Trending Content
Career Resources
WSO Free Modeling Series - Now Open Through
April 30, 2024
9 Free Financial Modeling Lessons
Get instant access to lessons taught by experienced private equity pros and bulge bracket investment bankers including financial statement modeling, DCF, M&A, LBO, Comps and Excel Modeling.
+Bonus: Get 27 financial modeling templates in swipe file
+Bonus: Get 27 financial modeling templates in swipe file
I personally like to do a couple of readthroughs of something I produce, looking at various things each time. For example: - Pass 1 = formatting errors (footnotes, page numbers, double spaces, colours, alignment etc) (easy win) - Pass 2 = read the words and see if they make sense (typos, grammar, readability etc) (easy win) - Pass 3 = check all the numbers / and charts and make sure everything looks right conceptually - Pass 4 = do the numbers and the text link up together
Thank you! What do you mean by your 4th pass, make sure numbers and text line up?
For example if you changed your EBITDA on the first slide, make sure all the other slides have the figure changed too. Can be easy to miss out changing the figures on other slides and your numbers won't match
Perhaps that the numbers make sense with the text - e.g. saying about revenue growth due to increased market share but showing numbers for cost reduction
I mean if you're talking about a certain number or growth rate etc, make sure it's consistent throughout all the text or in any tables. As others have said, when you do a model update or change your comps, its easy to miss numbers in the text on other pages which should also be updated.
The 4th pass could also be referring to:
Making sure that if the graph shows margin compression, your text doesn't say that the company had margin expansion, etc.
Making sure you're biz description doesn't call the company a market leader if your market share graph clearly shows that the company is not a leader. Or calling the company a global player when you have a sales by geography chart that clearly shows that 98% of sales are from the U.S., etc.
All of this is great advice. One thing I have as well is a list of everything I need to check, such as not having oxford commas, making sure United States is always abbreviated U.S., etc.
What's up with the oxford comma thing? I've always used them my whole life while writing, and ever since I entered IB I've noticed that they are conspicuously omitted. Never asked why though, just followed suit.
This is a great system - wish I'd thought of a systematic way like this! +SB
If your attention to detail is lacking then you need to identify all the all areas where you are missing things.
The worst place to keep track of that is in your head, you will forget and you will have selective memory or "write-off" certain mistakes as one-time.
Keep a running list of every time you miss something and either 1) realize it later, or [worse] 2) get called out on it. That way when you are reviewing your work you have a written list of all the areas you need to double check before submitting.
What will work 100% of the time is to build a running Excel list of things to check, and expand it every time you get a comment/turn that you miss. It will grow and grow, but you can use it as a master list to check through rather than rely on your memory. e.g. in each cell something like "double spaces" ; "check footnote numbers tie to footnotes" ; [ and ] ; currency conventions (USD USD$ $; CAD C$ CA$ etc...) and so on, you will have at least 100+ things to check.
Great question Saw23! Definitely something I've tried to focus on more lately. Severely underestimated printing everything out when I started! If I have time, I try to work on something else and come back to whatever I was working on. Helps get my head out of the grind for a minute and allows me to look at something more high level.
More mistakes happen when you are rushed, even more mistakes happen when you are rushed AND tired. Be careful with creating / editing the decks. Don't forget the formatting for numbers, %'s, $ signs, etc.
One thing that really helps me is keeping a notebook where I record every mistake I made. I then spend some time once a week going through all the mistakes that I made that week and figuring out ways how to never make them again. In order for this to work, you have to be religious about it and be brutally honest with yourself. Making mistakes is understandable, but making the same mistake over and over again is unforgivable and will get you fired.
I like that idea. So you physically write into an actual notebook?
My memory is better with hard copies than CPU.
I do. It might sound silly, but it really helps avoiding making the same mistake twice.
This helped me immensely. I was making too many errors until I 1) slowed down and 2) wrote down every error I made.
Slowing down is key. What good is finishinf a task quickly if it is ridden with errors?
I think it's important to wait for a while before delivering a final product to your manager (or whoever is reviewing your work). I try to get things done with plenty of time to spare, then go focus on something else for a few hours. Go back to your original project with fresh eyes and you'll catch errors you would never have seen otherwise.
Don't forget the big picture. It is definitely not okay to have inconsistent numbers in a presentation, but it's way worse to have the numbers all the same, but equally wrong. Most investment banking math is very simple. You can - and you should - be able to replicate your model on a clean piece of printer paper, with a pencil and calculator. Always do this. It's not cool if your model shows $0.39 Accretive on page 5 and $0.41 on page 36, but it's way worse when the actual answer is ($0.05) dilutive. And confirmation bias will cause you to fix the former but ignore / explain away the latter. I've seen it a thousand times.
saw23 I'll give you slightly different advice which should be applicable for just about everyone (in no particular order):
1.) After completing an intensive analysis, be it in Excel or PowerPoint take a break and go for a quick walk or just look away from the screen and give your mind and eyes a rest.
2.) Print out your spreadsheet, PPT, or other document and use a highlighter and or pen to go over every cell on slide and read aloud.
3.) Have a colleague glance through and get their feedback. This works well for PPT slides as someone unfamiliar with your work will be able to spot mistakes faster than you can.
Key is to read every line for PPT slides and verify formulas and links when dealing with spreadsheets. Think about common mistakes you've made in the past and focus on those. You also need to take a step back because often you're mind will gloss over things simply because you've worked on a particular analysis for so long its become repetitive.
On top of keeping a notebook with the mistakes you have made, write down which of your managers like/dislike certain things. Obviously you always want your work to be perfect but some managers will always be more anal than others and keeping track of that will make your life easier.
My biggest tip: Care more about getting it done right instead of getting it done as quickly as you can. I feel like a lot of errors occur because of people rushing to complete decks and cycling changes, When your director is breathing down your neck, that is tough to do at times.
PDF your PPT before you check it. have this PDF open while you go through the hardcopy. If you see a number on the hardcopy that should show up multiple places you can search for the number in the PDF version....if it doesn't show in all the right places you know something is wrong
Not applicable for every deck but in 100 page books with multi-scenario outputs on a summary page this can save a lot of time spent page flipping. Also applicable with the prior version so you can see if the numbers moved the right direction
On a related note: search the pdf for the names and tickers of all your other clients, or at least the ones you previously used the slides or models for. Those artifacts are common, and control-f in powerpoint won't pick up text in embedded images.
Make sure there are consistently two spaces after every period. I've seen firms do both, but the exercise I was taught was putting two spaces after every period, because it forces you to read through everything carefully and pay attention to nitty detail.
I suppose you may have to alter this if you're working on something internal and your firm consistently only does one space after every period, but I do frequently see a discrepancy between number of spaces after the period if the document has gone through several turns with different team members or companies. Same goes for Oxford comma consistency.
Always, always, always print out your model. You need to check everything, but one thing to focus on is any step-function changes in any line item. Most companies increase revenue on a somewhat gradual, or at least linear basis, and everything else in the financial statements flows from there. If your cash balance, fcf, eps, SH equity, AP, or anything else doubles, gets cut in half, or otherwise changes in a way that is inconsistent with the top line, you've probably got a modeling error. But even if you don't, you will need to explain the anomaly in very clear language, so you better understand it.
Totally being serious, try meditating for sometime every morning. Definitely helped me focus better.
Working in a non-banking role for sometime, you realize most people's attention to detail comparatively sucks donkey nuts. Banking is tough and demands perfection, but don't be TOO hard on yourself. Good work alone gets you there as a junior, but slightly worse work + good relationships does too.
Advice to your question: There isn't really a magic formula or method to get better at it, it just takes time and extra energy to make sure it all looks good.
But I usually look at it once on screen, once on paper, and once again on pdf.
Then pretend like you are your associate reviewing someone else's work. If you still missed something after that, sometimes it just happens, remember what the mistake was so you see where you are prone to slip up.
also the refreshing of charts/tables relating to macro or operational data once a change has been made, especially minor. In big decks it can be easy to just overlook some items considered already checked
just found this thread. i can own a model but i can't process simple tasks without making mistakes. good advice
Great post, thank you for the advice! Just wondering - if you have to deal with a crazy large model - >10k rows, columns that go to beyond CC, etc, how do you check for mistakes? It'll be crazy to pdf it and print it out. Any tips?
Yeah, you can insert little checker cells along the way. If you really want, you can put in checker cells with conditional formatting, lighting up red when something is off or green if it checks.
95% of the average model, no matter how complicated, is a recreation of 3 statements, whose basic assumptions (growth, margins, cashflow) are not in dispute. The EPS in your standalone model are going to match management's estimates or the Wall Street consensus.
What you do from there - an M&A deal, financing, both, neither, etc - can be sketched out on a single piece of printer paper, with a pencil and a calculator. You should always do this, both as a check to your more detailed calculations, and to provide you with a shorthand explanation for what they show. Start with standalone EPS (or EBITA), then add or subtract any or all of the following adjustments, tax affected: EBITA (in the case of an acquisition), interest expense, amort, and shares.
Thanks guys for the tips! Well.. I'm in a slightly more... niche group. I don't actually do 3 statement models often. My models consist of more... portfolio based models and can't be done on an A4 paper or done with conditional formatting. Imagine a REIT's portfolio. Rather than look at the 3 statement model, we structure every single mortgage within the portfolio instead. It's so numbers heavy, that an excel file can go up to 100MB purely with numbers alone. And yes, it crashes a lot because of that. My problem is, I don't know which errors to look for if I don't know which errors to look for. Unless I'm very experienced, I won't know the numbers look off because they never look that way before, or which numbers should look different from last month's. I have no doubt that after I've done this role for the next 2-3 years, I'll know what to check for (which can be said about any role), but before that, the numbers don't look familiar and I create unnecessary errors :(
Say for example, it's normal to have mortgage payment rates around x% with a variation of perhaps +/- 0.5%. This is applicable to tenor of y% who cannot have a variation of above 5%. Maybe a bit messy in explanation, but what I'm getting at is that these numbers are 'market conventions' and not exactly something someone who's just started would know to check for unless I check every single cell across x10+ sheets. Plausible to do it within the week, but not if I'm only given half a day to model the whole thing.
Perhaps checker cells will help though! More tips are also appreciated!
Voluptas ut sit et deleniti quod recusandae nam. Ipsa velit officia dolor architecto voluptatem. Est soluta officiis voluptas aut alias et veritatis. Tempora voluptatem dignissimos in.
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...
Esse quisquam autem ut non. Doloremque sit vel error facilis sequi dolorem. Sit architecto sequi doloremque sunt. Eum consequatur amet nisi fugiat consequatur.
Impedit corporis molestiae repudiandae minus. Dolorem ut placeat libero provident velit. Ad incidunt blanditiis atque quo modi. Harum deleniti hic non animi commodi voluptatum. Aliquam iure molestiae doloribus asperiores rerum voluptatum aliquam.
Exercitationem molestiae quas accusantium ex ea. Assumenda est quia aut ipsa eos. Blanditiis porro tempore minima inventore perspiciatis. Optio cum velit dicta ab ullam fugiat.