00:29 What is the role of Microsoft Excel in the supply chain industry?
02:19 Which are the key strengths of Excel?
04:43 What we are seeing in the industry is that these basic approximations are good enough and have been running on for decades. Is that right?
10:02 Is the replicated logic the reason behind it?
11:39 How can you move away from these spreadsheets?
13:59 What can we learn from what the Googles and Amazons have implemented?
17:10 What would you say to the sceptical supply chain practitioner? Is there really an incentive to move away from Microsoft Excel?
Forecasting demand and optimizing stocks are both usually done in Excel, even in large companies. Many CIOs have tried to move away from spreadsheets to no avail. Yet, most criticisms addressed to Excel and spreadsheets in general are incorrect. As a result, intended software replacements for Excel miss the point.
Microsoft Excel is a staple in the supply chain industry, used by everyone from the smallest retailers to the largest multinationals. Despite this, over the last decade many CIOs have made it their main aim to get rid of it in favour of more modern, scalable tools. In this episode of LokadTV, we discuss why so many CIOs have reached this decision yet Excel is still preferred by the vast majority of organizations.
You could say that Excel is the “swiss knife of supply chain”, due to its ubiquitousness and the fact that it’s used everywhere for practically everything. Why is it so widely used? For a start, there have been very few credible alternatives. Excel is an excellent tool and two of the major strengths that render it very powerful is its programming ability and the level of expressiveness that comes with this system, plus the fact that it can be heavily distributed throughout an organization. For example, supply chain practitioners managing different product lines in different locations around the world are able to craft their own heuristics.
Heuristics are numerical recipes that are only approximatively correct. An example of a supply chain heuristic could be keeping an amount of stock that equals twice the amount of units sold a year ago, at the same period, considering a window of 3 months. Why two times the stock? Why last year? Why a 3-month window? Heuristics are recipes that have been “battle tested” but aren’t strictly mathematically founded. But seeing as these heuristics are tried and tested and deemed “good enough”, supply chain practitioners aren’t motivated to change their methods, especially after using them for so many years.
However, Excel could be seen as a technological dead-end for supply chains, because Excel - and all spreadsheets in general - have a scalability issue. This is linked to their programming model, where a piece of logic is copy and pasted across the spreadsheet, thus from a programming perspective its a massive logic replication. For a spreadsheet with a few hundred products and only 2 or 3 heuristics, this works well. However, for a larger scope with 20 heuristics, the spreadsheet begins to have important complexity issues. Then if you try and go up to 100 heuristics and this is replicated throughout your entire organization, problems definitely start to occur. These overly replicated logics then render the spreadsheets clunky and slow.
There is also the important question of how to maintain a sheet with 100’s of different formulas, millions of lines with different segments and different heuristics that become almost impossible to consolidate. This is where the scalability problems lie, not on the data processing side, but on the complexity side. Spreadsheets do not handle increased complexity well. Therefore, debugging, maintenance and even just understanding what is going on within the spreadsheet becomes extremely difficult.
Yet, these spreadsheets contain years of information and structuring by organizations, so how can you move away from that? What are digital leaders like Google and Amazon doing? To conclude this episode, we discuss in more detail these crucial questions and more.