Why isn’t coding more popular for modeling?
General question which I am quite curious about. Please excuse my ignorance if I’m missing something really obvious, but why isn’t modeling done more thoroughly with coding given that CS fields have advanced immensely in the finance world since the 90’s? I’d imagine there are some hurdles it would face but I’m surprised financial engineering hasn’t really made immense strides in this area given they’re starting to overtake a lot. Are there some big limitations coding would face?
For building from revenue to FCF or EPS you don’t want something super hard. The technical building of the model isn’t difficult and excel can handle it just fine. Usually the limiting factor is human error so you need every to be easier laid out in front of you.
Excel is nicely laid out and you can format it to be easily understandable vs a coding program. Also since we usually don’t deal with too large of data sets, we don’t need the extra computing efficiency of more dedicated/specialized software.
Keeping it simple and being able to easily edit/understand the flow of dollars is much more important than being able to update a model more efficiently computer wise.
Also you can’t scale a model like you can a computer program. Each model has to be unique for each company in some way so it’s not worth the extra effort to straight code a financial model for presentations.
The model being easily understandable is more important than it being complex.
Also, just shows how some things are slow to change, even in industries like banking where things move fast (as opposed to say government technology). Problem is also low level associates and analyst change all the time, so you can have an analyst who is really good at coding, but the next one might not be, then you are stuck with a heavy coding model.
adding onto the comments above, every company has its nuances so building it out yourself on excel is also a good way for you to understand the business rather than just coding it
At first my thought is “because that is too complicated for a relatively simple workflow” but then it hit me. There is this program called Orange. It is a pretty advanced data analytics tool. It gives the user the option to do a “drag and drop” style workflow using modules (in the actual Orange program modules include things like heat mapping and classification trees so in an IB specific version you’d have modules like “operating model” or “M&A model”) or you can open up the more detailed views and actually do the coding. Something like that could pretty easily be done, and probably should be done, so long as the import data tool has an account matching tool (ie - so Spirit AeroSystems “737 operating revenue” account is mapped to Boeing’s “737 construction expense” account when running the M&A module for Boeing’s acquisition of Spirit
At the company I currently work for (F50), we regularly use scripting to wrangle data / large modeling exercises. Excel still has it's place, but operations of large companies are too big not to take advantage of scripting capabilities.
In my last role at this company, I worked with some external M&A people that were advising on a deal who exclusively used excel / power point, and it was generally not insightful work / the analysis was weak, but that's not what they were really hired for anyway. I think excel shines as a tool to test ideas / do quick work, but for any meaningful analysis of a large scale organization, it's largely inadequate. I suspect the financial services providers (sell side / FDD / TAS types) will continue to lean on excel / power point since (at least in my experience), their role isn't really about analysis / understanding a business, but getting a deal done / making things look pretty.
Optio delectus quam et et. Totam facere qui ipsum omnis.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...