Sambasivan G  

Having been in corporate life for over 35 years, there is one thing that I have learnt. In order to survive you need to continuously reinvent yourself……

 From manual accounting and auditing (when I started my articleship days in 1984) where tallying the trial balance was a great milestone for the company, to now, wherein you are deciding between Agentic AI and Intelligent automation, the world of finance has changed forever.

 I have travelled this journey with the help of my curiosity and continuous learning appetite. That is for another day.

 Today I like to talk about how you as a finance leader can decide on the projects/tasks wherein you can start using AI in a meaningful way.

 There are certain preconditions through this process. Firstly, To ensure operational efficiency, start by simplifying and streamlining your processes. Establish automated controls and monitor internal audit findings to confirm there are no significant deviations. Centralize your data to create a single source of truth and embed this as part of your organizational culture. Finally, ensure your team is skilled and up to date with the latest technologies and their applications.

 Having done all these (it could take anywhere between three to five years – it took me ten) to reach this stage when you can start working on implementing AI based solutions.

 Identify the problem you are trying to solve; today there are ‘n’ number of AI toolkits that are available; You may need to reimagine your entire process keeping in mind these toolkits; At every stage of the process, you can think of the solution that can be implemented. For example, if you want to use Agentic AI, identify the set of segmented purchases (could be tail end values) that you would like to experiment on.

 Then you can approach an expert (typically an outsourced agency) to develop a customised solution. It may take a few months as things are being done the first time. Before going live, please make sure that the UAT (user acceptance test) is as foolproof as possible.

 Then you implement the solution in a limited way and look for glitches; once you are confident, you can expand the same to the larger population.

 I can tell you that many a time, you may not be able to accurately calculate the ROI of such solutions; in such cases I use a mental concept called “LOF”, which is leap of faith; in general, many innovative solutions require some bit of leap of faith before you actually start reaping the results; So far, the results have not disappointed and the solutions that I have been able to implement always delivered ROI beyond my expectations.

 Regards

Samba