The usage and impact of Engineering Analytics across the product development lifecycle stages (Design, Test, Sustenance, and Support) is well understood and articulated.
How about looking at:
This blog series looks at some of the existing features/mechanisms rolled into the products we regularly use, as well as some future trends we could expect, all leveraging Engineering Analytics.
Imagine how many times you would have (and still do) sat in front of your TV, scanning/browsing the channels endlessly before settling on a content of interest? Traditional broadcast solutions (be it Cable, Satellite, or even the recent IPTV) did very little beyond offering Static Program guides to help us beat this boredom. Switch now to the alternate entertainment options (thanks to the online/broadband revolution) – be it Netflix, YouTube, or any one of the many Online Entertainment Applications. See the primary difference? Yes, they now ‘recommend’ to you, what could be of interest to you – based on a variety of ‘Analytics’ done in the backend. From ‘Popular’ content to ‘Similar’ content based on your viewing pattern, personalization is all about getting you the benefit of a typical “Engineering Analytics”.
For years, the traditional broadcast solutions have taken user feedback on viewing details (often without their knowledge) and used it for aspects like decision on popularity of content (TRP ratings), ad customization etc. Interestingly, Content Recommendation (initially perfected by Amazon for its online retail business) has now found its way into the traditional Set-top-boxes as well, thanks to the immense pressure on this “analytics-aided” feature parity with the online rivals
When we take a close look around, many such examples exist – from the frequently dialed numbers on your phone, frequently asked questions in every self-care portal, and the list keeps growing.
The question we as an engineering community should be asking now is not how Engineering Analytics is useful for us as Architects, Product Managers, Feature designers, Testers or Support engineers in carrying out our jobs more efficiently, but how do we make such analytics insights reach our end customers (consumers of the products we work on) so that they directly benefit from it.
An IP-PBX manufacturer could easily decide to prioritize the Feature work, Testing-Validation they do on over 100 variants of call forwarding based on the insights they gain from the configuration and specific instrumentation of feature usage on the System. They however, could take this a level closer to their enterprise customers by recommending (yes, I mean recommending !!) call forwarding flavors that the customer could potentially configure and use. Sample this: Existing sources of call logs, answering pattern, and times in the day the call traffic is active/unattended – could all be helpful in piecing together this recommendation, with little engineering effort on the product itself.
Similarly, there is a lot of interest today in the Connected Vehicle space – thanks to Telematics. while a lot of use cases are focused on Predictive Maintenance, Service Quality improvement, Insurance premium tied into driving patterns etc. there could potentially be a personalization angle as well. Analytics that could tell the driver (may be a sweet alert via the car audio) that they’re driving a bit unusual/rash, there’s a potential breakdown ahead, or the unavailability of gas stations in the neighborhood they’re about to drive through.
Successful accomplishment of all such use cases may not ring-in a lot of cash for the Product Vendors, but sure gives them a differentiation and a first mover advantage. After all, the data belongs to the customer – hence they should have the first-use of the Analytics benefit/advantage as well. All these use cases, as one would agree, requires a combination of the domain understanding, ecosystem knowledge, and a fair bit of data science/analytics application skill. More importantly, they also need the conviction to see the consumer benefit as a prime motive.
Given that there are a number of parallel efforts undertaken in any product company to leverage data from the products (be it during the engineering effort or the post production/deployment –from the field), the ‘User’ angle to the Engineering Analytics is often going to be seen as a redundant effort and hence face operational challenges. Unlike (Telecom, Retail etc) Service Providers, Product Manufacturers do not have a strong Digital Strategy that focuses on this aspect. Rarely, we have a structure like a Digital Office / have a Chief Digital Officer in these companies. That trend is for sure changing and a number of Tier-1 Product Manufacturers are creating a Digital Strategy and an organization structure that supports and executes this strategy.
This effort is not just another opportunity space for the Traditional ESOs / SI Partners of these organizations – it’s a litmus test for a comprehensive partnership that most claim it to be. Given that these vendors share a good amount of product knowledge and an appreciable understanding of the ecosystem where the products are deployed / used, the ability to identify, incubate, and demonstrate user-centric engineering analytics use cases would become the key differentiator. It’s a litmus test for the product organizations themselves as well, to leverage the relationship beyond a specific contract.
In the subsequent blogs, let’s look deeper at some of the technology enablers and business choices one could make for the successful accomplishment of user-centric engineering analytics.