Kratos — Blitz Index 47 (C)

Backlog +47% and a $1.45B record award — but unmanned systems run at 0.9% margin.

Last re-scored: November 5, 2026.

Score breakdown

Think we scored it wrong? Score it yourself →

Radar axes (0–100): Moat 40 · Speed 68 · Creative 60.

Watch item

Gates: G1 ✓ · G2 ✓ (borderline — all on production conversion). 19 points below the threshold — awards are real; the factory isn't.

Strengths

• Q2'26 revenue +30.5% YoY — accelerating
• Total backlog $2.084B (+47% YoY), 75.4% funded — order-backed growth
• MACH-TB 2.0 $1.45B — largest award ever, in-house tech converting to orders
• Valkyrie flying for USAF/USMC + MUX TACAIR CCA — in the loyal-wingman wave

Concerns

• Unmanned systems margin 0.9% (worsening) — the narrative's hero is the thinnest
• Zero Valkyrie/CCA production contracts — prime is Northrop, Kratos the airframe sub
• OCF −$42.1M (FY2025) — operating cash outflow amid revenue growth
• +24% dilution in 18 months, −52% vs offering price — dilution-sensitive funding

Research notes

Q3 2026 update

Q3 2026 update (company release, Nov 5, 2026; quarter ended September 27, 2026): revenue $471.8M (+35.7% YoY, +23.2% organic); Kratos Government Solutions $397.2M (+52.5% YoY, +35.8% organic); Adjusted EBITDA $50.9M (10.8% margin); GAAP operating loss $4.9M, net loss $0.2M, GAAP EPS $0.00; adjusted EPS $0.24 (vs $0.14 a year ago); bookings $1.037B, book-to-bill 2.2, backlog $2.65B (record). Guidance from the same release (forward-looking): FY2026 revenue $1.78–$1.81B (range tightened), Adjusted EBITDA raised to $176–$180M; initial FY2027 forecast revenue $2.1–$2.2B, Adjusted EBITDA $230–$240M. Score 47/C unchanged — this is a quarterly update note, not a rescore. Source: Kratos press release, https://www.kratosdefense.com/newsroom/kratos-reports-third-quarter-2026-financial-results (also filed as 8-K Ex-99.1, Nov 5, 2026, https://www.sec.gov/Archives/edgar/data/1069258/000106925826000085/ktos202609278kexhibit991.htm)

Thesis: the backlog is moving faster than the profit

Kratos Defense & Security Solutions enters the BlitzLab record with a first scoring of 47 out of 100, grade C, built from Speed 17 out of 25, Moat 24 out of 60, Creative Destruction 9 out of 15, and deductions of 3. The total sits 18 points below the 65-point line, and the shape of the score matters more than the headline: on the radar conversion used by the model, Speed converts to 68.0, Creative Destruction to 60.0, and Moat to only 40.0. This is a company whose order book is accelerating while its moat, measured on the frozen v1.0 anchors, remains the thinnest part of the case.

The central tension in the file is stated plainly in its key drivers. The growth story most often told about Kratos is a drone story, centered on the Valkyrie loyal-wingman aircraft and the wider idea of cheap, attritable unmanned aircraft flying alongside crewed fighters. The profit-and-loss statement tells a different story about where the money currently sits. The unmanned systems segment produced $292.0 million of revenue in fiscal 2025, about 22 percent of the total, at an operating margin of 0.9 percent. The larger body of the company is the traditional defense business, the KGS segment, at roughly 78 percent of revenue with a 5.7 percent operating margin. The consolidated operating margin for fiscal 2025 was 1.9 percent. A reader who starts from the narrative and a reader who starts from the segment margins will therefore reach very different impressions of the same company, and the score file refuses to let either impression stand alone. The B4 gross-profit item, scored 3 out of 12, is where that gap is priced into the assessment.

Growth itself is not in doubt on the figures cited. Revenue rose from $811.5 million in fiscal 2021 to $1,346.8 million in fiscal 2025, an increase of 18.5 percent in the final year of that span and a five-year compound annual growth rate of 13.5 percent. The most recent quarter in the file, the second quarter of fiscal 2026, reported revenue of $458.8 million, up 30.5 percent year on year, against $371.0 million in the first quarter of fiscal 2026. Trailing-twelve-month revenue stood at about $1.52 billion, and the score file puts the market capitalization at about $7.6 billion. Both gates pass, so the scoring proceeds, but the second gate passes only on a borderline, estimated basis, and that qualification runs through everything that follows. The question the file keeps returning to is not whether Kratos is growing, but whether growth of this kind, at these margins, from this position in the supply chain, can compound into the much larger company the gate test imagines.

The order book is the strongest evidence that something real is accelerating. Total backlog reached $2.084 billion in the second quarter of fiscal 2026, up 47 percent year on year, with funded backlog of $1.572 billion, or 75.4 percent of the total. For comparison, backlog stood at $1,244 million in fiscal 2023. The score file stresses a discipline in how that backlog is counted: under the IDIQ contract structure, only funded task orders are counted in backlog, so the funded portion is not a wish list but work that has been funded. That distinction matters for a business whose customers are governments, because an unfunded ceiling value can flatter a pipeline that funded backlog does not. Here the funded share is high, and the year-on-year growth in the total is large, which is why the file treats the backlog as order-based growth rather than expectation-based growth, explicitly contrasting Kratos with the JOBY pattern of pre-commercial, pre-certification expectation. Kratos, the file notes, has about $1.52 billion of trailing-twelve-month real revenue plus $1.57 billion of funded backlog behind it.

At the same time, the file is careful about what the backlog does not prove. Backlog is a promise of future revenue, not revenue itself, and the conversion of backlog into revenue, at what margin and on what schedule, is the open question. The score calls this stage the first act of a blitz, in its ramp-up phase: backlog moving before revenue, prototype work under way, and the path from about $1.5 billion of revenue toward a much larger scale still at its entrance. That framing is useful because it keeps the acceleration in view without mistaking it for arrival.

Speed: fast orders, funded backlog, and a single large award

Speed scores 17 out of 25, the strongest of the three radar axes on conversion, and its four items show where the pace comes from and where it is constrained. Growth, A1, scores 7 out of 10. The revenue trajectory and the backlog growth support the mark, but the file attaches a concentration warning to it: the engine of growth leans heavily on a single award, the MACH-TB 2.0 contract, a five-year OTA agreement valued at $1.45 billion, described as the largest award in company history. That value rests on the premise that all options are exercised, a premise the file states rather than assumes silently. A single award of that size can transform a backlog, and it can also concentrate the story: if the options, the execution, and the follow-on work do not unfold as the headline value implies, the growth rate has less underneath it than the backlog chart suggests. The file also notes that the unmanned systems segment, at 22 percent of revenue, is thinner in the profit-and-loss statement than the drone narrative implies, which tempers the growth item even as the headline rate accelerates.

Capital velocity, A2, scores 4 out of 5. In February 2026 the company raised about $1.38 billion through a public offering of 16.43 million shares at $84.00 per share, a figure confirmed in a SEC filing, and by the second quarter of fiscal 2026 cash stood at $1,437.6 million. Long-term borrowings were zero, against $174.6 million in fiscal 2024, fully repaid, and an immediately effective S-3ASR shelf registration remained available. On the financing side, then, the company moved quickly and in size when it chose to raise capital, and it did so without leaving a long-term debt balance behind. The other side of the same item is the price record that followed: the offering price of $84 compares with a price of $40.42 cited from market data dated October 16, 2026, a decline of 52 percent from the offering price. The file treats that gap as a sensitivity for any future capital raising, because a market that has seen that kind of decline after a large offering may respond to further dilution with caution. Speed of capital, in other words, was demonstrated once, at scale, and the cost of demonstrating it again is an open question.

Hiring and productivity, A3, scores 2 out of 5. The company employed about 4,300 people, with revenue per employee of about $313,000 in fiscal 2025, a profile the file characterizes as defense-hardware productivity. Hiring is not described as the bottleneck on growth, because the structure is government-contract based rather than a consumer hiring race, but neither is there evidence of the kind of productivity leverage that would lift the item higher. The mark is modest, and it fits the overall picture of a hardware business scaling through contracts and programs rather than through a software-like productivity curve.

Share-gain velocity, A4, scores 4 out of 5. Kratos occupies the prime position on MACH-TB 2.0 under its five-year agreement, and it retains sole-source production positions on three target-drone programs: the BQM-167A in its twenty-first year, the BQM-177A in its seventh year, and the MQM-178 in its thirteenth year. The Valkyrie has flown for the the USAF and the USMC, and in Europe the file cites, on a press-reported basis, a flight with Airbus and participation in a German next-generation unmanned fighter competition. A quantified market share is not confirmed in the file, so the item rests on the pace of program wins rather than a measured share figure, and the file says so. What can be said is that the company keeps winning and keeping positions in programs that last, which is the government-contracting version of distribution velocity: long incumbencies, renewed, rather than a viral adoption curve.

Moat: a government-contract franchise, not a network effect

Moat scores 24 out of 60, the weakest section relative to its weight, and the file gives it a name that clarifies the whole assessment: a government-contract franchise moat rather than a platform moat. Network effects, B1, score 6 out of 18, in the anchor range for data and ecosystem effects of five to nine. The moat consists of sole-source production contracts and long contracts of three to five years, which are durable in the sense that incumbency is hard to displace, but which do not compound the way a network or an ecosystem does. Each new contract must still be won, performed, and renewed, and the value of one program does not automatically make the next program cheaper to win or more valuable to the customer in the compounding sense that defines a strong network effect.

The GE Aerospace team-up, cited from the 10-K at Note 17, is the closest thing in the file to a platform element: joint development and production of engines, with a revenue- and cost-sharing structure. But the file immediately qualifies it, noting that in the loyal-wingman and CCA area, Kratos sits inside collaboration structures led by large primes, in the position of an airframe provider rather than the prime itself. That positioning lowers the quality of the moat, because the economics of a program, including margin, volume, and pricing power, tend to sit with the prime. The moat is real, in the sense that sole-source incumbencies and long contracts are difficult for a rival to take away quickly, but it is a franchise moat, renewed contract by contract, rather than a moat that widens on its own as more users join.

Distribution scalability, B2, scores 8 out of 18, at the upper end of the anchor range for paid, sales-dependent but repeatably expandable distribution. The channels are direct: the DoW and intelligence agencies, IDIQ contracts under which only funded task orders enter backlog, and firm commercial orders. Long contracts of three to five years form the repeating structure, and the file adds two extensions on a confirmed and press-reported mix: the Orbit acquisition, at $352.7 million, extending the company into satellite communications, and the European Valkyrie activity, including the flight with Airbus and the German competition participation on a press-reported basis. Distribution here does not scale by adding users at near-zero marginal cost; it scales by winning the next funded program, in the next country or the next service branch, with the same sales motion repeated. The mark of 8 reflects that the motion is repeatable and is in fact repeating, without claiming the near-free scaling of a software channel.

Market size, B3, scores 7 out of 12 and is labeled an estimate. The reasoning is tied directly to the second gate, which is addressed below, but the item-level point is that the ceiling is large enough to matter only if the wedge converts: loyal-wingman production, the MACH-TB ecosystem, and engines together must carry the company toward a scale far above its current revenue. The established the KGS segment base business is described as stable but with a lower growth ceiling, so it cannot carry the market-size case alone. The full mark, the file says, would require demonstrated production-contract wins for the CCA, which have not yet occurred. The 7 is therefore a mark for a plausible but unproven ceiling, not for a measured market.

Gross profit, B4, scores 3 out of 12, and this is the item that most disciplines the enthusiasm elsewhere in the file. The gross margin fell from 27.7 percent in fiscal 2021 to 22.9 percent in fiscal 2025, confirmed in the filings and checked against inline XBRL data. Segment operating margins for fiscal 2025 were 5.7 percent for the KGS segment and 0.9 percent for unmanned systems, with the unmanned margin deteriorating from 2.0 percent to 0.9 percent over the period described. The consolidated operating margin was 1.9 percent. The file states the resulting tension as the core mismatch of the case: growth-company valuation against subcontractor margins. That valuation observation itself, including a trailing price-to-earnings ratio of about 235 times cited in the file, is explicitly not reflected in the score, because the model scores the business and not the price. But the margin figures are reflected, and they are reflected harshly, because they show that the fastest-growing narrative segment is also the thinnest, and that the margin trend is moving in the wrong direction even as revenue accelerates.

Creative destruction: a real substitution thesis, still at prototype scale

Creative Destruction scores 9 out of 15, with all three items at 3 out of 5, a symmetry that captures a thesis that is coherent, partly demonstrated, and not yet proven at production scale. The order being displaced, C1, scores 3. The destruction hypothesis is that cheap, attritable unmanned aircraft can take over some missions now flown by expensive crewed fighters, with the Valkyrie cited at a unit cost in the millions of dollars against crewed fighters costing tens of millions to hundreds of millions of dollars. Target drones, the BQM series, are already at the completed-substitution stage in their niche: they exist precisely to be shot at, and they have replaced older approaches to that training mission. But tactical unmanned aircraft for combat missions are at the concept-to-demonstration stage, prototypes rather than production, and the file keeps the mark at 3 for that reason. The economics of substitution are visible; the substitution itself, at combat scale, is not yet.

Irreversible behavior change, C2, also scores 3. The backlog evidence is the early demonstration: backlog grew from $1,244 million in fiscal 2023 to $2,084 million in the second quarter of fiscal 2026, driven by MACH-TB 2.0, and the file presents MACH-TB as a case in which internally funded development, the Zeus and Erinyes flights, converted into an order. That is the pattern creative destruction needs in its early phase: a customer funding, at scale, something that was first built on internal money, because the internal work proved something the customer then wanted to own. Unmanned systems revenue also rebounded, from $212.2 million in 2023, a year of decline, to $292.0 million in 2025. The mark stays at 3 because program scale is still small relative to the order it would displace. A changed procurement pattern is visible in the backlog, but it is not yet the dominant pattern of the market.

Cost-structure change, C3, scores 3 as well, on a press-reported basis for the forward-looking production figures. Attritable airframes and low-cost engines, the GEK800 and GEK1500, give the destruction thesis a unit-cost logic that is structurally lower than crewed aircraft, so the economics of the thesis cohere. What does not yet cohere as evidence is the production plan attached to it in press reporting from October 2026: more than 3,000 units of parts orders in 2027 and a goal of 10,000 units per year by 2029, described in the file as advance investment without a production contract, a price, or a delivery schedule. The file is explicit that this is investment ahead of proof, and the mark of 3 reflects that discipline. A cost advantage that has not yet been produced at volume is a thesis about costs, not yet a demonstrated cost structure.

Deductions and risks: clean of debt, leaking cash, diluting fast

Deductions total only 3 points, so there is no FLAG, since the FLAG line in the model sits at 15 points of deductions, but the composition of those 3 points is more informative than the total. Debt, D1, scores minus 3 out of a possible minus 10. Long-term borrowings are zero, against $174.6 million in fiscal 2024, fully repaid, confirmed in the filings. On a debt-only reading, the balance sheet looks clean, and cash of $1,437.6 million reinforces that impression. The minus 3 corrects the impression rather than contradicting the debt figure. Operating cash flow was negative $42.1 million in fiscal 2025 and negative $38.4 million in the first half of fiscal 2026, meaning the company consumed operating cash while revenue grew. Unbilled receivables stood at $405.4 million in the second quarter of fiscal 2026, a working-capital burden that sits between revenue recognized and cash collected. And the share count rose from 151.1 million to 168.9 million and then to 187.6 million over eighteen months, an increase of 24 percent, driven by the large offering. The file calls this the correction of an illusion: the company is effectively free of long-term debt, but its cash-generation structure is weak, and its growth has been funded in part by dilution. The minus 3 is small, but it points at the right question, which is whether growth can become self-funding.

Revenue quality, D2, scores 0. The 10-K contains no mention of a material weakness, the auditor Deloitte gave an unqualified opinion as recorded in the filing, and the fact that about 80 percent of revenue comes from United States government-related sources is treated as a structural characteristic of the business, a standing item in the risk factors, rather than a trigger. The mark is recorded as a confirmed no-trigger, which is a meaningful statement for a government contractor: concentration in one customer type is the business model, not a newly discovered risk, and the file does not penalize the company for being what it is.

Governance, D3, scores 0 on a confirmed and unconfirmed mix. The chief executive is Eric M. DeMarco, named in the 10-K signature block, and the file records no mention of a dual-class voting structure, treating the structure as standard, with no unusual concentration confirmed. Regulatory and social cost, D4, also scores 0 on a press-reported basis: no confirmed securities litigation or fines were found, and the file records the price record, a 52-week range of $37.84 to $134.00, a decline of about 70 percent from the high, and a trailing price-to-earnings ratio of about 235 times, as a valuation-area observation that is not reflected in the score. The model scores the business, and the file keeps that boundary visible by recording the price facts without letting them move the total.

Taken together, the deductions section is unusually clean for a company with a weak cash-conversion profile, and that combination is itself a finding. The risks in this file are not accounting risks or governance risks on the evidence cited; they are operating and strategic risks, concentrated in margins, in subcontractor positioning, and in the conversion of backlog and prototypes into funded production at profitable rates.

The gates: both pass, and the second pass is the whole argument

Gate 1, sustained growth or a second growth curve, passes on a confirmed basis. Revenue grew from $811.5 million to $1,346.8 million over five fiscal years, the second quarter of fiscal 2026 accelerated to 30.5 percent year-on-year growth, and funded backlog of $1.57 billion supports the claim that growth rests on orders rather than expectation. The file is explicit that the quality of that growth, an operating margin of 1.9 percent and negative operating cash flow, is assessed separately in the B4 and D1 items, so the gate pass does not wash away the margin problem. It establishes only that growth is real and continuing, which the figures support.

Gate 2, a ten-year market ceiling at least ten times current revenue, passes on an estimated, borderline basis, and the file defines its wedge narrowly before doing the arithmetic, following the BMNR precedent of declaring the wedge before scoring. The whole defense market is excluded as a market definition, because counting all defense spending would make the pass trivial and the analysis meaningless. The bar is about $15.2 billion, ten times trailing-twelve-month revenue of about $1,522.5 million. The scenario figures are: first, MACH-TB 2.0 at $1.45 billion over five years, the largest award in company history, on the premise that all options are exercised; second, a CCA program measured in thousands of aircraft at millions of dollars per aircraft, a program worth tens of billions of dollars in total, in which Kratos would need a meaningful share as an airframe provider; third, engines through the GE team-up, with no production contract yet; and fourth, base growth in KGS, which at an assumed rate of about 10 percent per year would reach about $4 billion in ten years. Reaching the bar requires a ten-year compound annual growth rate of about 26 percent, against a current trajectory of 18.5 percent for the final year of the five-year span and 30.5 percent in the most recent quarter. The file judges that reachable only if the current trajectory extends and combines with wedge conversion, and it states the decisive variable plainly: production contracts for Valkyrie and the CCA currently number zero. The MUX TACAIR participation is a twenty-four-month prototype OTA agreement, and the prime is Northrop Grumman, with Kratos in the airframe-provider position. The file therefore attaches an explicit review trigger: a production-contract win requires a re-examination upward, while continued absence of production wins requires a re-evaluation of the gate itself. That trigger is the single most important sentence in the file for anyone tracking this company, because the gate pass and the score both depend on a conversion that has not yet happened.

Blitzscaling fit: an order blitz, led from the supply chain

The file's blitzscaling-fit analysis gives Kratos a distinctive label: a business-to-government order blitzer, riding the same wave as Anduril but from a different position in it. Where Anduril is described as the prime, Kratos is described as the supply chain. The main growth levers are distribution, meaning DoW program wins, and capital, meaning the $1.38 billion offering. The secondary lever, platforms in engines and airframes, is closer to component supply than to a platform, and the weak lever is network effects, because the contract-franchise moat does not compound on its own. The trade of speed against efficiency was made in two ways: speed was bought with the offering, producing a cash balance of about $1.44 billion at the cost of dilution and a subsequent price decline from the offering level, and speed was pursued through advance investment, in Valkyrie development and in the engine production goal, ahead of customer contracts. Efficiency, on the evidence, was not so much traded as structurally absent: an operating margin of 1.9 percent is not a temporary sacrifice for speed but the margin profile of the business as it currently stands.

The file's largest blitzscaling risk is stated as a single line: the protagonist is a subcontractor. The loyal-wingman story belongs, in narrative terms, to the Valkyrie, but the MUX TACAIR prime is Northrop Grumman, and Kratos provides the airframe. The economics of the program, margin, volume, and pricing power, therefore sit primarily with the prime, and it is possible for the narrative to appreciate while the profit-and-loss benefit accrues elsewhere. That is not a prediction about any program outcome; it is a structural observation about where value sits in a prime-led program, and it explains why the moat items score as they do despite the program wins.

Three open model questions are recorded in the file as version 1.1 candidates, with the model itself unchanged. The first is a weighting for government-contract backlog in the distribution item, separating funded from unfunded backlog and anchoring the backlog-to-revenue ratio, with Kratos at 75.4 percent funded and $2.084 billion as a reference case. The second is a rule for the gap between a growth narrative and profit-and-loss reality, detecting the pattern in which the narrative's leading segment has a lower margin than the consolidated business. The third is a finer treatment in the debt item for dilution-funded companies, linked to the BMNR candidate, for businesses that carry no debt but dilute heavily, as Kratos did by 24 percent in eighteen months. These are recorded as notes, not applied, and they show the file's own awareness that the current model may underweight exactly the features that define this company.

What to watch: conversion, margin, and cash

What to watch follows directly from the score, and each item is a conversion question rather than a narrative question. The first is production conversion: whether a CCA production contract arrives, and in what position, prime or provider, and on what economics, because that single event is the stated review trigger for the second gate and the clearest path to the full market-size mark. The second is backlog conversion: whether the $2.084 billion backlog, and the $1.572 billion funded portion within it, converts into revenue on schedule, and whether the MACH-TB 2.0 options that underpin the $1.45 billion headline value are exercised as the file's premise assumes. The third is margin conversion: whether the unmanned systems margin recovers from 0.9 percent toward, and then above, the consolidated level, and whether the gross margin decline from 27.7 percent to 22.9 percent stabilizes, because a growth story carried by its thinnest segment cannot compound indefinitely at a 1.9 percent consolidated operating margin. The fourth is cash conversion: whether operating cash flow turns from negative $42.1 million in fiscal 2025 and negative $38.4 million in the first half of fiscal 2026 toward self-funding, and whether unbilled receivables of $405.4 million narrow as work is billed and collected. The fifth is the engine program: whether the GE team-up and the low-cost GEK800 and GEK1500 engines move from joint development toward a production contract, and whether the press-reported production figures for 2027 and 2029 acquire the contracts, prices, and delivery schedules they currently lack. The sixth is the European channel: whether the Airbus flight and the German competition participation, both press-reported, convert into funded program positions comparable to the domestic sole-source incumbencies.

Two further markers belong in any watch list drawn from the file. The third-quarter fiscal 2026 results had not been released as of the scoring date and were excluded from this assessment, so they appear here only as a future marker, with no anticipated outcome attached. And the valuation-area facts recorded in the file, the 52-week range of $37.84 to $134.00, the decline of about 70 percent from the high, the price of $40.42 cited from October 16, 2026, and the trailing price-to-earnings ratio of about 235 times, are recorded without effect on the score, because the model assesses the business and not the price. A future rescoring will have new figures for revenue, backlog, margins, and cash flow, and the honest test of this 47-point assessment is whether those figures move the items that currently keep it down: B4, B1, and the production-dependent parts of B3 and the second gate. If production arrives, margins widen, and cash turns, the score has room to move; if prototypes stay prototypes and the thinnest segment stays the thinnest, the score explains why it has not.

More Frontier companies

Other Frontier names on the Blitz Index, highest score first: Rocket Lab (62/B), Tesla (62/B), Illumina, Inc. (61/B).

Next earnings

Q4 2026 · Expected

Coverage

Kratos Defense & Security Solutions, Inc. · Q3 2026 earnings Nov 5, 2026 (after close, confirmed) · Score 47 · Grade C

Embed this score

Copy the HTML below to show this score on your site. The badge links back to this page and updates automatically when the score changes.

<a href="https://blitzlab.app/companies/ktos/"><img src="https://blitzlab.app/badge/ktos.svg" alt="Kratos Blitz Index 47/C — BlitzLab"></a> Copy HTML

Preview:

BlitzLab scores businesses against a scholar’s standard — not investment advice. No price targets, no buy/sell recommendations. Figures are from SEC filings (XBRL) as labeled in the source scorecard.