, We boast a strong portfolio in both those
areas, driven by blockbusters for a few years to come. However, other areas of the portfolio have
Industry Profile 2003, www.phrma.org
Henry G. Grabowski and John M. Vernon. Returns to R&D on New Drug Introductions in the 1980s, Journal of Health
Economics, 13, 383-406, 1994.
P. Barry, Whats behind high drug prices in the U.S.?, AARP Bulletin, 41 (4), April 2000.
Project Portfolio Management at XYZ Pharma
suffered some setbacks: late-stage trials had to be terminated and some applications had trouble in
the regulatory arena. This has hit the central nervous system unit especially hard, which is not
expected to be a major growth driver anymore.
The Portfolio Management Board
Decisions concerning the project pipeline are taken by the Portfolio Management Board or PMB.
The PMB has two important functions: At its yearly meeting in August, it decides on the shape and
content of the project pipeline by accelerating and delaying projects, and, on a regular basis, the PMB
checks its evolution, says John Smith. My role is to prepare the portfolio data for these meetings,
and integrate the requirements from the different business units into a single portfolio from a company
perspective. Before the meeting, each of the business units submits individual business plans with
capital and resource requirements based on the projects within the unit. Input from Project Teams
(resources, milestones, risks), Strategic Marketing (market performance and potential revenue
streams) and Strategic Planning (disease area audits and benchmarking) begins in early December.
The portfolio group, led by John Smith, consolidates the business plans of the business units. It was
a hectic time this year, but we managed to finalise everything and build a provisional project budget in
two weeks time to be ready for the PMB meeting, remembers John. We needed to allow the PMB
five working days to review the documentation associated with the annual strategic plan.
The yearly PMB meeting deals with an annual budget of more than US$5 billion and considers
approximately 150 projects executed in ten development sites worldwide. Its main purpose is to
decide which compounds to develop and their priority. The resulting development budget for every
business unit is the basis for a contract, explains John. Any individual project can be singled out for
special attention concerning its expected profitability, strategic fit and contribution to portfolio or
pipeline balance. The projects are subsequently monitored by the PMB in quasi-monthly meetings,
which are held to evaluate the performance of the projects against the objectives established in
The PMBs decision process consists of two parts (Figure 2). The preparation of the yearly business
plan takes place during the planning period, from June until August, followed by the implementation
and control of the plan during the budgetary year, from January to January. The planning process
starts with the evaluation of the options in the light of the strategic plan and the analysis of perceived
opportunities. On the basis of this information, the decision makers agree on targets and the optimal
portfolio that enables them to reach their proposed objectives. These decisions are recorded in the
annual business plan at the PMB meeting end of August. During the execution of the plan,
milestones may be reached or opportunities and threats identified, requiring decisions to be taken.
The quasi-monthly meetings are held for that purpose and allow flexible project execution.
Project Portfolio Management at XYZ Pharma
Figure 2. Representation of the Decision Making Process
Portfolio Review Criteria
XYZs CEO had recently announced his expectations of double-digit growth rates for XYZs
pharmaceutical division, which is significantly above the industry average, and one of the PMBs main
concerns was on how to reach that target. To maintain growth, XYZ needed to deliver on its pipeline
and introduce new successful products, compensating for the decline in sales of mature and launched
products. Also, to sustain a continuous growth, the pipeline has to be balanced, says John Smith.
Balance is determined relative to the pipeline fill that is required to maintain the flow of product
launches given historic attrition rates of projects in the R&D funnel. The projects in the pipeline can
be subdivided into innovative and life cycle management projects, or NME versus LCM products.
Though R&D into new molecular entities are less likely make it to market, the reward is typically
higher, and blockbusters are usually found amongst NME projects rather than life cycle management
Although XYZs pharmaceutical division boasts a healthy profit margin, it is heavily reliant on a few
drugs that will recuperate their R&D expenses. The portfolio review group requires that all projects
asking for funds be accompanied by a Net Present Value (NPV) analysis. A projects potential value
is derived from the estimation of future resource requirements, timing of the R&D stages and market
launch, and the projections of sales revenues and associated marketing costs generated by the
Strategic Marketing Group. The sales forecasts are made based on a number of assumptions
concerning the indication and label of the drug, the disease population, the reimbursement potential of
the drug, potential market share and pricing.
Next to financial criteria, we also consider the strategic fit of any project under consideration, says
John Smith. Strategic alignment is assessed based on the strategic plan in which therapy areas of
interest have been highlighted as a result of a disease area and competitor analysis. As population
composition and disease prevalence change, pharmaceutical companies adapt their research focus.
This explains why many companies have been concentrating on chronic diseases such as
hypertension and cholesterol control since the mid-1990s, said John. However, even though most
Project Portfolio Management at XYZ Pharma
pharmaceutical companies have a strategic focus, they cannot necessarily enforce it, because the
R&D process is essentially opportunistic: funding of research in the strategic focus area does not
guarantee discovery of interesting compounds.
Pharmaceutical R&D activities are subject to a high level of risk, which is an essential ingredient of all
the reports presented to the PMB: project values are expressed as expected values, weighted with
the probabilities of reaching the successive stages and ultimately the market. According to John, In
PMB meetings, we only discuss expected values, e.g. expected sales or expected NPV. It is
meaningless to talk about a potential 5 billion drug without taking into account the probability of the
drug ever reaching the market. Unfortunately, there is little that can be done about the technical
success or failure of a project. Requiring a higher success probability before starting a project would
effectively rule out most projects, especially NMEs. Hence we rely on portfolio diversification.
However, it is essential that we monitor the risk in the portfolio, making sure that any decision taken
results in acceptable risk limits.
Next to technological risks, XYZ Pharma also faces considerable uncertainty about the sales that the
product will generate once launched. Initial projections are made for a distant future when the
compounds characteristics are still relatively unknown. A typical NPV valuation is presented in Exhibit
6, in the format used by XYZs portfolio group. The cash flows are discounted using a company-wide
weighted average cost of capital (WACC).
At the start of the meeting, the portfolio management group, represented by John Smith, presented a
summary of the current state of the project portfolio and pipeline. Some of the presented slides are
given in Exhibit 7. Slide 1 shows the number of current projects in each phase in the different
business units. Slide 2 shows the expected NPV per phase in the different business units (in
$millions). Slide 3 graphs the number of expected launches for the next 10 years. These figures take
into account the probability to market of each of the drugs due to be launched in that year.
Slide 4 presents an overview of the net present value of all the projects in each business unit,
represented by a cumulative probability distribution. The distributions show the likelihood of a
particular net present value based on the technical success of the projects in the portfolio. For
instance, Slide 4 shows that the net present value of the projects in therapy area 2 (the curve on the
right) is between approximately $3 billion and $13 billion, and shows that the probability of a net
present value of at least $7.5 billion is around 70%.
Several slides show the expected sales and sales growth, based on the median sales figure. John
Smith commented: We need to look at ranges when forecasting sales, instead of just focusing on the
most likely sales figure. However, this is a major challenge for the marketing group. The
decomposition of expected sales into therapeutic area, project type or brand name is also
communicated. The major drugs together account for about 40% of the sales of XYZ for the next 5 to
7 years, and John Smith claims that [this] means that XYZ is more diversified than most of the other
major pharmaceutical companies.
XYZ is also looking into the respective sales of LCM and NME projects, and of General Practitioner
(GP), Niche and Specialized products (Slides 5 and 6). John Smith adds: Because our current
blockbusters are pretty strong, they allow for interesting line extensions. However, XYZ also has
other promising NME projects due to be released in the next 5 years, especially in the cardiovascular
therapy area and the immune disorder and inflammation franchise. GP products account for the
majority of sales, but have a relatively low profitability, whereas Niche and Specialized products offer
higher profitability for a smaller sales potential. Different type of products might also react differently
to patent expiry: GP products are usually copied very quickly and market share loss can be severe.
Several slides contain financial information for each of the projects, such as NPV, expected NPV,
peak sales and expected contribution to sales growth, as in Slide 7. As John Smith explained: Last
year was the first time they made decisions heavily based on financials.
Project Portfolio Management at XYZ Pharma
Based on the information contained in the previous slides, the decision makers assign...