PE CIM Case Studies

Here's what I do - I compartmentalize into building blocks, because reading a CIM without a clear direction of what you're looking for is a total time waste.

• Skim exec summary for 5 minutes - basically to see the industry, and to see if there's anything funky about the business model or transaction that isn't a standard LBO (minority txn? weird structure?)
• Straight to financials, and immediate work on the model. Copy and paste whatever financials are available into excel, and just build the LBO. If it's a standard PE case study, you should be aiming to do this as quickly as possible, and you can literally paste in CIM forecast, and make any standard debt assumptions. Your goal here is to lock down the LBO structure, and to basically do it as quickly and robotically as possible. Basically, you are setting up your template so that you can go and fuck around with drivers later after reading the CIM. This should take hopefully 25-40 minutes.
• Once you've done the LBO (which will probably give you a nonsensical IRR), THEN you can start thinking about the revenue and cost builds. There should be a section in the CIM regarding business model - is it # stores, is it subscription-based, some sort of volume, etc? Then, you search for the key metrics in the CIM. Many times, you have to impute some figures which is fine. For example, maybe this is a services business that fundamentally operates as # projects X $ per project. CIM usually will mention 1 or the other, so you just impute what's missing. Maybe it's more complicated then that, but you need to build in layers so that you don't run out of time trying to run some stupidly complicated revenue build (e.g., first do projects x $ per project. Then if you've done that quickly, you can split it up into X number of segments if that's important). Do the same for costs, CIM will usually give indication of what's fixed COGS vs. variable COGS vs. Operating costs (usually can just do operating costs as % of revenue and scale it down over time). Then, you add this into the revenue line and make slight adjustments so you can see what the sell-side is "implying" for whatever driver you're focused on. They also usually give a page on their key assumptions so you can test your build and make sure it's consistent. This should take another 30 minutes.
• Your model should be fundamentally done after this, and the rest is actually reading the damn CIM and developing your base case. Ideally, you have at least another 30min-1 hour for this. As you mention, good base assumptions are triangulated using: close to historical growth (if not, need clear justification), in-line with market growth, very moderate cost scaling, cashflow needs in-line with historicals, etc). Then, slap on an entry and exit multiple that makes the returns somewhat palatable. Then, you sensitize your drivers
• Not gonna go over the write-up, but you should leave yourself 20min (more if it's a PPT) to clearly articulate investment thesis and risks. For thesis, take the top 3 highlights from the CIM and re-word them from your point of view. This should be easy because bankers usually put 6+ highlights. You can include another angle too if you think it's differentiated ("rare platform in saturated space, potential for M&A, whatever bullshit). For key risks, you can put the standard BS (growing competition, sustainable moat, etc). But you need at least 1-2 key risks that show that you've thought about this. Maybe IRR is highly sensitive to one of your revenue drivers, mention this, and extrapolate what this means in the real world (e.g., sensitive to pricing --> competition and price concessions, maybe macro factors, maybe whatever "moat" they highlight in the CIM is iffy especially if there's a history of pricing volatility)