Move from Excel to Python with Pandas Transcripts
Chapter: Welcome to the course
Lecture: Where do Excel errors come from?
0:00 Excel is widely used for many tasks.
0:02 Why does it lead to big and small errors and complaints from users?
0:06 I think part of the reason is that Excel is the only tool.
0:09 Many analysts know. And they tried to use it for all their tasks.
0:13 The quote, If your only tool is a hammer then every problem looks like a
0:17 nail is very appropriate in this context.
0:20 If you are an Excel user,
0:21 you probably used it for many things.
0:23 A simple database, a task tracker,
0:26 Project planner, Quick calculations complex modeling data cleaning you name it.
0:31 Some of these tasks are great for Excel,
0:33 others not so much because Excel can only do so much.
0:36 People try to use it for everything data related.
0:39 Instead of alternative options like python,
0:41 that might be more appropriate for the task.
0:45 The next problem with excel is that all the flexibility makes it difficult to trace the flow
0:49 of a spreadsheet Excel. You can have formulas that reference other spreadsheets or have complex
0:54 nested functions and visual basic behind the scenes.
0:57 The simple representation of a spreadsheet hides a lot of complexity that goes on behind the
1:02 scenes. Another related problem is that Excel formulas make it very easy to make subtle
1:07 mistakes. They're difficult to detect.
1:10 We already mentioned data formatting is one example.
1:13 I can almost guarantee you have had issues where you forgot to include the correct range,
1:16 no formula, Or maybe your data included zeros or blank values that gave you
1:21 erroneous results. Buying these types of errors can be very challenging.
1:26 The final issue with excels performance,
1:28 especially on large data sets. In addition to managing Data,
1:32 Excel stores information about formatting formulas and logic to analyze the data.
1:37 This overhead makes opening a large spreadsheet,
1:39 a slow process with modern computers.
1:42 We have gigabytes of memory available,
1:44 so this shouldn't be a limitation.