Selecting Comparable Companies: What Most Valuation Analysis Gets Wrong
Comparable company analysis is one of the most commonly used valuation tools in private markets. It is also one of the areas where judgement matters the most.
The theory behind comparable company analysis is simple - it involves deriving a valuation for a business by referencing how the market prices companies with similar characteristics in the public markets. A commonly adopted shortcut is to select a group of businesses operating in the same industry as the target company, and then to apply the average earnings multiple (typically EV/EBITDA or P/E, but this can vary by sector).
In practice, this often produces misleading results. Companies within the same sector can trade at very different valuations because the market is not pricing industry alone. Differences in growth rates, revenue visibility, operating model, and leverage tend to matter greatly.
The purpose of comparable analysis is therefore not simply to identify companies in the same industry, but to identify businesses that investors assess in broadly the same way.
What a Multiple Really Represents
A valuation multiple reflects a set of expectations about a business. Among other things, the market is implicitly pricing:
Future growth
Margin profile and scalability
Visibility and quality of revenue
Risk and cyclicality
Capital intensity
Scale and liquidity
This is why two businesses in the same sector, with similar headline revenue and EBITDA, can trade at very different valuations. The market isn’t being inconsistent – it’s reflecting real differences in risk and growth expectations. Where these characteristics differ materially between a target company and its peers, applying the observed multiples directly is unlikely to produce a defensible result.
What Makes a Good Comparable Company
A credible peer group doesn’t come from a broad screening exercise. It comes from understanding how the business being valued actually works, and then finding listed companies that investors approach with similar assumptions. In practice, that tends to mean alignment across the following areas.
1. Business model and revenue quality
The strongest comparables share similar underlying economics. This usually includes:
Recurring versus transactional revenue
Customer profile and concentration
Contract structure and retention characteristics
Route to market
Software is a good example of where this matters. Vertical SaaS businesses targeting a single industry with deep workflow integration tend to attract different valuations than horizontal platforms competing across multiple sectors. The vertical business may have stronger retention and more predictable revenues; the horizontal platform may have a larger addressable market and faster headline growth. Investors don’t view them interchangeably, so grouping them in the same peer set without further analysis creates noise rather than insight.
2. Scale, geography and market positioning
Size is a systematic valuation driver. Larger businesses attract a broader universe of buyers, carry lower concentration risk, and benefit from analyst coverage that creates price discovery and liquidity. Therefore, wherever possible, you should aim to select comparable companies of a similar size to the subject company. If none are available, you can make an explicit downward calibration to the observed multiples, and document the basis for that adjustment.
Geography also matters. A UK-listed business serving mid-market corporates is a more naturalreference point for a similar UK-based private company than a US counterpart trading on different structural multiples, in a different interest rate environment, with a different investor base. Proximity of economic conditions, cost structure and regulatory context all affect whether a peer company is genuinely comparable.
This creates a practical challenge when valuing private companies, as listed comparables are frequently larger, more liquid and sometimes geographically broader than the target business. Judgement is therefore required when translating public multiples into a private market context. Peers do not need to be identical in size, but should sit within a broadly comparable range.
3. Growth and margin evolution
Top-line growth trajectory is another key driver of valuation. Two companies generating the same level of earnings today can trade at very different valuations if one is growing materially faster than the other.
However, investors rarely look at growth in isolation. What also matters is how that growth translates into future profitability. Businesses that are expected to scale into higher margins or stronger cash generation typically command higher multiples than those where margins are expected to remain flat or constrained.
A useful comparison therefore considers both top-line growth rates and the expected evolution of margins over time, as this ultimately drives future earnings growth. Forward-looking measures are usually more informative than purely historical ones for this reason.
4. Financial leverage and risk profile
Differences in leverage can materially affect comparability. Multiples based on Enterprise Value such as EV/EBITDA are theoretically capital structure neutral, which is one reason they’re commonly used in private markets. In practice, however, a highly leveraged listed business can still trade differently from an unleveraged peer because leverage affects perceived earnings volatility and downside risk.
When selecting comparables, it is generally preferable to avoid peer groups where leverage profiles are materially different from the target business, or at minimum to recognise how this affects positioning within the observed range.
Choosing a Multiple When Many Comparables Exist
Where a large number of comparables are available, the difficulty is usually not lack of data but maintaining discipline in the selection process.
A distinction that can be useful is between the:
Core peer group, consisting of the closest operational comparables, and
Wider peer group, used to understand how the broader sector is trading.
The wider group tells you where the sector is broadly trading and provides useful context. The core group, typically three to six businesses with the closest operational and financial similarities to the target, is what actually drives the applied multiple. If you can’t articulate why each business is in the core group on its merits, it probably shouldn’t be there.
Placement within the range derived from the core group then comes down to how the target compares on growth, margin trajectory and risk profile. A business performing above the median on those dimensions belongs towards the top of the range; one underperforming sits lower.
When Comparable Data Is Relatively Limited
Some businesses operate in industries where directly comparable listed peers are relatively limited. In these situations, valuation inevitably involves broader judgement. Common mitigating approaches include:
Expanding the geographic scope of comparables
Considering businesses with similar economic characteristics in adjacent sectors (e.g. could multiples for a healthcare vertical SaaS company move with the broader healthcare sector?
Using transaction evidence to help frame an appropriate valuation range
Greater reliance on alternative approaches such as DCF analysis
The aim is not to create artificial precision, but to arrive at a conclusion that reflects how a reasonable market participant would approach valuation given the available information.
Conclusion
Comparable company analysis works best when it reflects how investors actually assess businesses in practice. The most useful comparables tend to align not only on sector or geography, but on business model, revenue characteristics, growth outlook and financial leverage.
Where peer sets are deep, discipline in narrowing the group becomes critical. Where they are limited, broader judgement and triangulation across valuation approaches become more important. In both cases, the objective remains the same: to ensure that the applied multiple reflects how the market would price a business with similar underlying characteristics.
