An interesting yet strategic thing happened the other day. Facebook teamed up with IBM to help top brands achieve advertising personalization at scale by using IBM’s marketing cloud. What’s curious is that Facebook’s cloud computing credentials are impressive; it could not have grown to anywhere near 1.44 billion active users without complete mastery of the cloud.
So what’s going on here? It’s really all about big data analytics and the cloud is simply the best place to do it. Facebook is in fact not the first social media company to tap into IBM’s cloud-based analytics capabilities. Back in October we announced a partnership with Twitter to help transform how businesses and institutions understand their customers, markets and trends – and thus inform business decision.
While the nature of these collaborations is different – with Facebook we are delivering a marketing service while the Twitter relationship is focused on market intelligence – what they have in common is the refining of massive amounts of social media data into actionable insight that is sold as a service. Incidentally, this gives a new and far more powerful meaning to the acronym IaaS: Insight –as-a-Service.
In essence data is the what, cloud is the how, and insight is the why. And that applies as much to mainstream enterprises – which are sitting on their own mountains of valuable data, as yet largely unmined – as it does to the Web 2.0 crowd. According to a Nucleus Research study, analytics pays back $13.01 for every dollar invested.
By why do it in the cloud? The simple answer is to reverse the question: why wouldn’t you? After all, the operational efficiencies behind the rapid rise of cloud to become the dominant architecture for computing resource apply as much to analytics as they do to other kinds of workloads; perhaps more so given its peaky, compute-intensive nature. Little wonder then that IDC predicts spending on cloud-based big data and analytics solutions will grow three times faster than spending for on-premise solutions over the next five years.
There is, however, rather more to it than that. Once you start down the path of big data analytics pretty soon you want to add external data into the mix. While there’s plenty of data in public domain there’s also an awful lot that belongs to someone else. Having an efficient and secure method of accessing/sharing such data therefore becomes very important, which is where hybrid cloud solutions come into play.
The new Facebook/IBM partnership is a great illustration of this. Facebook obviously doesn’t share its raw customer data; that remains safely locked away on their own infrastructure. Its data is being anonymised for consumption in the IBM marketing cloud. By combining Facebook’s ad technologies with IBM’s Journey Analytics, brands can more accurately determine which groups of customers are among Facebook users and establish correlations in aggregate between their interests and interactions across multiple channels. These insights can then be brought to life through IBM’s Journey Designer solution, guiding brands to deliver more compelling messages on Facebook and other mediums.
So this is a collaborative process involving Facebook, IBM and the hundreds of brands that use the IBM marketing cloud. Both Facebook and the brands have sensitive information that somehow has to be brought together and analysed in order to achieve commercially valuable results.
It is clear that this kind of data collaboration is going to become an increasingly common practice and it is hard to conceive of how it could be done on anything except a shared but fully secure cloud service platform such as the IBM marketing cloud. Or as IDC puts it, hybrid on/off premise big data and analytics deployments will become a requirement.
Brian Groen, Cloud Business Executive