The Pentagon keeps reorganizing for speed. Delivery keeps slowing
Summary
GAO's 24th annual weapons assessment puts the average time to field a capability at just over 12 years — up from 10 in the 2024 edition — across a $2.4 trillion portfolio, and says the number will likely rise because programs are holding delivery dates static while delaying the milestones in between. The measurement lands eight months after the Secretary of War ordered the acquisition system onto "a wartime footing."
The documents
Five primary documents, read against each other: three consecutive Weapon Systems Annual Assessments (2024⧉, 2025⧉, 2026⧉ — the 24th in the series), and the two November 7, 2025 reform documents: the Secretary of War's transformation memorandum⧉ and the Acquisition Transformation Strategy⧉ it ordered — 'Rebuilding the Arsenal of Freedom,' whose first stated outcome is to 'field technology and modernize systems at a rate that outpaces our adversaries.' The Comptroller General's 2025 transmittal supplies the longitudinal verdict: findings 'over my 15 years have grown increasingly dire… notwithstanding recent reforms.'
The money
The portfolio under measurement: 104 of DOD's costliest programs, over $2.4 trillion in planned investment⧉. The slippage is not free — the 2025 edition's headline was a single program, the Sentinel ICBM, reporting a $36 billion cost increase⧉. And the lane built to escape the slowness is inflating: the middle tier of acquisition, a pathway designed to deliver in five years, now carries at least $49 billion across 23 of its largest programs — up from $35 billion across 20 in the 2024 assessment — mainly, notes, because more of the largest and most time-consuming programs moved onto the rapid pathway. Almost half of MTA programs, across multiple assessments, entered that pathway with technologies too immature to deliver on its five-year clock.
View data as table
| 2024 assessment | 35 | 20 programs |
|---|---|---|
| 2026 assessment | 49 | 23 programs; almost half entered with immature technologies |
The measurement
's cycle-time numbers deserve the scrutiny it applies to everyone else's, and the report itself supplies it. The just-over-12-year portfolio average excludes six programs⧉ that have no new delivery date at all — most of them rebaselining after extensive delays. Across the 75 major programs: the 43 that have delivered took 132 months on average; the 17 still waiting already project 149 months. And because programs keep initial-capability dates fixed while compressing everything between here and there, 's own conclusion is that "the 12-year average will likely increase in the future." The published number is the floor, and the auditor says so.
View data as table
The disagreement between documents
November 7, 2025: the Secretary of War directs the department to "put our Acquisition System and Enterprise on a wartime footing and dramatically accelerate the fielding of new technology," and the strategy issued the same day promises fielding "at a rate that outpaces our adversaries," a culture shift "from a culture of compliance to one of speed and execution," five pillars, three outcomes. July 2, 2026: 's measurement arrives — the average is up again, and the report notes drily that it found the same thing about the previous speed reforms in June 2024 and June 2025: "struggled to execute" them, leaving "a status quo of slow, linear development approaches." The fairest reading, which the piece adopts: the new strategy is too young for this year's data to judge — which is precisely what was said about the last reform, and the one before it, while the trend line ran 10, almost-12, just-over-12.
View data as table
| 2024 assessment | 10 | "incompatible with maintaining military advantage" |
|---|---|---|
| 2025 assessment | 11.9 | "up to almost 12 years" — up a year and a half |
| 2026 assessment | 12.1 | "just over 12 years" — GAO: likely to increase |
What happens next
's recommendation this year targets the entry gate of the rapid lane: require programs on pathways like MTA to start with mature technologies, or develop the immature pieces separately — the practice its leading-practices work has urged since 2022. The next annual assessment, in mid-2027, will be the first with a full year of the Acquisition Transformation Strategy in the data, and its test is already written: either the delivery average bends below 12 years, or the wartime-footing directive joins the June 2024 and June 2025 findings in the pile of reforms the trend line never noticed. Six programs still owe the public a delivery date at all.
- 's average time-to-field across 's costliest programs: 10 years (2024 assessment) → almost 12 (2025) → just over 12 (2026) — while a new speed reform was announced between each measurement.
- says the 12-year figure will likely rise: programs hold delivery dates static while delaying interim milestones, and six programs with no new date are excluded from the average.
- The 5-year 'rapid' pathway now carries $49B across 23 programs — up 40% from the 2024 assessment — and almost half its programs entered with immature technologies.
- The November 2025 'wartime footing' directive is eight months old; its first real test is the mid-2027 assessment.
Method notes. Cycle-time figures are 's portfolio averages for major defense acquisition programs (expected time from program start to initial operational capability); portfolio composition changes between editions, and the 2026 average excludes six programs without a new delivery date — both caveats appear in the reports themselves. The 2026 assessment's data largely predates implementation of the November 2025 strategy; the cross-examination is with the reforms measured in its June 2024 and June 2025 editions, and the piece states this explicitly. MTA portfolio figures are floors over slightly different samples (20 vs. 23 largest programs). In the trend chart, 's verbal figures are encoded numerically for plotting — 'almost 12 years' as 11.9 and 'just over 12 years' as 12.1 — with 's verbatim wording in the bar notes. The strategy and memorandum captures are Wayback snapshots — media.defense.gov blocks non-residential fetches.
Sources(5) ▾
- U.S. Government Accountability Office, Weapon Systems Annual Assessment: Requiring Mature Technologies Could Enable Shift to Rapid Delivery (GAO-26-108457) (2026-07-02) — 's 24th annual assessment, covering 104 of 's costliest programs ($2.4 trillion planned investment). Source of the expected-cycle-time increase to just over 12 years, the 132-vs-149-month IOC cohorts across 75 MDAPs, the six programs with no new delivery date, the $49 billion / 23-program MTA portfolio, and the finding that prior speed reforms left 'a status quo of slow, linear development approaches.' gao.gov · original document
- U.S. Government Accountability Office, Weapon Systems Annual Assessment: DOD Leaders Should Ensure That Newer Programs Are Structured for Speed and Innovation (GAO-25-107569) (2025-06-11) — The 2025 edition: average expected time to initial capability rose by a year and a half, to almost 12 years; the Sentinel ICBM program reported a $36 billion cost increase; the Comptroller General's transmittal notes findings 'over my 15 years have grown increasingly dire… notwithstanding recent reforms.' gao.gov · original document
- U.S. Government Accountability Office, Weapon Systems Annual Assessment: DOD Is Not Yet Well-Positioned to Field Systems with Speed (GAO-24-106831) (2024-06-17) — The 2024 edition: average expected time to deliver even an initial capability was 10 years — 'a time frame incompatible with maintaining military advantage'; MDAPs that delivered capability took 11 years against 8 planned; the MTA sample then stood at $35 billion across 20 programs. gao.gov · original document
- U.S. Department of War, Office of the Under Secretary for Acquisition & Sustainment, Rebuilding the Arsenal of Freedom: Acquisition Transformation Strategy (2025-11-07) — The November 2025 reform strategy: three overarching outcomes ('field technology and modernize systems at a rate that outpaces our adversaries'; wartime surge capacity; 'a wartime footing… from a culture of compliance to one of speed and execution') and five pillars. The promise document the 2026 assessment is read against. Dated per the companion memorandum's November 7, 2025 stamp (files posted November 10). Captured via Wayback (media.defense.gov blocks datacenter fetches). media.defense.gov · original document
- U.S. Department of War, Office of the Secretary, Transforming the Defense Acquisition System into the Warfighting Acquisition System to Accelerate Fielding of Urgently Needed Capabilities to Our Warriors (Secretary of War memorandum) (2025-11-07) — The Secretary of War's November 7, 2025 order directing the transformation: 'put our Acquisition System and Enterprise on a wartime footing and dramatically accelerate the fielding of new technology.' Captured via Wayback. media.defense.gov · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
Every June or July, measures the same thing: how long the Pentagon's costliest weapon programs take to put an initial capability in a warfighter's hands. The 2024 edition⧉ said 10 years — "a time frame incompatible with maintaining military advantage." The 2025 edition⧉ said almost 12, up a year and a half. The 2026 edition⧉, released July 2, says just over 12 — and warns the number is optimistic by construction, because programs are delaying every interim milestone while refusing to move their delivery dates. In between the last two measurements, the Secretary of War ordered the entire acquisition system onto "a wartime footing"⧉ — the newest entry in a stack of speed reforms that 's own trend line has yet to register.