What’s Inhibiting The Adoption of NFV?
It has been three years since an Industry Specifications Group (ISG) for Network Functions Virtualization (NFV ISG) was formed under the auspices of the European Telecommunications Standards Institute (ETSI NFV ISG). A lot has happened in those three years. The ISG has driven a number of POCs, and open source communities such the OPNFV formation. Industry organizations such as the TM Forum have launched major initiatives to enable NFV adoption., and service providers like AT&T have announced major programs with the goal of implementing NFV. In spite of all of this activity, there has been very little adoption of NFV in production networks. That raises a key question: What is causing the gap between all of the NFV-related activities and the broad adoption of NFV?
I recently published a report titled: Network Functions Virtualization: A Status Update. The report contains the results of a survey that was distributed in December 2015 and January 2016. I gave 144 survey respondents a set of 20 possible inhibitors and asked them to indicate the 3 biggest inhibitors to their company’s adoption of NFV. The top 5 responses and the percentage of the survey respondents who indicated that response are shown in following figure.
Top 5 impediments to NFV adoption
Most of the inhibitors listed in Figure 1 don’t particularly concern me. I say that because using the immaturity of current products as an example, most of the inhibitors listed in Figure 1 will dissipate over time as part of the normal product development cycle.
The lack of a compelling business case concerns me because it's the most frequently mentioned inhibitor and it seems counter-intuitive that it would be a major inhibitor. It's counter-intuitive because so much has been written about the need for IT agility in order to enable businesses of all types to compete in a rapidly changing, global economy. In addition, it's widely accepted that NFV facilitates IT agility by enabling organizations to deploy network services more quickly. So what’s the problem? Isn’t deploying network services more quickly a compelling business benefit?
The answer to that question is a resounding no. In order to build a compelling business case for NFV adoption, organizations must identify and quantify the business outcomes that results from being able to deploy network services more quickly. Does it increase revenue? Improve margins? Reduce churn? In any and all cases, by how much?
Building a business case for a new technology or architecture is never easy. If an organization is deploying NFV just for a particular use case, such as virtual CPE, they need a creditable forecast for the adoption rate as well accurate insight into the customer’s willingness to pay. The challenge becomes even more complex if the organizations wants to take a big bang approach to NFV implementation and hence deploy a single platform that can support any future NFV use case. In this case, they will likely have to make investments that won’t pay off in the short term.
Independent of how organizations intend on deploying NFV, without a solid business case few will be successful. Unfortunately, merely saying that NFV makes you more agile is not a business case.