Of course, there are some risks. These risks include protecting the intellectual property of the startup and making sure that funds flow correctly through the large vendor to the startup.

The other approach is for the CSP to deal directly with the startup. This can be more efficient, effective and produce faster results. Dealing directly with a few startups may also help control some of the problems and resistance in large vendors. If large vendors see that their obstruction efforts will result in CSPs dealing directly with the startups and cutting them out, they may become less obstructive.  At the same time, it is only fair to point out that dealing directly with startups may be a bigger change in business processes for the CSP.

One way to get around this problem is to hire startups as consultants (a well-established business practice) to do the early studies that confirm the viability of a particular technology and lay the groundwork for a demonstration project. Once viability has been established, it may be easier to make the decision about direct engagement with the startup, or combination with a skunk works.

Building an innovation pipeline

The objective is to create a constantly replenishing pipeline of innovation. To do this, the pipeline (as shown in Figure 3) acts as a funnel and a filter to, stage by stage, move successful innovation closer to implementation. At the top of the funnel are a relatively large number of efforts with relatively small amounts of funding. As projects move down the funnel, there is a winnowing process that reduces the number while increasing the level of funding. To do this effectively, there have to be clearly measurable criteria for determining success for each project. These criteria must be explicit, measurable, and clearly stated at or before the beginning of the project.  Successfully meeting the criteria has the trigger of moving the project down to the next level.

Figure 3. The Innovation Pipeline

The top of the funnel has to be constantly reseeded with new projects. These new projects may come from completely different sources. Or, they may come from projects that failed to meet success criteria, learned from their failure, and reinvented themselves to start over. Similarly, there is also a feedback loop in the pipeline process such that the process itself learns from experience and makes adjustments over time.  These adjustments can be anywhere in the process but are most likely to be found in the construction and use of success criteria.

Creating change

Implementing the process described here requires CSPs and the large vendors that serve them to make substantial changes. These include changes in procurement, contracting, advanced technology, operations, and intellectual property. The history of CSPs and associated regulatory and standards organizations has been one of tight control.  Against this background, the break-up of monopolies, and the rise of Internet and cellular technology has shown that CSPs and their large vendors can make dramatic changes.

So, we can say that the CSP vendor ecosystem is both resistant to change and at the same time has made great changes. Even so, it is important not to minimize the potential difficulty of making the changes needed to create a sustainable innovation ecosystem. One way to ease the transition is to start with smaller efforts that pilot the process.