LTM versus NTM Multiples in Precedent Transaction Analysis

A multiple labeled “EV/EBITDA” answers one question only if you know what EBITDA the analyst used. A transaction that traded at 10× trailing EBITDA and a different transaction that traded at 10× forward EBITDA are measuring different things. Put them in the same column of a precedent table without noting the difference and you have not built a comparable set — you have assembled a collection of numbers that happen to use the same label.

The financial period is the part of a multiple practitioners are most likely to skip over and most likely to be questioned on. Getting it right for every deal in a set is not a detail. It is the arithmetic foundation that makes the rest of the analysis usable.

What LTM means and how it is calculated

LTM stands for last twelve months. It represents the most recent full year of operating results, measured backwards from a specified reference date — typically the last reported quarter before the deal announcement.

The standard calculation is:

LTM = last full fiscal year + year-to-date period − comparable prior-year period

For example: a company with a December 31 fiscal year reports a deal in September 2024. The LTM period ends June 30, 2024 (the last completed quarter before announcement). The LTM EBITDA is the full-year 2023 figure, plus the first half of 2024, minus the first half of 2023.

The reference date matters. Using December 31, 2023 EBITDA for a deal announced in September 2024 is the last fiscal year, not LTM. For a company growing at 20% annually, those two figures can differ by roughly 15%. Calling both “LTM” without checking the date is an error in the set.

LTM is the default period for precedent analysis because it reflects actual reported results rather than a forecast. The inputs are the target’s public financial statements or disclosed figures in regulatory filings. A counterparty can reproduce the calculation. A dispute about the multiple then becomes a dispute about the disclosed source figures rather than about whose estimate was more reasonable — which is a much narrower argument.

What NTM means and where it appears

NTM stands for next twelve months. It uses projected earnings rather than reported ones — typically consensus sell-side estimates or disclosed management guidance for the forward period beginning at announcement.

NTM multiples appear most often in research notes and financial-data terminals, where the forward multiple is calculated automatically from consensus estimates. They also appear in disclosed fairness opinions where an investment bank presents both trailing and forward multiples as part of its analysis range. A deal may be described as “10× LTM EBITDA” in the press release and “8× NTM EBITDA” in the bank’s fairness opinion, with both numbers referring to the same transaction price.

Using NTM multiples in a precedent set is not inherently wrong. The problem is verifiability. The estimate used for a forward multiple may have come from analyst consensus at the announcement date, from management’s internal projections, from an adjusted consensus that strips non-recurring items, or from a calendarized estimate that adjusts a non-December fiscal year to match the rest of the set. None of those sources are identical, and none of them are available from the face of a press release.

Unless the source you are working from explicitly states the NTM figure and its basis, you cannot reconstruct an NTM multiple reliably. You can construct an LTM multiple from the target’s audited financial statements, but you cannot construct an NTM multiple without access to the forward estimate that existed at the announcement date.

When each period is more useful

LTM is appropriate when:

You need a multiple that a counterparty or board can verify independently.
The target had stable or predictable revenue and earnings in the trailing period.
The deal set includes privately-held targets where no consensus estimates existed.
You are presenting a fairness analysis where the arithmetic must survive scrutiny.

NTM is appropriate when:

A source explicitly discloses the forward estimate used to calculate the multiple.
The sector standard is to value on forward earnings — certain technology, SaaS, and high-growth sub-sectors are typically valued on forward revenue or EBITDA, making NTM the more meaningful reference.
You are cross-checking your trailing-based range against a forward view and labeling it clearly as supplementary.

In most precedent sets built from publicly available sources, LTM is the primary period. NTM multiples sourced from fairness opinions are a useful supplement when the transaction discloses them. NTM multiples sourced from a data terminal’s consensus estimates should be used cautiously, because the estimates loaded at any point after the announcement date may not match what existed at the time the deal was struck.

How to identify the period a source used

A disciplined check of the source before accepting its multiple is not optional. Here is what to look for in the most common document types.

Press releases. The best-disclosed deals state the period explicitly: “10.2 times the target’s EBITDA for the twelve months ended March 31, 2024.” Many do not. When the period is not stated, treat the multiple as unverified until you calculate it yourself from the target’s filings.

Proxy statements (Schedule 14A or equivalent). The fairness opinion appended to a proxy will typically present a precedent transaction table with the period stated in a footnote or column header. This is often the most reliable source for the period and the components of the calculation. It will also usually note whether the EBITDA figure was as reported or adjusted for non-recurring items.

Regulatory filings (schemes of arrangement, tender offer documents). Public tender offer documents and UK scheme documents often contain a financial analysis prepared by the acquirer’s or target’s bank, with multiples and their period stated in a table footnote. These are frequently more detailed than a press release.

Data terminals and financial databases. Terminals calculate multiples using their own standard methodology, which may or may not match what the deal parties used. A terminal’s LTM EBITDA for a given deal will typically be the target’s last reported trailing period at announcement, calculated from the filed financial statements. That is usually a reasonable proxy, but it may differ from the EBITDA disclosed in the deal announcement if the parties applied adjustments. When you use a terminal figure, note that it is a calculated value, not a disclosed one.

Calendarized periods and non-December fiscal years

A target with a March fiscal year has periods that do not align with a December-end peer in the same set. Some analysts calendarize: they adjust the non-December target’s financials to a December year-end equivalent. Others leave each target on its own fiscal calendar. Either approach is defensible, but applying them inconsistently across a set is not.

Before constructing your set, decide whether you will standardize periods. If you calendarize, note the method. If you do not, label the period end for each transaction in the table so the reader can see the variation. What you cannot do is compare a LTM ending March 2024 in one row to a LTM ending December 2023 in the next row without identifying the difference — for a growing company, that is nearly a full year of earnings separating two numbers presented as the same measure.

Normalizing a mixed set

Most practitioners encounter precedent databases that include multiples from multiple sources. Some entries will have LTM multiples calculated from filings, some will have NTM multiples disclosed in fairness opinions, and some will have a multiple and no period stated at all. The normalization problem is how to make them comparable.

The practical options are:

Recalculate to a common period. For every transaction where you can access the target’s financial statements, calculate the LTM multiple yourself using a consistent reference date rule — for example, the last reported quarter ended at least 30 days before the announcement. This eliminates variation from different sources’ period choices, but it requires access to the underlying financials for each target.

Flag period separately from the multiple. If recalculation is not practical, create a column in your precedent table that records the period for each multiple. A reviewer can then see at a glance that five transactions are LTM and two are NTM, and weight the comparison accordingly.

Exclude rather than guess. If a multiple’s period cannot be identified from the source and cannot be calculated independently, exclude it from the quantitative set or present it in a separate unadjusted reference table. A multiple of unknown vintage is not a data point — it is noise that makes the rest of the set look less rigorous.

A common error: treating the announced multiple as LTM

Deal announcements sometimes lead with an enterprise value and describe it as a multiple of EBITDA without specifying the period. The implied EBITDA is whatever produces the stated multiple. That number may be a trailing period, a calendarized period, a consensus estimate, or a management- adjusted EBITDA that excludes items the parties agreed to treat as non-recurring. Using that announced multiple in your set without verification means accepting an unknown period and an unknown EBITDA definition.

The correct response when a source gives you only a headline multiple is to test it: divide the disclosed enterprise value by what you can calculate from the target’s financials. If the number matches the announced multiple, you now know which period produces it. If it does not match, you have found a non-standard EBITDA figure — and that is important information about how the parties presented the deal.

What to record for each transaction

For a precedent set that can be audited, capture the following for every multiple you include:

the financial period end date and whether it is LTM, last fiscal year, or forward;
the EBITDA figure in absolute terms and its currency;
whether the EBITDA is as reported or adjusted, and what was adjusted if disclosed;
the enterprise value used in the numerator and its components;
the source of the multiple (press release, proxy, fairness opinion, calculated);
whether the multiple was reported by the source or calculated by the analyst.

This information makes the set reproducible and makes any subsequent challenge narrower. A reviewer who disputes a multiple in a set with full period documentation has to identify a specific figure in a specific filing that differs from your record. That is a tractable disagreement. A reviewer disputing a multiple in a set with no period documentation has a much wider target.

PrivyLogic is a searchable database of precedent M&A transactions with available deal terms and valuation multiples. Records include the financial period and the basis of the multiple where disclosed; undisclosed fields are left blank rather than estimated. For selecting a comparable set and checking the multiples for financial period consistency, see the available plans. For enterprise access, contact the team.

For background on the enterprise value and equity value distinction that sits behind every multiple, see Equity Value versus Enterprise Value in an Acquisition. For how cash, debt and earnouts affect the numerator before you divide by EBITDA, see How Cash, Debt, and Earnouts Affect Transaction Value.

Sources

Financial Measures and Ratios — New York University Stern School of Business; accessed September 7, 2026