A Partnership is Not a Purchase Order


By Ben Gomes-Casseres, Advisor to Powerlinx, and author of Remix Strategy: The Three Laws of Business Combinations

Scott McNealy, ex-CEO of Sun Microsystems, tweeted this message some time ago:

“Most over used phrase in business is ‘strategic partner.’ Favorite partnership for me is a purchase order. Defined charter, beginning, end.”

Unfortunately, this view is precisely why many external partnerships fail. True collaboration is much more than a purchase order. Setting up an external partnership as if it were a PO, at best, leaves value on the table. At worst, it leads to conflict and value destruction.

The difference between a partner and a vendor is now well understood by leading practitioners and scholars, yet many companies still proceed into external collaboration with McNealy’s blinders. Perhaps they can’t be blamed, as a purchase order gives them the semblance of control. But most true collaborations cannot be controlled with fixed terms that are defined in advance — they must be managed through relational contracts that allow you to respond flexibly to new information.

McNealy is right about the need for a charter and time frame for any relationship. But he is wrong in expecting these specs to be as well-defined and as stable as they would be in a typical PO. True collaboration always has open-ended elements — ranging from precisely how new innovations will be implemented, to how products will fare in the market, and even to what priorities partners will pursue in the face of changes in the environment. A true partnership creates ways to manage these uncertainties, and does not default to executing against a predefined contract, as one would with a classic purchase order.

Sun’s own experience is a case in point. Under McNealy, Sun ran its “partnerships” in the spirit of purchase orders. In its heyday, when Sun sold servers running its proprietary architecture and software, the company bought semiconductors from major vendors, such as Fujitsu in Japan. But for many years it took a hands-off approach to new product development — dictating needs, but letting vendors bid for a slot on the Sun server and sink or swim after that. One of the results of this approach was that vendors failed to invest sufficiently to produce advanced products for Sun. Recognizing this, Sun began to work more closely with Fujitsu and Texas Instruments.

This kind of approach is common for firms that are in dominant positions in their industries. They tend to play vendors off against each other and shy away from partnerships that require some sharing of power. But firms seeking to expand into new fields or facing an onslaught of new competition often turn to alliances. Their challenge then is to learn how to manage these partnerships differently than the vendor relationships they had before.

Sun did briefly maintain an arrangement that it termed an “alliance,” one of the few times it used that term publicly. The Sun-Netscape Alliance was formed when AOL acquired Netscape in 1998 but then sought to spin out the software operations of Netscape. For tax reasons, it could not do so immediately, and instead entered into a three-year deal with Sun. The attraction to McNealy, no doubt, was that this deal came with a purchase order from AOL to buy $500 million of Sun products. At the end of the three years, all operations reverted to Sun and the “collaboration” ended, leaving but a trace of itself in Sun’s product line. That trace survives today, buried deeply in the software offerings of Oracle, which acquired Sun after the latter’s long decline. But one wonders whether a much ballyhooed “alliance” by two tech giants of their day couldn’t have made more of an impact if true collaboration were involved.

What does external collaboration look like, if not a purchase order? It depends on the nature of the industry and task at hand. Here are a few examples that illustrate the range of options in external collaboration:

  • Most major pharmaceutical companies are well-known for their joint R&D deals with biotech start-ups. These collaborations take many forms, from sharing personnel to minority investments and licensing of intellectual property. Most of the partnerships involve highly uncertain outcomes and research trajectories, but these uncertainties are structured and managed to share risk and new investments.
  • Airlines depend on coordination and marketing relationships with other airlines outside their home regions and on organizations that manage key groupings such as Star, oneworld, and SkyTeam.
  • Boeing and Airbus both depend on a huge network of suppliers for major components and subsystems (as do automobile companies), while Airbus is itself a consortium of multiple European companies.
  • Google uses a different approach to working with multiple external parties. Its Android smartphones have quickly surpassed Apple’s and Nokia’s phones in terms of sales. Google’s approach was to license its technology widely and work with partners to develop an ecosystem of user applications, operating software, handset hardware, and telecommunication services. This landscape is evolving rapidly, in ways not even Google could foresee.

The key difference between such partnerships and a purchase order lies in how each structure deals with joint decision making. In a purchase order, the parties agree on terms and then execute the deal against this contract. But in true collaboration, new conditions will arise that may well nullify any initial deal. The best that the parties can do in these situations is to set some terms, but then agree to agree later on future decisions. To facilitate future agreements, a well-designed partnership sets up processes for joint decision making and cultivates relationships that help partners communicate.

Not every partnership enjoys these helpful features. When new conditions arise, these partnerships then often default to the managerial equivalent of the flight-or-fight response: Exit in a huff or butt heads until the other side gives in. This usually compounds the error — not only does the collaboration come to an abrupt end, but there may be costly dissolution fees and collateral damage to reputations.

Before following the popular call to make more use of external resources, make sure that you know what you are in for. Design and manage collaboration projects as true partnerships, and leave simple transactions to the purchase department.

An earlier version of this article appeared online in Harvard Business Review.


Let Powerlinx help you find the right businesses for a long-term, mutually beneficial partnership – sign up or log in today.