Much has been written about Data Science and Analytics, but at the end of the day, there is a single reason why Data Analytics has staying power, and continues to attract investment across all industries -and that is its compelling ROI.

Yes, industries are awash with a lot of Data, but that is no reason to do anything about it. It is not even the insights this Data possesses that is propelling investment, but instead, quite plainly, it is its monetization potential that is attracting sustained attention.

Let us do a quick & dirty back of the envelope ROI calculation:

Assume that a distributor spends$100M annually moving goods from place A to place B. Being a fan of IoT, he collects a lot of data that covers each touch point in the journey. He hires (or outsources) some data analysts/modelers who build a recommendation engine for him that eventually saves him 0.5% of his cost (an uplift of 0.5%). This translates to$500K in savings annually. His investment in the modelers, recommendation engine and changed business process is$500K in the first year, and$150K on an annual basis thereafter.

Assuming, over a 3 year period:

Returns =$1M (benefits realized in Years 2 & 3)

Costs =$800k ($500K year 1;$300K in Years 2&3)

This would mean on a 3 year basis: ROI (3 years) = (1.0M-0.8M)/0.8M = 25%

And this number will only improve in subsequent years.

I assumed an uplift of only 0.5% because spaces like distribution are fairly well optimized already. However, there are many spaces in various industries where the uplift tends to be in the double-digit range (15%-20%), and if the spend amount is large, the ROI routinely tops 100%.

This then lies at the heart of the appeal of Analytics, and its potential to significantly raise returns or reduce spend in a range of spaces.

Analytics projects however must bear certain characteristics to realize these returns:

a) Good quality data is/can be collected

b) The overall spend on those processes/areas is already significant (~$10M+)

c) The space has not seen a lot of investment in prior optimization projects (leaving plenty of room for upside)

d) Change Management: The business is ready to make changes in alignment with the data

e) A cycle of create/test/refine is applied towards modeling before a complete roll-out is effected

f) Results are tracked carefully

These are some of the reasons why those experienced in the art of driving positive ROIs from Analytics projects highly recommend that Analytics be applied towards the heart of the business, for that is where the  most significant changes will be seen.

Here is a quick viewing on getting started with successful Analytics in an org:

https://www.youtube.com/watch?v=SVwjtabLAFM