Forum Posts

Prashant Sanghal
Feb 07, 2019
In Sourcing
One of the biggest barriers to sourcing strategically is 'Just Buying' mindset. It forces us to shift our emphasis on getting things done quickly, leaving us very little time to explore innovations within and beyond our supply networks. Why is that the case? We as organizations might be dealing with time and budget constraints, competitive pressures, market difficulties and resistance to change as contributing factors. We could also be dealing with consequential risks such as loss of revenue, production delays and safety related risks escalating it to a rush type situation. Not having something which was needed yesterday is a tough problem we face every day. That broken pump, leaking pipe, missing parts and not having experienced personnel to do the job, makes us think should we plan a strategy or just get it done right away? What can be done? Portfolio Analysis might help us get started. We can divide our spend into 4 different quadrants and define labels accordingly. Example: Spend that is labelled strategic would mean enterprise wide, long term, 3rd-party contributive while critical on the other hand would mean not enough supply, difficulty switching suppliers and so on. The benefit of labelling our portfolio is that we can develop strategies in advance and decide how we chose to react when that broken pump situation arises. Who are our preferred suppliers we can go to, than relying on spot purchases. How much of the spend can be consolidated and sourced from fewer suppliers than many? This gives us the power to choose and act! Please feel free to share your comments, thoughts or just say hi.
0
1
43
Prashant Sanghal
Feb 07, 2019
In Technological Trends
Classifying organizationals spend is not easy. It takes time and effort to define boundaries and put strategies in place to maximize returns. But the question is how we do it today? Is there may be a better tool out there which can help us? Machine learning has the answer. It comes in very handy when it comes to discovering patterns in data which otherwise would be difficult to explore. How about when business conditions change daily, monthly or all the time, can we spot patterns quickly? Good thing ML algorithms can learn from data in real time and can categorize spend accurately when those changes occur. All we need to do is acquire the data, do some data preparation, explore questions we want to find, acquire the data again and apply the algorithm. It's that simple! So what is the difficult part then? Demand for Data scientists who can interpret all the data is high. It takes time to learn coding and applying it to problems, which can be the difficult part. Please feel free to share your comments, thoughts, or just say hi.
2
0
30

Prashant Sanghal

More actions