Done BIWS/WSO course - How to build modelling intuition / discernment from scratch
Hey everyone,
I’ve done the BIWS / WSO courses, I’m comfortable in Excel, I can link the 3 statements, create debt schedules, build in checks, and format models properly. But when I sit down with a blank 10-K or annual report and try to build something from scratch — I freeze.
It’s not that I don’t know how to model. It’s that I don’t know what to model.
Specifically:
- I can pull the financials and set up the structure easily.
- But I struggle with judgment — figuring out which addbacks are reasonable, which expenses to normalize, how to interpret management’s “adjusted” metrics, and what’s actually recurring.
- The hardest part is the revenue build — deciding what the real drivers are. BIWS gives you clean examples like “Volume × Price,” but when I open a real 10-K, the company gives you segments, FX effects, and product lines with almost no guidance.
- I end up second-guessing everything: am I over-complicating this? am I missing what really moves the needle?
In other words, I can execute the mechanics, but I lack the discernment - the ability to translate the business narrative into numbers and assumptions.
For those who’ve made it through this stage - how did you build that intuition? Was it just repetition and seeing enough sell-side models? Did you start by copying existing models and then stripping them down?
Any frameworks, habits, or mental models you used to develop this “business operator’s mindset” would really help.
Thanks in advance I feel like this is the stage that separates “good with Excel” from actually thinking like an analyst.
To develop the intuition and discernment needed to build financial models from scratch, here’s what the most helpful WSO content suggests:
1. Understand the Business First
2. Start with Qualitative Analysis
3. Revenue Build Framework
4. Normalize and Adjust Thoughtfully
5. Repetition and Reverse Engineering
6. Think Like an Investor
7. Practice, Practice, Practice
8. Leverage Resources
By combining these strategies—understanding the business, focusing on key drivers, practicing with real-world examples, and thinking like an investor—you’ll gradually develop the discernment and intuition needed to build models with confidence.
Sources: To All Freshman and Sophomores, Scared Shitless: How Did You Actually Learn How To Competently Build Financial Models, How to Convince My Boss to Include Excel Modeling Into the Job?, Q&A: From Sellside Research to Strategy & Finance at Rapid Growth Start-up
Wondering the same thing. How do you know where to start your revenue builds?
#1, Don't worry too much about building a model from complete 0 as that's something you'll almost never have to do. Yes, it's cool to have the skill, but if you are a bit of a junior, don't start by jumping in the deep end, build the mechanics in each section first and then when comfortable, combine them.
#2 Assumptions are the hardest parts of models, you are literally trying to predict the future which is impossible. However, there are good and bad predictions. I'd say step 1 of making predictions is to look at management guidance. Who knows the company better than... the company itself. Obviously there will be underlying motives here (is mngmnt trying to set low expectations to outperform later or are they trying to boost the share price now for a share issuance despite knowing they're having a shitter), but these usually give you a pretty decent ball park estimate. Plus you can look historically to see if management usually over or under shoots guidance and bake that in.
Next for assumptions, I like to simply apply a cagr of the last 5 yrs to the next 5-10yrs and see if that makes commercial sense. If the cagr of the prev 5 years means in the next 10 yrs they'll become the next Nike or Amazon, growth should probably drop off a bit (unless you want to say that they will reach supersonic growth).
Something else I like to do (this is hard if you're just a student but try your best) is to try and read a few ER reports on the company. See what the analysts say, what everyone agrees on and what they disagree on and then choose which one you agree on. If ER team 1, 2 and 3 all think margins will fall bc of tariffs but ER team 4 and 5 think it's being overblown, read each of their rationales and see who you agree/disagree with. Also feel free to mix and match here, you might like 1's rev projections, 2's margins, 3's Capex and 4's adjustments so just go through them and see which justification you agree with.
Lastly, don't focus too much on the small stuff (adjustments, FX effects, derivative gains/losses, other BS) as that stuff is usually inconsistent and will flip year to year anyway so in the grandscheme, it all evens out, but do spend time digging into the your big expenses. Focus on the trend of cogs being a % of rev, how and if SG&A is broken down and what goes into it, where is interest expense coming from and is it a fixed or floating rate. By ignoring the small things, you can spend more time on the actual important lines that Do matter.
Hope this helps
Thank you sir! Appreciate it.
The reason I asked about the starting from scratch is because typically LBO models for PE recruiting may ask you to build out 3SM from scratch from what I have heard.
But I found this very useful especially wrt to the trends
Ok that explains it. I'm not a PE expert so I'll let others chime in. But, from what I've seen, the type model required in PE recruiting is almost its own skillset
IMO. The comparison would be like if someone tried to maximize their overall body strength compared to someone just trying to max out bench press. Yes, benching a lot will make you stronger overall and being stronger overall will make you a better bencher, but depending on what your actual goal is you should focus on that. The tips I shared above will make you a better modeller but in a time crunch probably a bit overkill. For specific PE recruiting needs, I'd focus on strictly on building your models under time pressure with strong intuition and back of the napkin math and maybe a few scenario analysis to cover your bases opposed to the deep analytical rigor I mentioned.
How helpful did you find BIWS and WSO modelling courses? I've done some of the WST, but struggle to remember it as I am still in undergrad and not working on modelling outside my own time.
recommendationsAny reccomendations on how to build skills and have them stick?
Tbh, I think apart from just the idea of setting up a model and very big picture structuring and considerations, even after doing a few of the courses, I never found them that helpful (for reference, most of the ones I did were from the Marquee group which was acquired by Training the Street). This isn't to speak down on the teachers or the courses themselves, I just don't think I learned the best that way.
This is kinda cringe but I found the best way I learned modelling was through our school's finance club bc in junior years, I got to own basic stuff like just inputting historicals and making sure stuff linked while watching seniors do the actual analysis to then owning parts of the model in middle years to then being the driver of the model in my final year. This was also great bc in my first 2 yrs i could kinda just watch and ask stupid Q's to the seniors and felt they could explain it better as a student than a prof or specialized teacher.
In terms of retaining/building the skills. I'd say use one of the templates you have, and build a model for a firm with a super easy and understandable business model. Don't rush it, if it takes 6 months so be it. And then from that point on, every time that Co has their earnings report, update your model and assumptions. If at first you modeled out to Q4 2030, after their earnings, model out to Q1 2031 and change whatever assumption you have for the interim years you have so that it reflects the trend or guidance from earnings. Doing this will serve a few purposes. #1: You'll have a model where you should know 100% how everything flows into everything, #2: It will train arguably the more common part of modelling which is not building the model but rather updating it. #3: Going forward, you will always have a model you can include in networking emails/maybe even applications/something to speak to in interviews that is updated.
Hope this helps
Aut esse et eos aut. Ad incidunt aliquid ea dolorem. Molestias sed quibusdam odit quibusdam vel et. Ratione iusto perspiciatis soluta magni. Quae iure est nihil ducimus.
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...