Part 3 in a 5 part series from Sukumar Rajgopal, CIO at Cognizant. The kick off column was here. Part 2 is here and here is Part 4 and Part 5
As described in the previous post, our goal was to produce a OneClick timesheet. This proved to be a tough challenge because the foundational software was a third-party application. But our team persisted. We brought in experts from various parts of Cognizant and solved the problem.
Now we had the germ of the idea proven. But where do we begin? There were so many application systems on myriad technology platforms. One of our teammates suggested that we do an ethnographic (or anthropological) study to understand our users better. That was a great idea. We pursued it.
We conducted this study by identifying the top 25 tasks that associates perform using corporate IT applications. We decided that the minimum viable product should be to OneClick-enable the top 25 tasks.
Is this sufficient to produce associate delight? We went over the five major problems, which we outlined in the first post, again and decided that we needed to a few constraints to guide our designers. We came up with five that would enable us to produce delightful designs:
1. 500 millisec response time for the applications.
2. 500% productivity gain from the previous way of performing tasks.
3. No change management effort – no mail blasts, no marketing campaigns, no training/elearning, etc.
4. No email alerts from the platform.
5. Apps should be rolled out in a few weeks instead of the typical several months or years.
I am sure you see that all five constraints are “seemingly impossible.” When we first mentioned the 500% productivity gain, the response was eye-rolling skepticism. Though the OneClick timesheet proof-of-concept did improve productivity significantly, the team found it hard to accept that 500% gain was feasible or even attainable.
We needed some inspiration. We scoured the history of software development, history being one of our favorite sources of inspiration. We struck pay dirt with CDDB . Understanding CDDB’s brilliant solution is key to understanding how to produce 500%-plus productivity gain and it was a very simple idea. CDDB’s designers figured out:
1. A unique signature for a CD can be created by concatenating the track number and track length of all the tracks on the CD.
2. When a CD is ripped for the first time, the ripper enters the title, composer, singer information, which is stored in CDDB with the CD’s unique signature.
3. The next time someone else rips the same CD, CDDB furnishes the title, composer info which means the second and Nth ripper of the same CD need not key in the information.
4. Combine this with the fact that the music industry is hits-driven, so the CD collections of people have remarkable similarity.
5.Since title and composer information is stored in the database after the first instance is supplied, productivity gain, for mp3 music listeners, is incalculably high.
This design of CDDB, we believe, is the essence of social design. In case you are not impressed by CDDB, consider the fact that it was sold for $260MM to Sony in 2008 .
CDDB inspired us sufficiently to build apps that delivered similarly significant productivity gains.
Over the next few months, the team built out the MVP for OneCognizant and we rolled it out to a few thousand select associates. We were in stealth mode, another lean startup technique. Since there was no launch mailer or any marketing communication, only those few thousand associates knew of the existence of OneCognizant.
Feedback was slowly trickling in and it was quite positive overall. But we hit a major roadblock that threatened our 500msec goal which we will cover in the next installment of this series tomorrow.
