How to avoid making silly mistakes

I work in a lean staffed investment firm, I work directly for one of the MDs. I work a lot of hours.

I keep mistaking silly makes on my excel models and because there is really no one else to check my work, the MD chewed me out a few times.

My question is, how do you guys avoid making these silly mistakes? like adding an extra digit to a number or making a stupid grammar mistake?

 

Two things that help me a lot: 1- Print it and read the numbers out loud in a different language (I do understand this is not an option for some) 2- Take a break, do something else and return to it. Even a 5 min break from looking at the same thing helps.

I may not be on the Jedi Council, but I sure am great with the Force. See my WSO blog posts
 
Best Response

Printing out things and looking them over is good. However, if the buck stops with you (as is frequently the case once you're out of the analyst banking environment) you need to develop a few more error proofing tools. I've been thinking about how to do this better myself a lot lately. It's an interesting issue when you really think about it - if you put together a model with 1,000 calculations and you have a 99% accuracy rate (which is very high), you'll still have 10 errors in your model. This could obviously be very bad depending on what the errors are. So how do you prevent this? Here are a few observations from me - would be interested to hear other techniques:

(1) Focus on what you are doing. Sounds obvious but you'd be surprised how often you aren't 100% focused on the task at hand. Instead you're modeling or writing rather mindlessly while simultaneously checking email, fielding phone calls, or surfing the web. Work in blocks of 1 or 2 hours where you don't respond to emails or fuck around. At then end of those blocks respond to email or take a break, then get back to it.

(2) Lay out in detail what you plan to put together before you fire up Excel and start modeling. Think about what you are trying to accomplish overall, where the pitfalls may be, where some logical complexities might reside, how the analysis may change over time (and what functionality you need to incorporate as a result). This separates the planning from the execution of modeling, which allows you to focus on one thing at a time. When your brain is able to focus on one thing at a time, you generally make fewer mistakes.

(3) Accept that you are not going to be perfect and will make mistakes. Some people say if you make a mistake your MD / VP / Partner / group will not trust you again and your career will be over. These are not good people to listen to. If you are the one putting together your analysis and also checking it, it's unrealistic to assume you won't make any mistakes. Accept this and stop worrying about it. You'll find the less you worry about this the fewer mistakes you will make.

(4) Build-in checks to all of your models. Calculate important, headline numbers multiple ways to make sure everything pencils. Put the check outputs in an easily recognized form (i.e. "OK" or "0.000") where a quick glance lets you know if something's amiss.

(5) Check the punchline numbers back-of-the envelope style. Think about it as if you had to explain the calculation in a meeting. That will give you a firm grasp on the logic and expose errors that tend to get lost among thousands of Excel cells.

 
labanker:
Printing out things and looking them over is good. However, if the buck stops with you (as is frequently the case once you're out of the analyst banking environment) you need to develop a few more error proofing tools. I've been thinking about how to do this better myself a lot lately. It's an interesting issue when you really think about it - if you put together a model with 1,000 calculations and you have a 99% accuracy rate (which is very high), you'll still have 10 errors in your model. This could obviously be very bad depending on what the errors are. So how do you prevent this? Here are a few observations from me - would be interested to hear other techniques:

(1) Focus on what you are doing. Sounds obvious but you'd be surprised how often you aren't 100% focused on the task at hand. Instead you're modeling or writing rather mindlessly while simultaneously checking email, fielding phone calls, or surfing the web. Work in blocks of 1 or 2 hours where you don't respond to emails or fuck around. At then end of those blocks respond to email or take a break, then get back to it.

(2) Lay out in detail what you plan to put together before you fire up Excel and start modeling. Think about what you are trying to accomplish overall, where the pitfalls may be, where some logical complexities might reside, how the analysis may change over time (and what functionality you need to incorporate as a result). This separates the planning from the execution of modeling, which allows you to focus on one thing at a time. When your brain is able to focus on one thing at a time, you generally make fewer mistakes.

(3) Accept that you are not going to be perfect and will make mistakes. Some people say if you make a mistake your MD / VP / Partner / group will not trust you again and your career will be over. These are not good people to listen to. If you are the one putting together your analysis and also checking it, it's unrealistic to assume you won't make any mistakes. Accept this and stop worrying about it. You'll find the less you worry about this the fewer mistakes you will make.

(4) Build-in checks to all of your models. Calculate important, headline numbers multiple ways to make sure everything pencils. Put the check outputs in an easily recognized form (i.e. "OK" or "0.000") where a quick glance lets you know if something's amiss.

(5) Check the punchline numbers back-of-the envelope style. Think about it as if you had to explain the calculation in a meeting. That will give you a firm grasp on the logic and expose errors that tend to get lost among thousands of Excel cells.

Thanks. SB'd.

 
labanker:
Printing out things and looking them over is good. However, if the buck stops with you (as is frequently the case once you're out of the analyst banking environment) you need to develop a few more error proofing tools. I've been thinking about how to do this better myself a lot lately. It's an interesting issue when you really think about it - if you put together a model with 1,000 calculations and you have a 99% accuracy rate (which is very high), you'll still have 10 errors in your model. This could obviously be very bad depending on what the errors are. So how do you prevent this? Here are a few observations from me - would be interested to hear other techniques:

(1) Focus on what you are doing. Sounds obvious but you'd be surprised how often you aren't 100% focused on the task at hand. Instead you're modeling or writing rather mindlessly while simultaneously checking email, fielding phone calls, or surfing the web. Work in blocks of 1 or 2 hours where you don't respond to emails or fuck around. At then end of those blocks respond to email or take a break, then get back to it.

(2) Lay out in detail what you plan to put together before you fire up Excel and start modeling. Think about what you are trying to accomplish overall, where the pitfalls may be, where some logical complexities might reside, how the analysis may change over time (and what functionality you need to incorporate as a result). This separates the planning from the execution of modeling, which allows you to focus on one thing at a time. When your brain is able to focus on one thing at a time, you generally make fewer mistakes.

(3) Accept that you are not going to be perfect and will make mistakes. Some people say if you make a mistake your MD / VP / Partner / group will not trust you again and your career will be over. These are not good people to listen to. If you are the one putting together your analysis and also checking it, it's unrealistic to assume you won't make any mistakes. Accept this and stop worrying about it. You'll find the less you worry about this the fewer mistakes you will make.

(4) Build-in checks to all of your models. Calculate important, headline numbers multiple ways to make sure everything pencils. Put the check outputs in an easily recognized form (i.e. "OK" or "0.000") where a quick glance lets you know if something's amiss.

(5) Check the punchline numbers back-of-the envelope style. Think about it as if you had to explain the calculation in a meeting. That will give you a firm grasp on the logic and expose errors that tend to get lost among thousands of Excel cells.

Awesome post if I could I would SB you

As far as building in checks are concerned, I guess you could easily flag stuff like A =/= L+E etc. Do you have a list of all the checks that you typically build into your models?

Simplicity is the highest form of sophistication ~ Leonardo da Vinci
 

Take a deep breath and go through everything. Everyone makes mistakes.

"If they can't handle you at your worst, maybe they don't deserve you at your best." HAHAHAHAHAHAHAAHAH Ok that was pretty funny - I read that off some female's facebook. The dumb things they say.

 

When summarizing an analysis in email form, I have found it very helpful to email it to myself in advance and see if anything looks funny. Surprisingly effective - things just look different once in your inbox.

 

tricky situation, the problem is once you have seen a slide/excel file too many times you can't spot your own mistakes any more. Some hints:

1) Pay attention to what the MD pays attention to, in what areas does he find mistakes? Then check those areas intensively

2) Always print out

3) Review your own work as if you were reviewing someone elses and trying to screw them as much as possible (this one helps alot)

 
DeanPortman:
print it out and read it while you take a dump...i'll be right there next to you proofreading as well

Well Done...I'm not even made, I'm impressed

Eventus stultorum magister.
 
DeanPortman:
print it out and read it while you take a dump...i'll be right there next to you proofreading as well

Well Done...I'm not even made, I'm impressed

Eventus stultorum magister.
 
DeanPortman:
print it out and read it while you take a dump...i'll be right there next to you proofreading as well

Well Done...I'm not even made, I'm impressed

Eventus stultorum magister.
 

1) Financial statements - check your cash flow numbers match the company's (given you will be building your model in a different way to how the company laid it out). If not, analyse and find where differences lie.

2) Balance sheet check - Assets / Liabilities + Equity = 100.00000000% (I always go to a ridiculous amount of decimal places)

3) Your valuation makes sense. Look at comps EBITDA, EV, and multiples, and compare to yours

4) Make sure the capital structure and other debt add up to what company reports as total debt on the balance sheet

5) Avoid using plugs in the model e.g. you know total Opex was 300m. You break out two items (X - 78m, Y - 120m, and make Z = 300-(X+Y). I try to avoid using plugs like Z as if there's an error in what you input as X or Y, it won't be obvious when you're checking the model. If you add up the numbers for Z individually (even if just in the same cell), then any mistakes on X or Y will feed through to cash flow when you notice your net cash flow is different to what the company reported.

6) Rest is qualitative stuff about the business which you can only check by doing your diligence properly

I rarely find mistakes in #1-5. Where I do make mistakes, are in my judgement of a business' key value drivers and value proposition. But that comes with trying to learn a new industry every other week. When you're in an investment seat, you'll realise how little 1-5 really matter (an obvious mistake should be picked up by who you're pitching it to, assuming you're not sole PM and analyst). #6 is where money will be made and lost.

 

Great post from New Guy.

I didn't have problems with the financials during my internship, but really understanding the business was not easy.

Take your time the get to know the market, the players and the company.

I always used to print a lot as well, even if it was just to see whether formatting was correct.

 
suchislife:
And for grammar mistakes don't be naive!

RULE OF THUMB #4: Always use F-7 (a.k.a. excel spellcheck)!!

If you don't typos can sneak in given that excel does not mark your typos like word does!

It's spellcheck in everything dude; outlook etc...
"After you work on Wall Street it’s a choice, would you rather work at McDonalds or on the sell-side? I would choose McDonalds over the sell-side.” - David Tepper
 

Printing out beforehand and stepping away from your work for a bit are both helpful.

Also, you should hard code as little as possible and use formulas for everything beyond whatever data you originally entered.

 

Aliquam ea quia quam harum corrupti ab occaecati impedit. Est fugiat placeat voluptate quia vel. Et esse enim sit aspernatur ut dolores.

Voluptatem debitis ab libero quia pariatur. Dignissimos sed architecto sapiente non qui. Voluptas nihil cum ratione. Cumque quas totam quae veniam rerum. Numquam aperiam alias occaecati quia fuga sed.

Career Advancement Opportunities

March 2024 Investment Banking

  • Jefferies & Company 02 99.4%
  • Goldman Sachs 19 98.8%
  • Harris Williams & Co. (++) 98.3%
  • Lazard Freres 02 97.7%
  • JPMorgan Chase 03 97.1%

Overall Employee Satisfaction

March 2024 Investment Banking

  • Harris Williams & Co. 18 99.4%
  • JPMorgan Chase 10 98.8%
  • Lazard Freres 05 98.3%
  • Morgan Stanley 07 97.7%
  • William Blair 03 97.1%

Professional Growth Opportunities

March 2024 Investment Banking

  • Lazard Freres 01 99.4%
  • Jefferies & Company 02 98.8%
  • Goldman Sachs 17 98.3%
  • Moelis & Company 07 97.7%
  • JPMorgan Chase 05 97.1%

Total Avg Compensation

March 2024 Investment Banking

  • Director/MD (5) $648
  • Vice President (19) $385
  • Associates (86) $261
  • 3rd+ Year Analyst (13) $181
  • Intern/Summer Associate (33) $170
  • 2nd Year Analyst (66) $168
  • 1st Year Analyst (202) $159
  • Intern/Summer Analyst (144) $101
notes
16 IB Interviews Notes

“... there’s no excuse to not take advantage of the resources out there available to you. Best value for your $ are the...”

Leaderboard

1
redever's picture
redever
99.2
2
Secyh62's picture
Secyh62
99.0
3
BankonBanking's picture
BankonBanking
99.0
4
Betsy Massar's picture
Betsy Massar
99.0
5
kanon's picture
kanon
98.9
6
dosk17's picture
dosk17
98.9
7
GameTheory's picture
GameTheory
98.9
8
CompBanker's picture
CompBanker
98.9
9
DrApeman's picture
DrApeman
98.9
10
Jamoldo's picture
Jamoldo
98.8
success
From 10 rejections to 1 dream investment banking internship

“... I believe it was the single biggest reason why I ended up with an offer...”