Sense checks for models (project finance/leverage/gearing)
When modelling, what are some sense checks or interpretations you do for your model to check everything is correct?
for example:
- sources and uses of funding should be equal
- unlevered IRR should be higher than levered IRR
- the interest rate for debt financing is the rate of return required by lenders
Other than creating 3 statements and checking they all balance, what sense checks do you use while modelling? Thank you
Based on the most helpful WSO content, here are some essential sense checks and interpretations to ensure your financial model is accurate and reliable:
Balance Sheet Should Always Balance:
Sources and Uses of Funding Should Be Equal:
Unlevered IRR vs. Levered IRR:
Interest Rate for Debt Financing:
Sanity Checks on Key Metrics:
Consistency in Presentation:
Version Control:
Cross-Verification with External Data:
Detailed Review of Formulas:
Review of Key Performance Indicators (KPIs):
Working Capital Movements:
Debt and Asset Management:
Continuous Communication with Management:
By implementing these sense checks, you can significantly improve the accuracy and reliability of your financial models.
Sources: Ways of Underperformance - and how to avoid them (Part 1), What lessons you learned the hard way so everyone else doesn't have to?, How to Convince My Boss to Include Excel Modeling Into the Job?, Scared Shitless: How Did You Actually Learn How To Competently Build Financial Models, https://www.wallstreetoasis.com/forum/asset-management/qa-equity-research-analyst-at-top-3-am?customgpt=1
patriot18 I did some research on this a while back with 15 interviews trying to understand what IB/PE folks do to check their models.
Here is a link to my original post on Linked-in: Linked-in post
Attaching a snapshot of their techniques below:
1. Start with a robust model design.
a. Create a consistent layout structure across tabs.
b. Color-code numbers based on their source
c. Maintain consistent formulas across rows, this way you need to inspect only one cell per row.
d. Avoid hardcoded numbers inside formulas, create visible assumption cells instead.
2. Build check lines. Trust but verify.
As you create or update building blocks, build check lines that show if your model is accurate.
a. Framework based (such as BS balancing, interconnections between the 3 statement model).
b. Content driven (whatever it is you are modelling, try to arrive at the same result from 2 different angles).
c. Technical (when creating summary tabs, make sure that total results match per line between tabs).
3. Stress test the model by changing the assumptions.
Test logic by changing assumptions and observing how the outputs behave. This process can identify potentially weak or incorrect logic. Here are a few tricks that we learned:
a. Set all assumptions to zero to identify any anomalies in the model.
b. Gradually change each assumption by a degree and observe how certain outputs behave.
c. Evaluate if the changes make sense. Do they affect the right outputs and by the expected degree or not?
4. Build performance alerts.
a. Connect two model outputs to create meaningful KPIs that highlight unusual results. Example: financial - link revenue and customer acquisition cost to track profitability over time.
b. Set up alerts for significant period-to-period changes in key metrics. Example: financial - flag revenue growth rates above 50%; operations - alert on inventory turnover decreases of 20% or more.
c. Benchmark against industry norms to highlight performance gaps. Example: financial - compare current ratio to industry averages.
5. Create visual model maps.
Creating a visual representation of your model away from Excel helps you question your own logic, but it also invites your stakeholders to contribute on a deeper level:
a. Draw a flowchart or diagram showing the model's structure and key calculations
b. Use shapes and arrows to illustrate relationships between different parts of the model
c. Include brief explanations of critical formulas or assumptions
6. Ask a colleague for support. The four-eye principle.
A fresh perspective can identify overlooked aspects. Your colleague could help catch mistakes and challenge you where you might have considered the job done. Provide your colleague with appropriate context and let them get to work independently. Discuss issues, debate solutions, and implement changes.
7. Print and review. Jump away from the screen.
Printing your model allows you to step back from Excel and see the big picture. Physical copies make it easier to spot patterns or inconsistencies that might be missed on screen. Get a cup of coffee. Zoom out. Assess if numbers make sense together. Look for visual anomalies in numbers.
Hope this helps!
Dicta ipsum quisquam qui quod. At inventore a omnis est. Ut sunt ut nulla quasi sequi reiciendis in. Et autem dolores eum autem.
Quia praesentium possimus excepturi. Debitis placeat quia illum autem aspernatur tenetur blanditiis. Molestiae similique accusantium cupiditate. Adipisci dicta ut est.
Culpa quasi id vel labore beatae voluptas. Voluptatem id cupiditate veniam et et ut odio. Sapiente repellendus numquam non quia nesciunt.
Voluptatem et incidunt perspiciatis. Ut provident sit quam aut velit. Dicta vitae explicabo ut repellat consequuntur inventore ab. Molestiae quae non quidem sequi officiis architecto magni.
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...