Fixed snapshot · 2026-Q3
2026-Q3M&A valuation multiples, lower middle market
EV/EBITDA and EV/Revenue quartiles by industry for US transactions with a target at or below $25M of EBITDA, as of . These figures were restated on 2026-08-27 and do not change otherwise. The correction is set out below, with the superseded numbers, so anyone who cited the original can see exactly what moved.
Correction — restated 2026-08-27
Restated after two data defects were found and fixed. First, two import paths were inserting the same transaction twice — 61 duplicate rows corpus-wide, three of them carrying contradictory multiples for a single deal — so the original sample sizes and medians were computed over deals counted more than once. Second, several sector term lists were matching industry labels belonging to other sectors: aerospace 'MRO' was matching industrial maintenance-repair-operations distributors, 'pet' was matching 'petroleum', and the industrials group was absorbing food-manufacturing and home-medical-equipment businesses. Both are fixed and covered by tests. (restated 2026-08-25) A further correction on 2026-08-26, of the same two defects. The deal-identity rule and the sector classification table each existed in two places and had drifted: one Utz acquisition was in the corpus four times at three different multiples, and sectorFor still classified industrial MRO distributors as aerospace after the peer groups had stopped. Both now have a single definition, 55 rows reclassify to the correct sector, and Aerospace & Defense publishes for the first time. (restated 2026-08-26) Trailing-basis normalization: a review of every row whose stored multiple contradicted its own EV and earnings arithmetic found 14 multiples on undisclosed different bases — post-synergy (EV over EBITDA plus expected synergies), forward-year estimates, net-of-tax-benefit prices, one announced-vs-closed price, and one currency mismatch (CAD figures stored as USD). Each was normalized to the gross trailing basis this dataset publishes, with the source's own wording preserved on the record; the affected multiples moved UP in most cases, since post-synergy figures understate what buyers actually paid relative to trailing earnings.
What the 2026-Q3 snapshot said before the correction, for anyone who cited it:
| Sector | Superseded median | Restated median |
|---|---|---|
| Aerospace & Defense | not published | 10.46x (n=14) |
| Industrials & Manufacturing | 8x (n=48) | 8.5x (n=49) |
| Software & Technology | 7.56x (n=13) | 7.79x (n=22) |
| Healthcare | 8.9x (n=19) | 8.85x (n=20) |
| Consumer & Food | 7.2x (n=19) | 9.6x (n=15) |
| Business Services | 8.65x (n=14) | 7.91x (n=20) |
| Distribution & Logistics | 5.64x (n=10) | 5.55x (n=16) |
| Construction & Engineering | 6.42x (n=16) | 6.12x (n=22) |
| Packaging & Plastics | 8.8x (n=9) | 8.5x (n=15) |
| Energy & Utilities | 6.43x (n=10) | 6x (n=17) |
| Industry | EV/EBITDA (p25 – median – p75) | n | EV/Revenue (p25 – median – p75) | n |
|---|---|---|---|---|
| Aerospace & Defense | 9.7× – 10.46× – 12.89× | 14 | 1.4× – 1.78× – 2.99× | 12 |
| Industrials & Manufacturing | 6.2× – 8.5× – 12× | 49 | 0.88× – 1.5× – 2.3× | 36 |
| Software & Technology | 6.7× – 7.79× – 10× | 22 | 1.26× – 1.99× – 2.96× | 19 |
| Healthcare | 6.75× – 8.85× – 11.68× | 20 | 1.19× – 1.5× – 2.4× | 16 |
| Consumer & Food | 7.57× – 9.6× – 12.25× | 15 | n/a (thin sample) | — |
| Business Services | 6.46× – 7.91× – 10.03× | 20 | 0.54× – 1.28× – 1.92× | 15 |
| Distribution & Logistics | 3.75× – 5.55× – 7.4× | 16 | 0.51× – 0.74× – 0.9× | 13 |
| Construction & Engineering | 5.13× – 6.12× – 9.02× | 22 | 0.56× – 0.84× – 1.16× | 16 |
| Packaging & Plastics | 6.25× – 8.5× – 9.6× | 15 | 0.51× – 1.13× – 1.94× | 11 |
| Energy & Utilities | 4.31× – 6× – 9× | 17 | 0.63× – 0.92× – 1.28× | 10 |
Method
- Scope. Transactions with a target at or below $25M of EBITDA — the lower middle market.These are not all-size M&A multiples. Across the same database, disclosed deals above $50M of EBITDA run several turns higher, because large public acquirers are the ones obliged to disclose.
- Source. Computed from 1,048 individually sourced transactions, of which 299 carry a disclosed or arithmetically implied EV/EBITDA multiple. Every underlying record has a source URL — an SEC filing, a company press release, or trade press.
- Period. Deals announced or closed 2012–2026, 82% of the size-banded sample in 2021 or later.
- Sample sizes. n is the number of transactions behind each figure. Sectors with fewer than eight are withheld rather than published: a handful of deals is not a benchmark.
- Quartiles, not averages. p25 / median / p75. Use the spread. A single point estimate hides the dispersion that actually determines what any one business is worth.
- Full rules. Sourcing, eligibility, de-duplication and correction policy are set out in the methodology, which does not change from quarter to quarter.
Download the data
The full sector table for 2026-Q3, in both formats. These files are frozen with the page, so a link to one will return the same numbers later.
The JSON carries its own provenance — licence, permalink, methodology and corpus coverage — so it still explains itself once it has been saved somewhere else.
Licence and citation
Published under Creative Commons Attribution 4.0 International (CC BY 4.0). You may republish these figures, including commercially and in adapted form, provided you credit the source and link to it. Because the snapshot is frozen, that link keeps saying the same thing.
Bankerly. "2026-Q3 M&A Valuation Multiples, Lower Middle Market." Published 2026-08-21. https://bankerly.ai/data/ma-multiples-by-industry/2026-q3
How the figures are produced, and what they do and do not cover, is set out in the methodology. Sample sizes are published per sector above; several are small, and the table says so rather than hiding it.
Other quarters
For figures that update as the transaction database grows, see the live multiples page.
This is the first published snapshot.