Four artifacts, shown in full. Everything on this page is a constructed demonstration — the same made-up bidder as the Replay sample (“Harbor Data Systems”), the same made-up state DOT solicitation, this time with a draft technical volume run through the board twelve days before the deadline. The pursuit is synthetic. The method is the real one.
The Gatekeeper runs before anything is scored, because real boards screen before they read. Every Section L instruction, graded. synthetic demonstration
| Section L instruction | Status | Finding |
|---|---|---|
| L.4.2 — Volume II shall not exceed 30 pages | FAILED | 31.5 pages once resumes are counted. Amendment 2 changed the counting rule to include appendices; the draft still assumes the original rule. This alone is grounds for elimination. |
| L.4.5 — 12-point font, 1-inch margins | met | Compliant throughout, including tables. |
| L.5.1 — Acknowledge all amendments on SF30 | AT RISK | Amendments 1 and 2 acknowledged; Amendment 3 (posted 4 days ago) is not. The form is drafted but unsigned. |
| L.6.3 — Staffing plan as Attachment J-4 format | met | Format matches; two labor categories use titles that differ from J-4's — flagged for consistency, not compliance. |
| L.7.1 — Submit via the state portal by 14:00 CT | AT RISK | No dry-run submission has been made. Portal accounts lapse after 90 days; last login was 6 months ago. |
Eight minutes, the way a tired evaluator actually reads: headings, topic sentences, callouts, captions. The trace records what was seen, in order, and what never was. synthetic demonstration
| Minute | What the Skimmer saw | Result |
|---|---|---|
| 0–1 | Executive summary, read in full | Win theme 1 (state-portal integration record) lands — it's the first sentence. |
| 1–3 | §2 headings + topic sentences | §2.1's heading promises a 'proven approach' with no proof in the first sentence. Skipped to the figure. The figure caption carries no claim. |
| 3–5 | §3 staffing — table scanned, prose skipped | Win theme 3 (named PM with DOT history) is buried in paragraph four. Never seen. |
| 5–7 | §4 transition plan headings | The 30/60/90 callout box lands. Win theme 5 (zero-downtime cutover) is findable. |
| 7–8 | §5 skimmed, appendices skipped | Win themes 6 and 7 live in an appendix. An evaluator giving you eight minutes does not read appendices. |
Verdict: 3 of 7 win themes findable in one pass.Nothing here is a compliance problem. All of it is points left on the table — and it’s fixable by moving sentences, not rewriting sections.
Each evaluator writes findings in debrief language, tied to the sentence — or the silence — that caused them. An excerpt across three of the ten worksheets. synthetic demonstration
The terms are FAR 15.001’s: a weakness increases the risk of unsuccessful performance; a significant weakness appreciably increases it; a deficiency is a failure to meet a requirement.
| Evaluator | Factor | Finding | Note |
|---|---|---|---|
| The SME | Technical approach | Weakness | §2.3 claims a 40% reduction in manual review 'based on prior deployments' — no deployment, customer, or measurement is named. Unsubstantiated as written. |
| The SME | Technical approach | Strength | §4's cutover plan names the rollback trigger, the owner, and the threshold. Checkable and specific — evaluators can score this. |
| The Points Accountant | Management | Significant weakness | M.3.2 awards points for 'demonstrated key-personnel retention.' The draft never addresses retention. Best material on this sits in the appendix, which M does not evaluate. |
| The Past-Performance Hawk | Past performance | Weakness | Reference 2 is relevant and recent but one-fifth the contract value. The stated test includes scale; the draft argues similarity instead of addressing it. |
| The Cost Realist | Cross-volume | Weakness | The staffing table totals 11,840 hours; the narrative commits to 'day-one full staffing' that the ramp in the table contradicts. One of them is wrong. |
The board’s combined read, and the fix list ranked by score impact — what actually changes points, in order. synthetic demonstration
| # | Fix | Why it ranks here |
|---|---|---|
| 1 | Cut Volume II to 30 pages under Amendment 2's counting rule | Elimination risk. Nothing else on this list matters if the volume is thrown out. |
| 2 | Acknowledge Amendment 3; sign the SF30; dry-run the portal | Two more screens closed in under an hour. |
| 3 | Answer M.3.2 retention in the management section, not the appendix | Clears a significant weakness in a factor worth 25 points. |
| 4 | Name the deployment behind the 40% claim, or cut the number | An unsubstantiated claim reads worse than a modest one. |
| 5 | Move win themes 3, 6, 7 into first sentences under their headings | Raises findability from 3 of 7 to 7 of 7 without adding a page. |
What the board didn’t say:whether this DOT trusts Harbor, what the incumbent will bid, or whether the price wins. It can’t know those. A mock evaluation is a measured read of a draft under stated conditions, not a prediction of award — the judgment calls stay with people, which is the point.
Book a call and we’ll run the board on a live volume with you — screen, traces, worksheets, and the ranked fixes, before the deadline instead of after the debrief.
Run one on your draft →