Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a consulting proposal that wins the engagement — outcomes over hours. Use when asked to write a consulting proposal, a project proposal, a pitch for a client engagement, or to respond to an RFP. Produces a proposal — the client's problem in their words, your approach & deliverables, outcomes/value, timeline & phases, investment with options, and why-you — framed around results, not a task list. Ready to export as a designed PDF.
.claude/skills/mohitagw15856-consulting-proposal/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 76% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -63% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -41% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 96% | 0% |
Clients don't buy hours — they buy an outcome and the confidence you'll deliver it. Losing proposals lead with the consultant's process and a flat day-rate; winning ones lead with the client's problem and the value of solving it. This skill writes a proposal framed around results, with tiered options that anchor on value — ready to drop into the themed PDF export.
Ask for these only if they aren't already provided:
1. The problem (their words) — restate their situation and the cost of the status quo. Show you get it before you pitch. This earns the read.
2. Objectives & outcomes — what success looks like, in their metrics. Lead with the value, not the activity.
3. Approach & deliverables — the phases and the concrete artifacts they'll receive. Enough detail to build confidence, not a padded task list.
4. Timeline — phases with milestones and rough dates.
5. Investment — the price, framed against the value/cost-of-inaction. Offer 2–3 tiered options (e.g. core / recommended / comprehensive) — options shift the conversation from "yes/no" to "which," and anchor on the bigger one. State what's included per tier and what's out of scope.
6. Why me/us — relevant proof: comparable results, credentials, a short case reference. Brief.
7. Next step — one clear action to move forward (sign, a kickoff call, a deposit).
Value-based consulting-proposal practice (Alan Weiss-style outcomes-over-hours, tiered options, anchor on value).
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 14,420 | 16,176 | +12% | 1 | 1 | 0% | 2,659 | 3,616 | +36% | 0 | 0 | — |
case-02 | fail→pass | 7,926 | 10,300 | +30% | 1 | 1 | 0% | 1,469 | 2,592 | +76% | 0 | 0 | — |
case-03 | pass→pass | 10,986 | 6,868 | -37% | 1 | 1 | 0% | 1,967 | 1,783 | -9% | 0 | 0 | — |
case-04 | pass→pass | 12,517 | 13,693 | +9% | 1 | 1 | 0% | 2,415 | 3,075 | +27% | 0 | 0 | — |
case-05 | fail→fail | 6,326 | 5,638 | -11% | 1 | 1 | 0% | 1,071 | 1,685 | +57% | 0 | 0 | — |
case-06 | pass→pass | 11,930 | 14,084 | +18% | 1 | 1 | 0% | 2,066 | 3,185 | +54% | 0 | 0 | — |
case-07 | pass→pass | 14,072 | 14,943 | +6% | 1 | 1 | 0% | 2,700 | 3,359 | +24% | 0 | 0 | — |
case-08 | fail→pass | 31,985 | 9,601 | -70% | 1 | 1 | 0% | 5,774 | 2,121 | -63% | 0 | 0 | — |
case-09 | pass→pass | 22,070 | 15,604 | -29% | 1 | 1 | 0% | 4,164 | 3,321 | -20% | 0 | 0 | — |
case-14 | pass→pass | 4,934 | 8,321 | +69% | 1 | 1 | 0% | 798 | 1,987 | +149% | 0 | 0 | — |
case-10 | pass→pass | 11,160 | 7,629 | -32% | 1 | 1 | 0% | 2,070 | 1,896 | -8% | 0 | 0 | — |
case-11 | pass→pass | 15,569 | 10,879 | -30% | 1 | 1 | 0% | 2,812 | 2,432 | -14% | 0 | 0 | — |
case-12 | fail→fail | 6,255 | 7,868 | +26% | 1 | 1 | 0% | 1,190 | 2,079 | +75% | 0 | 0 | — |
case-13 | fail→fail | 11,664 | 13,158 | +13% | 1 | 1 | 0% | 1,799 | 2,915 | +62% | 0 | 0 | — |
case-15 | fail→pass | 21,981 | 13,396 | -39% | 1 | 1 | 0% | 4,864 | 2,889 | -41% | 0 | 0 | — |
case-16 | pass→pass | 19,383 | 18,778 | -3% | 1 | 1 | 0% | 3,784 | 4,021 | +6% | 0 | 0 | — |
case-17 | pass→pass | 5,681 | 5,831 | +3% | 1 | 1 | 0% | 1,058 | 1,662 | +57% | 0 | 0 | — |
case-18 | pass→pass | 14,827 | 12,783 | -14% | 1 | 1 | 0% | 2,527 | 2,814 | +11% | 0 | 0 | — |
case-19 | fail→fail | 8,419 | 10,027 | +19% | 1 | 1 | 0% | 1,535 | 2,364 | +54% | 0 | 0 | — |
case-20 | pass→pass | 13,832 | 12,313 | -11% | 1 | 1 | 0% | 2,276 | 2,725 | +20% | 0 | 0 | — |
case-21 | fail→pass | 7,471 | 10,735 | +44% | 1 | 1 | 0% | 1,174 | 2,300 | +96% | 0 | 0 | — |
case-22 | pass→pass | 11,298 | 9,607 | -15% | 1 | 1 | 0% | 1,994 | 2,389 | +20% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +23 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.