Generally, businesses consider and vendors push a typical set of reasons to use cloud-based contact center solutions. Well-established benefits of cloud-based contact centers include:
- Evergreen technology (the cloud vendor takes care of upgrading the various platform components, allowing clients to avoid the costs of deploying and testing new versions)
- Prebuilt integrations (cloud vendors can amortize the cost of integrating to various third-party components across multiple clients, so each client can get access to many integrations at much lower risk and cost than they would see if they did the integrations themselves)
- Sandbox (businesses can try new technologies and configurations quickly and cheaply in a cloud platform, and then either scale up there or move the tested solution to its own platform)
- Financial flexibility (cloud contact center costs are typically variable rather than fixed, and operating expense vice capital expense)
- Cost-effective and rapid scale-up and scale-down (many business have cyclic demand, and using cloud-based contact center allows them to add capacity when needed and to shed it as soon as it is not needed)
There is another benefit that generally has received very little attention, but that ultimately may be the most compelling reason to adopt (at least partially) cloud-based contact center technology. That reason is what I call cloudsourcing (with a very definite hat tip to "crowdsourcing"), which is the concentration of expertise in the cloud. This is particularly relevant when one considers doing deep analysis of contact center operations.
As I have mentioned before, contact centers are intrinsically complex systems or business processes. As such, it is very challenging to accurately separate causes from effects, or to get to the root causes of a good or bad set of results.
Put another way, while it is certainly possible to manage a contact center using a single metric, it is impossible to effectively manage a contact center that way. No complex system, whether it is the economy, the climate, or your own contact center, can be effectively understood or managed using only a single metric.
Yet when one tries to use more metrics, the problem of interdependency arises. Most metrics involved in complex systems affect each other (this is a hallmark of nonlinear systems, which contact centers very much are). The reality is that, in order to get even close to a true picture of how a contact center works (and more importantly, how to improve how a contact center works), it is necessary to first obtain a large amount of very granular data from a large number of separate systems, then study that data using difficult mathematical techniques.
Furthermore, a single snapshot rarely captures the richness of behavior exhibited by complex systems, so it is necessary to spend considerable time studying data covering various operating conditions to develop a reasonably full understanding of how a complex system works.
And since the best method for getting to the core of a complex system's dynamics is to use a discovery-centric approach rather than a design-centric approach, one has to be familiar with how to look for natural experiments (situations where one or two key variables are directly observable without interference from other variables, which is very powerful but also very hard to achieve), and how to set up experiments that give useful results and that do not impact operations while conducting them. Deep analytics of complex systems is an advanced skill, and it is not one that is readily at hand, even in very large enterprises.
In many visits to contact centers of all sizes, around the world, I have found only one or two that came even close to having the skill sets on hand needed to study large data sets from a complex systems perspective. And I am not surprised, because it is what one would expect.
Very few contact centers have the budget to hire advanced applied mathematicians, and even those that might do so usually end up seeing them pulled off onto "strategic projects" rather than developing a deep and abiding interest in and familiarity with the center's operational data.
And even if these hurdles were overcome, the lucky deep analytics expert would only be able to learn from one set of data (the expert's own organization's data), which means the expert will only be exposed to the range of issues and phenomena that occur within the one enterprise.
All of these factors have made it impractical for businesses to carry out an ongoing practice of deep analysis of contact center operations, despite the fact that routine productivity gains of 5% or more per annum can be obtained through such work.
When you place contact center technology in the cloud, however, it becomes very practical to carry out ongoing deep analysis on behalf of all clients of the cloud service provider. Much as Google does analysis to finely tune their advertising engine (using a wealth of data about consumer behaviors and preferences), so cloud contact center providers should be able to provide ongoing deep analysis service for clients, leveraging massive data obtained from many clients (taking care obviously to anonymize all data and to protect competitive and confidential information of each client).
It is practical, for the cloud vendor, to put together a team of advanced deep analysts, and it is unlikely they will be pulled off for special projects, since the provision of cloud-based contact center platform and services is the core business of the service provider.
Moreover, because data is available from a wide range of clients of different sizes, styles, and industry sectors, the deep analysts operating "in the cloud" are able to learn from a rich data set that presents unlimited opportunity for skills growth and ongoing learning.So, to the list of great reasons to consider cloud contact center solutions provided above, I encourage you to consider cloudsourcing of expertise as yet another benefit. Certainly it is at the core of what we are doing at NVM Labs, and we believe all of our clients will benefit greatly as a result.