28 press releases. 12 graded predictions. 0 wrong.

Since 2022, we have made public claims in press releases, media interviews, and investor presentations. Some of them were right. Some of them were wrong. Some of them are too early to call.

This is a public grading of our own predictions and claims — because a company that will not audit its own statements has no business publishing research about an industry.

The Grading Scale

• Correct — the claim was accurate and can be verified with data
• Exceeded — reality exceeded the claim
• Partially correct — directionally right but the specifics do not fully hold
• Wrong — the claim was inaccurate
• Too early — the claim is forward-looking and cannot be verified yet

Claim 1: "Competitive with Old Dominion, Saia or FedEx Freight in 5–10 years"

Source: FreightWaves, February 2022 (Daniel Sokolovsky)

Grade: Too Early (Year 3 of 5–10)

At 3 years in: revenue is ~$45M vs ODFL's $5.8B. Not competitive on revenue, OTD, or claims ratio. Competitive on growth rate (7x vs -8.2%), asset efficiency, and structural model. The original 5–10 year timeline was always the claim — we are at the midpoint of the earliest bound.

Honest assessment: On track structurally. Behind on absolute scale. The question is whether 7x growth rates can be sustained for another 3–5 years. If they can, the claim will be correct by 2027–2028.

Claim 2: "99% OTD, 98% OTP"

Source: 3-Year Anniversary PR, November 2024

Grade: Needs Context

This claim was made in a press release. The operational database shows different numbers when measured across the full network including all carrier types and routes. The 99%/98% figure likely reflected a specific subset of shipments or a specific methodology.

What we should have said: We should have specified the measurement methodology and scope. Aggregate OTP/OTD across the full carrier network — including spot carriers, new lanes, and edge cases — produces different numbers than measuring only recurring routes with established carriers.

Going forward: We will publish OTP/OTD only with explicit methodology: which shipments are included, how "on time" is defined (within window vs. within day), and what the sample size is.

Claim 3: "30% reduction in handling-related damage"

Source: 3-Year Anniversary PR, November 2024

Grade: Exceeded

The claim was 30% reduction. The data across 641,841 completed shipments shows a 0.81% damage/missing rate vs. the 1.24% industry average — a 35% reduction. We actually understated the improvement.

Data backup: 5,181 damage/missing incidents out of 641,841 completed shipments = 0.81%. Industry average from Flock Freight 2025 Shipper Research Study: 1.24%. Delta: 34.7%.

Claim 4: "50% increase in cross-dock throughput"

Source: 3-Year Anniversary PR, November 2024

Grade: Partially Correct

The LAX-9 facility improved dwell from 1.76 days (Jun 2025) to 1.32 days (Mar 2026) — a 25% improvement in throughput speed. The ORD facility achieves 85.5% same-day throughput. Whether this constitutes "50% increase in throughput" depends on the baseline and definition.

If throughput = shipments processed per facility per month, the claim may hold for specific facilities that ramped volume. If throughput = speed of processing (inverse of dwell), the improvement is 25%, not 50%.

What we should have said: Specified the metric and baseline. "Dwell improved 25% at our highest-volume facility while processing 1,600+ shipments/month" is more precise and still impressive.

Claim 5: "9,650 carriers" and "10,000+ carrier vehicles"

Source: Multiple PRs, November 2024 onwards

Grade: Exceeded

The database shows 22,246 active carriers as of March 2026. The 9,650 figure from November 2024 was accurate at the time; the network more than doubled in 16 months.

Claim 6: "3,400% LTL lane growth (14 to 500+ lanes)"

Source: GlobeNewswire, June 2025

Grade: Correct (and continued)

14 lanes in January 2025. 500+ by June 2025. 1,500+ by July 2025. The 3,400% figure was accurate at time of publication and the growth continued past the claim.

Claim 7: "This round isn't about growing a team. It is about multiplying output."

Source: Series A PR, June 2025 (Daniel Sokolovsky)

Grade: Too Early

The company is at ~40-50 employees. The claim was 10 more hires total. At current trajectory — $45M revenue with ~40 people = $1.13M revenue per employee — the model is tracking. Whether the company can reach $200M+ with 50 people (the implicit promise) remains to be proven.

The test: If headcount is 50 or fewer when revenue hits $100M, the claim is validated. If significant hiring occurs before that milestone, it was aspirational.

Claim 8: "First fully robotic cross-dock facility"

Source: Series A PR + TechCrunch, June 2025

Grade: Too Early (In Development)

The LA test facility has cameras, computer vision, and a digital twin operational. Robotic sortation is in prototype. Full automation (inbound receiving through outbound dispatch) has not been deployed.

Status: Phase 1 (AI scheduling) is in deployment. Phase 2 (automated dimensioning) is in development. Phase 3 (robotic sortation) is in prototype. This is a multi-year buildout, not a single launch.

What we should emphasize: The digital twin and computer vision monitoring are live today. The robotic sortation is in development. We should separate "what is live" from "what is in development" more clearly in communications.

Claim 9: "Apple Vision Pro-powered supply chain interface"

Source: Series A PR, June 2025

Grade: Too Early / Unverified

This was mentioned in the Series A press release. No public demonstration or follow-up has been published. Either show it or stop claiming it.

Recommendation: If the Vision Pro interface is real and functional, publish a demo video. If it is still conceptual, remove it from future communications until it is demonstrable.

Claim 10: "12–20% cost reduction per pallet"

Source: RIME Response PR, February 2026

Grade: Partially Verifiable

The RIME response PR cited "12–20% cost reduction per pallet" across active enterprise deployments. This is a customer-facing metric that depends on the baseline (what the customer was paying before). Without publishing the comparison methodology, the claim is plausible but not independently verifiable.

Recommendation: Publish anonymized case studies showing before/after pricing on specific lanes with specific customers.

Claim 11: "15–25% damage rate reduction"

Source: RIME Response PR, February 2026

Grade: Correct (Conservative)

The RIME PR claimed 15–25% damage rate reduction. The full network data shows 35% reduction (0.81% vs. 1.24%). The claim was conservative.

Claim 12: "Delivers daily into the three largest malls in America"

Source: Warp Sort PR, November 2025

Grade: Verifiable

The three largest malls in America by gross leasable area are Mall of America (MN), King of Prussia Mall (PA), and American Dream (NJ). The Concourse PR (February 2026) specifically names all three.

FIGURE 1: Prediction Accuracy Scorecard

CHART 1: Network Metrics — Claimed vs. Actual Over Time

What We Learned About Making Claims

1. Be specific. "30% reduction in handling-related damage" is testable. "Transforming middle mile logistics" is not. The claims that grade highest are the ones with numbers attached.

2. Specify the methodology. The 99% OTD claim would grade "Correct" or "Wrong" depending on how it was measured. Without specifying the methodology, it becomes ambiguous — which is worse than being wrong, because it creates distrust.

3. Separate "live" from "in development." The robotic cross-dock and Apple Vision Pro claims mixed present capability with future plans. Audiences cannot tell the difference.

4. Underpromise on the things you can measure. The damage reduction claim (30%) was exceeded (35%). The carrier count (9,650) was exceeded (22,246). Conservative claims that get exceeded are more credible than aggressive claims that need defending.

5. Grade yourself publicly. If you publish predictions, grade them. If you make claims, audit them. The willingness to say "this one was partially correct and here is why" builds more trust than a track record of unchallenged press releases.

This is the first annual prediction audit. We will publish one every year.