EBITDA Multiple Analysis in the Lower Middle Market

EBITDA multiples in the lower middle market are primarily determined by three factors: sector comparables, business quality characteristics, and deal-specific dynamics. Sector comparables set the floor and ceiling—industrials businesses typically trade at 4–6x EBITDA while software businesses trade at 8–14x EBITDA in the same revenue range. Within any sector, business quality characteristics drive variation: recurring revenue versus transactional, customer diversification, management depth, and margin consistency.

Deal dynamics—whether the process is competitive, the buyer's financial position, and the seller's motivation—can move multiples by 0.5–1.5x in either direction within any comparable set. A motivated seller in a sole-source process may accept a 4.5x multiple in a sector where comparable businesses trade at 5.5–6.0x. A highly competitive auction process for a quality business with a strategic buyer in the mix may clear at 7.5x in the same sector.

Sophisticated buyers do not apply a single sector multiple uniformly—they adjust it for the specific quality characteristics of the business. The most common quality adjustments: a premium of 0.5–1.0x for recurring revenue above 70% of total revenue, a premium of 0.5–0.75x for growth above 15% annually with maintained margins, a discount of 0.5–1.5x for customer concentration above 30% in a single customer, and a discount of 0.5–1.0x for significant owner dependence with no clear management succession.

These adjustments are not formulaic—they require judgment about the magnitude and sustainability of the quality characteristics being evaluated. A business with 85% recurring revenue but declining NRR may not deserve the full recurring revenue premium. A business growing at 20% per year through a single large contract win may not warrant the same growth premium as one growing at 12% across fifty new customers.

Precedent transaction analysis in the LMM is complicated by the limited availability of reliable transaction data. Unlike public M&A, most LMM deals are private and disclosed transaction multiples are often unavailable or estimated. The best data sources are proprietary databases that compile disclosed transaction data (GF Data, Pitchbook, Capital IQ) supplemented by your own transaction comps from deals you have reviewed or completed.

When using precedent transaction data, be attentive to the EBITDA base used in the reported multiple. Was it trailing twelve months? Seller-adjusted? QoE-adjusted? A reported 6.5x multiple on seller-adjusted EBITDA may represent a very different transaction than a 6.5x multiple on QoE-adjusted EBITDA, and aggregating these without adjusting for methodology will produce misleading comparables.

One of the most useful disciplines in multiple analysis is reverse-engineering the implied returns at different entry multiples. At what purchase multiple do your base case assumptions generate target IRR? At what multiple does the deal break even under your bear case assumptions? This framework converts the abstract question of 'is 6x too much?' into the concrete question of 'at 6x, what operating performance is required to hit my return target, and how confident am I in that performance?'

This approach also reveals how much of the return is driven by multiple expansion versus operational value creation. If 60% of your base case return is attributable to buying at 6x and selling at 8x, you have a multiple expansion bet. If 60% is attributable to EBITDA growth from operational improvements, you have an operational value creation story. The former is more dependent on market conditions at exit; the latter is more within the management team's control.