Managing Co-Development Projects:
Strategic Alliance or Alliance Strategy?
Perspectives from Ben Gomes-Casseres, co-author, Mastering Alliance
Strategy As co-development becomes
an increasingly important strategy, a body of research has grown up around this practice
area. From managing a portfolio of alliances, to the finer points of Joint Development
Agreements, alliance strategy has emerged as a field of study that is more and more
relevant to product development. Ben Gomes-Casseres, a professor at Brandeis University
(Waltham, Mass.), is an authority on alliance strategy and management, with a focus on
technology companies. He has researched this topic for 20 years, taught it to MBAs and
executives, and consulted with major companies in the United States and abroad. In the
following article, we present some highlights from his ongoing research into best
practices in the domain of alliances.
One of the most important, high-level
findings of Gomes-Casseress research is that it is necessary to focus not only on "doing
the deal," but also on an overall alliance strategy that is tied to
overall business strategy. Individual alliance agreements must then fit within the
parameters of the alliance strategy, which itself may evolve over time. The Arc of
Alliance Strategy [see graphic] shows the four key elements of an alliance strategy: (1)
alliance design, (2) alliance management, (3) alliance constellations, and (4) alliance
capabilities. Gomes-Casseres encourages companies that engage in co-development projects
to consider the entire value chain and to carefully weigh the resources they will need to
devote to the alliance to realize the benefits they expect.

Uncertainty and "Open-ended" Projects in
Co-Development
One facet of alliance strategy involves choosing a position
for each project on the spectrum of possible types of alliances. Some of these types
involve clear-cut deliverables and tasks defined up-front and others are, by nature, more
"open-ended." On one side of the spectrum is, for example, a simple contract
between one organization and another to supply a component for a product. This type of
agreement is simple, straightforward and definable. On the other end of the spectrum is
"blue sky" research where there is a great deal of uncertainty around
technologies and markets. This type of project, in which both risks and uncertainties are
significant, lends itself to a more open-ended type of alliance that is more difficult to
manage.
"The open-endedness is not something you choose, it
is imposed on you because of the task," says Gomes-Casseres. "If the task
is not open-ended, you can define the requirements, timetable, milestones, payment and
penalties clearly at the start of the project. If the task at hand is not well defined,
then its, by nature, open-ended, and you need to manage it as such. In these cases
you need to create a governance structure that allows you the flexibility to go with the
new information that you will gather over time. A flexible structure allows you to deal
with things later that you cant foresee now."
Between these two extremes there are numerous shades of
gray all along the spectrum. "One project might have some amount of certainty, but
there will be uncertainty around some aspect of the project such as usage, cost, or
materials design, and that particular aspect might need to be left open."
Gomes-Casseres points out that, in addition to needing to have a governance structure
which allows for flexibility, projects with large degrees of uncertainty require that the
parties involved meet more often, and not tie themselves to early decisions when new
critical information is uncovered as events unfold.
Not surprisingly, open-ended projects require a great deal
more relationship management than simpler contracts. In these open-ended projects, he
explains, "You need to set, in the agreement, the way you will relate to each
other, in addition to the things you expect each other to do. What you expect each other
to do might change, because you dont know how everything is going to work out yet.
The way you relate to each other, however, can be established." For
example, Gomes-Casseres has observed that some firms with a good alliance strategy...
- Have personnel assigned to project manage the alliance,
- They have some decision rules that say how they plan to
resolve potential conflicts with partners,
- They have conflict management and escalation rules for the
times when difficult differences arise between partners, and
- They have sufficient people-to-people contacts which greatly
helps in managing the flexible part of the relationship.
In addition, alliances require that both partners have
sufficient internal alignment, communication and coordination. Says Gomes-Casseres,
"One success factor is managing the relationship with the other party; another is
managing the alliance process within your own firm. Managing alliances requires good
internal alignment and communication as well as attention to the relationship
management."
Key Capabilities for Managing Co-Development
Alliances
But in addition to these general principles, there are some
key capabilities that make a firm successful in managing co-development relationships.
What are these capabilities and what are some common pitfalls to avoid?
According to Gomes-Casseres, there have to be incentives
within the organization that favor success in co-development. Peculiarities built into the
incentives and processes in some organizations can inhibit successful co-development
alliances. For example, says Gomes-Casseres, "Some companies are set up in such a
way that rules about contracting, or about how the P&L will be charged, may create
biases against working externally, even when an external solution is demonstrably
better." Managers should check the incentive structure to make sure it does not
put obstacles in the way of co-development success.
In addition, successful firms recognize that it takes
personnel, time and investment in order to properly manage these external resources. "It
does not happen automatically. Once you create an alliance, you have not gotten rid of a
management task you have created a new management task," he says. He
argues that "outsourcing" can be a misleading term, "since it
makes it sound like it is on somebody elses books and the main partner no longer
needs to think about it." On the contrary, he says, outsourcing of complex
processes or project components typically "requires a lot of management for it to
work well, and to ensure that it is properly integrated into what you are doing
internally."
According to Gomes-Casseres, one of the most common errors
firms make in managing alliances is that, after they conclude an agreement, they
under-resource the ongoing management of the relationship. When it is done right, adequate
human resources must be devoted to managing the alliance. Gomes-Casseres has observed that
many firms have alliance managers whose task it is to nurture and watch over the
relationship, to surface problems and fix them, and to make sure that communication is
working well. Some companies gather together a group of alliance managers into an alliance
management unit. "When you group these people together, they share best
practice and they get better at it," he observes. "Its a
learning process, just like any other capability."
Another common pitfall is the "Not Invented
Here" syndrome which, says Gomes-Casseres, "can doom this sort of
relationship. There has to be the attitude that the partner is capable and has sufficient
technology in its field. You have to be open to use the best there is, be it your own or
someone elses technology." He points out that many leading firms do not
automatically assume an open and accepting attitude toward outsiders. This is another area
where learning and internal development are necessary if alliances are to thrive.
Measuring the Performance of Co-development Alliances
But how can the success of open-ended, co-development
alliances be measured? Research and practice suggests that there are two broad classes
of measurements associated with alliances. The first class of metrics measures the success
of the project in technical and business output. This is "not all that different
from measuring the output of an internal project," he explains. These types of
metrics are the basic measures many organizations are already using, such as adherence to
the schedule, meeting project requirements, customer satisfaction, and so on.
Another set of metrics can be used to measure the progress
of the relationship. These metrics monitor the quality of communications, the ability of
the partners to understand each other clearly, etc. Gomes-Casseres believes it is
important to separate these two types of measurements because it is easy for partners to
blame each other for some technical result that may be in the nature of the technology or
the marketplace. "Its easy to confuse the value of the relationship with the
output of the project, but they are two different things. The job of someone who is an
alliance manager is to makes sure that the quality of the relationship is not a hindrance
to the success of the overall enterprise," says Gomes-Casseres.
The purpose of measuring the progress of the relationship
is to help reduce avoidable friction between the partners, knowing that there will
be unavoidable successes and failures in the market and technology. "What
you are trying to do is limit the total risk in the project," explains
Gomes-Casseres, "so you have to limit the degree to which involving outside
parties will add to your risk."
In fact, alliances, in themselves, can be a method of
managing the risk inherent in new product development projects. "Sometimes
alliances allow you to spread the risk around," says Gomes-Casseres. "But
they do it at the cost of raising relationship risk. When you deal with partners you
create uncertainties, in part because you cant fully control your partners
behavior. Good alliance management tries to lower the relationship risk, so that you can
engage in projects with significant technical or market risk."
Being a Good Partner Choosing a Good Partner
Gomes-Casseres emphasizes that alliances must always be a
two-way street. To benefit from co-development partnerships, and to attract good partners,
it is necessary to be a good partner. He recommends that firms who wish to be
successful must step in the shoes of the partner and see themselves from the
partners standpoint. "Remember that the reason you are working with another
company is because you think they can do something you cant do or cant do as
well. They have answers and you have to listen to their point of view."
A good partner, says Gomes-Casseres, asks the key question,
"how can we, together, create something more valuable than either of us can
do separately? If you are not comfortable with this approach then stick to narrower
vendor-type relationships. But you will not get out of these what you might get out of a
well-managed alliance." By definition, suggests Gomes-Casseres, these
relationships cannot be maintained through the usual command and control structures that
executives have been taught to use in managing a company.
But what about choosing partners who will be able to travel
down the "two-way street"? Are there signs to look for that will indicate
whether or not a potential partner will be an effective one? According to
Gomes-Casseres, reputation is a good indicator: "In this world, where there
are more and more partners, the ability to partner can become a corporate asset or a
liability. If you are a small firm dealing with a large partner, try to find out how they
have dealt with other firms of comparable size. Track what has happened."
In addition to a good reputation, another positive
indicator is whether or not a potential partner is willing to commit the resources
necessary to manage the relationship effectively.
Gomes-Casseres proposes a few questions to think about when
selecting partners:
- When they work with you, do they have the attitude of
creating a partnership?
- Are they concerned with building the relationship or just
with negotiating the hard facts?
- Do they devote attention to your views and the quality of
the relationship?
- Will they put an alliance manager in place?
- Is the structure they propose balanced or one-sided?
Gomes-Casseres also indicates that the information
available about potential partners depends, to a certain extent, upon the industry in
question. For example, in the pharmaceutical industry, partnerships between large firms
and much smaller entities are extremely common. The ability to be a good partner to a
smaller firm is a corporate capability on which larger firms compete. In this industry,
market-research firms have been known to produce rankings of firms rating their
performance as partners. In other industries, information about potential partners might
not be discovered as easily.
Alliance Constellations: A Portfolio Management
Approach to Co-Development
In addition to managing individual relationships,
Gomes-Casseres also counsels co-developers to look at their entire suite of alliances
and to manage them as a portfolio. This portfolio-based approach, what he calls
managing "alliance constellations," is a way of handling the multiplicity
of alliances that many companies now require. "Usually, companies that are
co-developing products are dealing with more than one entity," observes
Gomes-Casseres. "When you think about these partnerships, you need to think of the
total value chain that is going to be delivering this product or service. Managing this
chain is managing the constellation. It is a portfolio, a group of alliances that have to
work together."
Gomes-Casseres points out that there are many different
types of alliance constellations. For example, there are constellations to create products
and services; there are constellations that are formed to create a standard within a given
industry; there are constellations to create vertical markets (for example for a
technology or a material that has many different applications in various industries or in
a variety of geographical locations); there are also constellations of market allies.
Gomes-Casseres suggests that someone in the company has to pay attention to managing this
alliance portfolio. "It goes beyond striking deals," he says,
"and includes managing the portfolio over time to see how its growing, which
alliances tend to be the most beneficial and which could be closed, how resources should
be allocated, and so on."
This question brings him back to the theme that underlies
his work: Rather than focus on stand-alone strategic alliances, he argues that
firms should pursue a comprehensive alliance strategy. Says Gomes-Casseres, "Its
not just a question of did this alliance succeed or fail? but what is the
bigger picture? Will this portfolio get you the result youre looking for?"
In sum, he suggests the following key points, (see also the
other tips found in the side bar on the previous page):
- Realize that certain tasks are inherently open-ended and
require a partnership approach
- Focus on the whole alliance strategy rather than the single
deal
- Devote resources to managing alliances; be sure that the
relationship gets the proper attention.
|