Explore how to identify and adjust relevant M&A deals for a robust market-based valuation of private companies, focusing on the nuances of synergy potential, growth rates, and intangible assets.
Selecting comparable transactions can be a bit of a puzzle—like rummaging through a box of old sports cards and trying to figure out which cards (and their prices) most closely match the one you’re trying to trade. In the context of private company valuation, you typically don’t have a readily observable share price or public float. Instead, you rely on the prices paid in recent mergers and acquisitions (M&A) of similar companies, adjusting for factors like minority interest, control premiums, growth differentials, even intangible brand values.
Anyway, let’s walk through a vignette-style approach so you can see how this typically appears on the CFA exam. We’ll talk about the big items you need to watch for: how the data is presented, which multiples to consider, and how to make proper adjustments for synergy potential or market conditions that may differ from the target you’re analyzing.
Imagine a scenario where you’re analyzing a closely held consumer goods company—call it HomeMade Supplies, Inc. The exam vignette might lay out facts like:
• HomeMade Supplies is a manufacturer of eco-friendly household products, including cleaning solutions and personal care items.
• They have stable, but modest, annual revenue growth of 5%.
• The industry is consolidating rapidly, with larger players absorbing small to mid-sized niche manufacturers.
• You have a list of recent private-company acquisitions in the same sector. Each transaction summary might include the target’s revenue, EBITDA, growth rates, brand recognition, synergy potential, and the transaction’s implied multiples (e.g., EV/EBITDA, EV/Revenue).
Your mission: figure out which of these transactions are the best comparables for HomeMade Supplies, incorporate the relevant multiples, and arrive at a valuation range. This is typical of what you might see in an exam item set (vignette) question. Let’s explore how you peel back the layers step by step.
Selecting comparable transactions isn’t just about picking deals from the same industry. You want to focus on a handful of critical factors:
• Company Size and Growth: A large, established target with 15% revenue growth isn’t exactly comparable to a small, steady 5% growth firm—unless you adjust the multiples accordingly.
• Profit Margins: EBITDA margins can vary widely, even in the same industry. You’ll see widely different EV/EBITDA levels if you compare an 8% margin company to one at 22%.
• Geographic Market: Comparable transactions in different regions might have different regulatory environments, consumer preferences, or cost structures.
• Timing and Market Conditions: Market sentiment can fluctuate quickly. A transaction from two years ago might reflect outdated market multiples if there’s been a surge or slump in investor sentiment.
• Deal Type: Distressed sales, majority control acquisitions, or minority stake investments can all reflect different transaction dynamics. You might need to adjust for control premiums or discounts for lack of marketability accordingly.
Below is a simple flowchart that shows how you might approach selecting comparable transactions when faced with a big list in an exam or in real-world practice:
flowchart TB A["Start <br/>with all possible transactions"] --> B["Filter by <br/>industry, geography, <br/>business model"] B --> C["Review key <br/>financial ratios <br/>(margins, growth)"] C --> D["Eliminate outliers <br/>(extreme multiples, <br/>distressed deals)"] D --> E["Adjust multiples for <br/>control premiums, <br/>marketability"] E --> F["Apply final multiples to <br/>target’s normalized metrics"]
Following this flow keeps your selection process systematic. Resist the urge to include every transaction in your final analysis—sometimes you only want two or three that best match your target’s profile.
One of the first things that can trip you up is forgetting to normalize the target’s financial metrics. If your target’s EBITDA includes large one-time expenses (e.g., a legal settlement) or extraordinary gains (e.g., a sale of a non-core asset), you have to adjust these out. M&A transaction multiples often reference a target’s “steady-state” or “normalized” earnings. So you want to:
• Remove non-recurring items (unusual legal litigation expenses, etc.)
• Adjust for prospective synergy benefits if your scenario specifically includes synergy assumptions in the comparable deals
• Reflect the capital structure differences if necessary (for instance, removing unusual interest costs or minority interest)
Those adjustments let you use apples-to-apples multiples, thereby focusing on a stable operating performance rather than one-time noise in the numbers.
In many exam vignettes—and certainly in real life—the nature of the transaction can significantly influence pricing. A controlling interest acquisition often comes with a control premium, which can drive up the multiple relative to deals that involve only a minority stake. Keep an eye out for:
• Transactions that gave the acquirer a controlling stake (typically >50%).
• Adjustments for synergy potential or significant business model reconfigurations.
• The size of the control premium to reflect real corporate governance changes.
If you ignore these differences, you might overestimate or underestimate your target’s value. The exam might specifically prompt you with lines like “Acquirer X purchased a 20% stake, representing a minority, non-controlling position,” or “Investor Y acquired 80% equity interest and integrated the target’s distribution channels.” Listen (or read) carefully for those hints.
Outlier multiples—either extremely high or uncharacteristically low—can arise for many reasons: distressed sales, unusual synergy potential, extremely rapid growth, or intangible brand value well above industry norms. The exam usually wants you to recognize and exclude or at least reduce the weight of these outliers unless you have compelling evidence the target is actually that unique.
An example might be a tiny brand with a cultlike following that is acquired at a sky-high EV/Revenue ratio because the buyer sees synergy with an existing product line. If your target doesn’t have that brand halo, then using that multiple would inflate the resulting valuation. The exam might slip in that detail: “Brand X was known for strong brand loyalty,” to see if you catch that this transaction probably doesn’t reflect your more ordinary target.
Let’s say our target, HomeMade Supplies, has a normalized EBITDA of $25 million. You have four recent acquisition transactions with EV/EBITDA multiples ranging from 6× to 11×. A straightforward approach (though the exam might require more nuance) is:
You then apply the relevant multiple (8× to 9×) to HomeMade’s $25 million EBITDA to derive a valuation range:
• Low End: 8× × 25 million = $200 million
• High End: 9× × 25 million = $225 million
At that point, you might factor in a minor control premium if, for instance, the target is receiving a controlling share acquisition. If that premium is around 20%, you’d increase each range accordingly. The resulting final bracket might be $240 million to $270 million. This is precisely the type of logic an exam vignette might require you to demonstrate quickly and accurately.
Here’s a personal anecdote: I once valued a small artisanal coffee roaster that had intangible brand loyalty, leading a potential acquirer to pay significantly above typical EV/EBITDA multiples. The roaster’s “community vibe” and local presence were intangible. If you see a scenario in the exam referencing intangible brand or intellectual property (IP) strength, that’s usually your cue to expect a premium. Become comfortable with adjusting your final multiples or weighting those transactions differently.
• Neglecting to remove one-time or non-recurring expenses during normalization.
• Mixing up controlling acquisition multiples with minority stake multiples without adjusting for control premiums or minority discounts.
• Overusing outlier multiples that distort the final valuation range.
• Ignoring the timeline of transactions in dynamic market environments.
When you see a big list, be rigorous in your elimination process. The exam often tests your ability to weed out transactions that aren’t truly relevant. If you can do that thoroughly, you’ll likely arrive at a more credible valuation.
Let’s distill the process into a concise outline that you can use on exam day or in real-world practice:
Sometimes folks like to do a quick average or a weighted approach in a spreadsheet or Python script. While not required in the exam, you might come across it in your day-to-day job. A quick snippet:
1import numpy as np
2
3transaction_multiples = [6.0, 7.5, 9.0, 11.0] # EV/EBITDA Multiples
4filtered_multiples = [x for x in transaction_multiples if 6.5 <= x <= 10.0]
5average_multiple = np.mean(filtered_multiples)
6
7normalized_ebitda = 25.0 # In millions
8valuation_estimate = average_multiple * normalized_ebitda
9
10print(f"Selected Multiples: {filtered_multiples}")
11print(f"Average Multiple: {average_multiple:.2f}x")
12print(f"Valuation Estimate: ${valuation_estimate:.1f} million")
While you won’t do Python coding in the exam, the logic is the same: systematically apply filters, compute your multiples, and apply them to the target’s normalized metric.
• Always read the footnotes, disclaimers, or small print in the vignette. The exam loves burying information about synergy assumptions or intangible asset considerations in a line or two.
• If the vignette provides you with a handful of transactions but highlights something “unique,” consider whether that unique factor disqualifies or modifies the multiple.
• Show all your calculations clearly. The exam often wants to see not just the final number but your intermediate steps.
• Rosenbaum, J., & Pearl, J. “Investment Banking: Valuation, Leveraged Buyouts, and Mergers & Acquisitions” – A practical guide with detailed transaction analyses.
• Damodaran, A. “Investment Valuation” – Particularly helpful for adjusting outliers and synergy assumptions.
• Sample CFA Institute Vignette Questions on Private Company Valuation – Practice with official style mock item sets.
Navigating comparable transactions is part art and part science. You’ll have to juggle the raw numbers (multiples, growth rates, margins) with the intangible elements (branding, synergy potential, unique technology). Especially on exam day, the key is to follow a logical process, highlight your assumptions, weed out outliers, and clearly link your final valuation range to the multiples you’ve deemed most appropriate.
And that’s it, folks—hopefully you now feel more comfortable tackling “Selecting Comparable Transactions” in a real-world or exam setting. Practice with as many transaction data sets as you can, so that by the time you see a CFA vignette, you’ll know exactly how to parse it, compare apples to apples, and produce a solid, defendable valuation.
Important Notice: FinancialAnalystGuide.com provides supplemental CFA study materials, including mock exams, sample exam questions, and other practice resources to aid your exam preparation. These resources are not affiliated with or endorsed by the CFA Institute. CFA® and Chartered Financial Analyst® are registered trademarks owned exclusively by CFA Institute. Our content is independent, and we do not guarantee exam success. CFA Institute does not endorse, promote, or warrant the accuracy or quality of our products.