InvestorsFree
Comparable Company and Multiple Analysis
Send the screen and the set you kept, and get the range priced twice: once on the whole screen, once on your selection.
Every comparable company analysis template ranking for this query is the same object: a grid with tickers down the side and EV to EBITDA across the top, and instructions to pick companies of similar size, sector and growth. The grid is fine. The problem is that it only has room for the companies you kept, and a comparable set is decided by the ones you dropped. Nothing on page one asks for that list, so nothing on page one prices the decision.
Halden Fluidics is an invented equipment maker with 42.6m of revenue and 8.1m of adjusted EBITDA. The screen returned nineteen companies. Two had negative EBITDA, so no multiple exists. Four disclose no segment that isolates the comparable line, three earn most of their revenue from services, and two were dropped as not comparable in scale or focus. Eight survived, and their median EV to EBITDA is 11.4 times. That is the number the analysis reports.
The seventeen companies with a defined multiple have a median of 9.2 times. On 8.1m of EBITDA, that is 74.5m against the selected set's 92.3m. The selection is worth 17.8m of enterprise value, or 23.91 percent, and it is the single largest input on the page. The run records every exclusion with the reason and the document that tests it, so the 17.8m is attributable to named decisions rather than buried in a set nobody can reconstruct.
Adjusted EBITDA is not one measure, and eight comparables use eight of them
The second half of the problem is the denominator. Each of the eight discloses adjusted EBITDA on its own definition, and in filings the SEC requires the most directly comparable GAAP measure to appear with equal or greater prominence, plus a reconciliation. That reconciliation table is the useful artifact, because it lists the add-backs one by one. Count them across the set and the eight companies use between three and eleven lines each, with only two line items appearing in all eight of them.
Restate every company to those two common add-backs and the median multiple goes from 11.4 times to 12.9 times, because a stricter definition lowers each denominator. Run the same restatement on Halden's own accounts and adjusted EBITDA goes from 8.1m to 6.9m. Multiply the restated pair and enterprise value is 89.0m. The mixed pair, a set median on eight definitions against a target on a ninth, gives 92.3m. The mismatch is 3.33m.
That 3.33m is not a rounding artifact, it is a definition. Halden's own add-backs are 14.81 percent of its EBITDA and the median comparable's are 11.63 percent, a spread of 3.19 points that lands entirely in the answer. Regulation G is blunt about the standard a non-GAAP measure is held to: it must not omit a material fact necessary to make the presentation not misleading. A multiple built across nine definitions is where that gets tested.
How it works
Send the whole screen
Every company the screen returned, not the set you kept, plus the criteria the screen ran on.
Price the exclusions
The median on the full universe against the median on your set, in turns and in currency.
Unify the definition
Add-backs counted from each reconciliation table, then every multiple restated onto the common ones.
Write the rationale
One line per company saying why it belongs, and which filing exhibit supports that line.
What you get
- An exclusion register with every company the screen returned and the reason it was dropped
- The range priced twice, once on the whole screened universe and once on your set
- Each exclusion reason marked testable against a filing, or asserted with no criterion
- Every multiple restated onto one add-back definition taken from the reconciliation tables
- The target restated onto that same definition, so both sides of the ratio agree
- The inclusion rationale written per company, in the form a reviewer will challenge
Common questions
Why does the exclusion list matter more than the inclusion list?
Because the inclusion list is the answer and the exclusion list is the reasoning. On the worked example the eight kept companies imply 92.3m and the seventeen with a defined multiple imply 74.5m. That 17.8m came from eleven decisions, and a template with only eight rows on it cannot show a single one of them.
Is dropping a company from the screen not just normal judgement?
Yes, and most of these are defensible. Negative EBITDA leaves no multiple to use, and a services revenue majority is a real difference you can point at in a segment note. The register separates those from the two dropped as not comparable in scale or focus, which is a conclusion with no criterion attached and therefore nothing a reviewer can test.
Does this produce a valuation?
No. It produces a set, a range and the reasoning behind both, which is the input a valuation professional then judges. Concluding on value is their act and they sign it. Where the conclusion has to be documented for an audit, the valuation methodology pack carries the memo and the input register that record it. The same multiple later prices an exit's own findings instead of a fresh one.
How do I restate every comparable onto one definition?
From the reconciliation each filer already publishes. Filings must show the GAAP measure with equal or greater prominence and reconcile the difference, so the add-back lines are itemised. Take the intersection across your set, apply it uniformly, and apply it to the target too. On the worked example that intersection was two lines out of a range of three to eleven.
What if the set has to match the one used last quarter?
Then continuity becomes an input and the register is what makes it visible. Compare the sets by name rather than by count, because a set can hold eight members at both dates with three names changed. The valuation methodology pack treats that membership change as its own line in the mark bridge rather than letting it arrive inside a repriced multiple.
Where do the target's own adjustments come from?
Its accounts, itemised, with a reason per line. That is the same discipline normalising a target's earnings applies on the diligence side, and the two have to agree or the ratio is measuring two companies. Halden's add-backs are 14.81 percent of EBITDA, which is 3.19 points above the median comparable's. On a fund interest rather than an asset, secondary sale analysis does the equivalent restatement.
Does it read the filings itself?
It reads what you send and it searches for public filings where you name a company, then says which figures came from a document and which came from you. Where a number cannot be traced to a filing it is marked unsourced rather than used quietly. For the fund-level metrics these sets eventually feed, fund performance and benchmark analysis is the next stop.
Comparable Company and Multiple Analysis
Fill in the form and your workspace opens with the work already underway.