Government Customer Concentration: The Defense Startup Metric That Quietly Kills Deals

Most defense tech founders obsess over the metrics VCs talk about publicly: ARR growth, gross margin, burn multiple. Those matter. But there's a number that kills deals quietly, often late in due diligence, that founders consistently underestimate: government customer concentration.

Photo by Amar Preciado on Pexels.

Spend enough time doing defense tech diligence and you start to recognize the pattern. A company has $8M in revenue, clean unit economics, a genuinely differentiated product, and then you pull the customer breakdown. One program office. One contracting vehicle. Sometimes one contracting officer who is a champion for the product. That's not a business. That's a grant with product-market validation theater layered on top.

This is more common than people admit.

Why Concentration Hits Differently in Defense

In commercial software, a 40% customer concentration from a single Fortune 500 client is a yellow flag. VCs will note it, ask about diversification plans, and often move forward if the trajectory looks right. The churn risk is real, but the customer has reasons to stay, and there are hundreds of comparable potential buyers.

Government contracts operate on different physics entirely. A single defense program can vanish because of a continuing resolution, a political priority shift, a FYDP rescission, or a new program executive officer who has different vendors in mind. None of those events are correlated with your product quality. They're correlated with budget politics, bureaucratic relationships, and congressional committee priorities. Your best engineer cannot fix a budget mark.

The typical commercial playbook for managing concentration (build a bigger sales team, expand the ICP, run more outbound) doesn't map cleanly to a world where your next customer acquisition timeline is measured in fiscal years, not quarters.

What the Cap Table Sees

From a venture perspective, high government concentration creates three compounding problems.

First, revenue predictability collapses. Government contracts have end dates. Options are exercised at the government's discretion. A startup showing $10M in revenue can have a credible path to $3M in eighteen months if a recompete goes sideways or a continuing resolution delays option exercise. That volatility is incompatible with the growth assumptions embedded in most Series B or C valuations.

Second, valuation multiples compress. Enterprise SaaS might trade at 8-12x ARR when growth is strong. A defense services business with 70% revenue from one agency branch might clear 3-4x, if the buyer is optimistic. The ceiling matters when you're building toward an exit.

Third, follow-on investor appetite narrows. Crossover funds and growth equity investors who might otherwise write checks at scale get spooked by the customer concentration disclosure. They've seen the movie where a program pivot wipes out two-thirds of revenue in a single quarter.

The Paths Out

Concentration isn't automatically fatal, but it requires an honest roadmap before it shows up in a data room. A few approaches actually work.

Branch diversification within a single service is the easiest first step. Selling to one Air Force program office and expanding to two more within the same command is not glamorous, but it meaningfully reduces single-point-of-failure risk without requiring an entirely new customer acquisition motion.

Interagency expansion is harder but more durable. A company with DoD revenue that can credibly serve DHS, the intelligence community, or law enforcement has a fundamentally different risk profile. The sales cycles are still long; the buyer dynamics differ enough that a budget shock in one department won't propagate to the others.

Dual-use revenue is the premium option for the right product categories. If your core technology has a genuine commercial application (and many do, in sensing, communications, and data processing), building even 15-20% commercial revenue changes the investor conversation significantly. It proves the technology generalizes and provides a valuation floor that survives government budget turbulence.

graph TD
    A[Single Program Revenue] --> B{Recompete or Budget Cut?}
    B --> C[Revenue Cliff]
    B --> D[Option Exercised]
    D --> E[12-Month Runway Extended]
    C --> F[Recapitalization Pressure]
    E --> G[Pursue Branch Diversification]
    G --> H(Reduced Concentration Risk)
    F --> H

What Founders Should Do Before the Data Room Opens

Start tracking customer concentration the same way you track ARR. Build it into your board reporting. If a single customer represents more than 35% of revenue, treat that as an active risk item with a quarterly update on mitigation progress.

Diversification doesn't happen fast. The right time to start is before a VC asks the question, not in response to it. Investors can distinguish between a founder who has been managing this deliberately for eighteen months and one who just noticed it during term sheet negotiations.

The defense market rewards patient capital and patient founders. But patience applied only to sales cycles and not to portfolio risk management is how good companies get stuck at Series A with no clear path forward.

Concentration is a solvable problem. Leave it unsolved long enough, though, and it becomes the reason the deal doesn't close.