---
name: sponsorship-package-evaluator
description: Evaluates event sponsorship and exhibitor packages from the buyer's side. Breaks a prospectus into deliverables (booth, speaking slot, attendee list access, lead scans, branding, passes, meetings), estimates value and cost per expected meeting or lead, compares tiers, flags vague or unmeasurable items, and drafts negotiation asks with a go, renegotiate, test small or decline recommendation. Use when deciding whether to sponsor, comparing sponsorship tiers or negotiating with an event organizer.
license: MIT
metadata:
  author: eventmarketingsoftware.com
  version: "1.0"
  category: sponsorship
  source: https://eventmarketingsoftware.com/skills/sponsorship-package-evaluator
---

# Sponsorship Package Evaluator

Reads a sponsorship prospectus the way a buyer should: what you're really buying, which pieces create buyer conversations, what each conversation costs, and what to ask for before signing. Prospectus decks are written to upsell. This skill gives you an evidence-based answer instead.

## Intake

If the user has an `event-context.md` file, read it first.

Ask for whatever is missing in ONE message:

1. The prospectus (PDF, pasted text or a list of tiers with prices and items). Several offers are fine.
2. Your goal for this event: pipeline meetings, brand awareness in a segment, customer retention, partner or recruiting. Pick one primary.
3. Your ICP: industries, company size, titles you need to reach.
4. Budget range and whether booth build, travel, staff time and promo are in a separate budget.
5. The organizer's audience data: attendance, share in your ICP, seniority mix. Note whether it's audited, survey-based or a claim.
6. Last year's results if you've been before: meetings, leads, opportunities, pipeline, closed revenue.
7. Your sales follow-up capacity (how many leads your team can work in the first two weeks).

If the organizer gave no audience data, say so plainly and make getting it the first negotiation ask.

## Data sources (connectors optional)

- If a Google Sheets connector is available, read past event results from it and offer to write the deliverables table to a new tab.
- Otherwise use the uploaded PDF, pasted text or CSV/XLSX, and return Markdown tables plus a downloadable CSV.
- Never require a paid service. Never fetch URLs, including the organizer's site; ask the user to paste what you need.

Sample input: [assets/sample-prospectus-items.csv](assets/sample-prospectus-items.csv).

## Method

1. **Itemize.** One row per deliverable per tier. Use these types: Booth or space, Speaking or content, Meetings (hosted buyer, matchmaking, appointments), Lead capture (scanner licenses, session scans), Attendee data (pre or post list, opt-in only or full), Branding (logos, signage, lanyards, app banners), Hospitality (dinner, reception, lounge), Passes, Digital (email, newsletter, social, app push), Exclusivity, Other.
2. **Rate measurability.** For each item: `Measurable` (a count you can verify: meetings, scans, check-ins, list size), `Proxy` (impressions, reach claimed by organizer) or `Vague` ("premium visibility", "thought leadership positioning"). Vague items get a clarifying question.
3. **Estimate interactions.** For items that create buyer conversations, estimate a low, likely and high count of ICP interactions using the user's history first, then the organizer's data, then a stated assumption. Show the formula. Example: booth traffic estimate = attendees x ICP share x expected booth visit rate. Mark every assumption.
4. **Assign value.** Use the user's own benchmark if they have one (their cost per meeting or cost per lead from other channels). Value an item at what it would cost to get the same interactions elsewhere. Branding and Proxy items get a value only if the user gives a rationale; otherwise list them as "unpriced, nice to have". See [references/valuation-method.md](references/valuation-method.md).
5. **Add hidden costs.** Booth build and shipping, drayage and show services, travel and lodging per staffer, staff time, promo and giveaways, lead retrieval licenses if not included. Ask the user for numbers or leave clearly labeled placeholders. Never present the sponsorship fee as the full cost.
6. **Compute unit costs.** Total cost per expected ICP meeting and per expected qualified lead, for low, likely and high cases. Compare against the user's benchmark and other channels if given.
7. **Compare tiers.** Show where the next tier up adds measurable interactions vs only branding. Often the best value is a lower tier plus one a la carte item that creates meetings.
8. **Flag risks.** Attendee list without opt-in (GDPR and the UK GDPR make it hard to use for EU or UK contacts), no audited attendance, lead scans limited to one device, speaking slot at a low-attendance time, no cap on competing sponsors in your category, cancellation or force majeure terms that keep your money.
9. **Draft negotiation asks.** Prioritize asks that add measurable interactions at low cost to the organizer: guaranteed meetings with named ICP accounts, extra scanner licenses, a better session time, session attendee scans, an opt-in attendee list, competitor exclusivity, extra passes, payment terms, a make-good clause for missed deliverables, early access to next year's renewal at this year's price.
10. **Recommend.** One of `Go`, `Renegotiate`, `Test small` (lowest tier or attend and book meetings only) or `Decline`, with the deciding numbers and what would change the answer.

Domain rules that matter:
- Cost per meeting beats cost per lead for B2B pipeline goals. A badge scan isn't a meeting.
- Expected value assumes your team follows up. If follow-up capacity is lower than expected leads, cap the value at capacity.
- Treat organizer attendance numbers as claims unless audited. Ask whether the number is registrations, unique attendees or total badge scans.
- First-year sponsorships carry more risk; `Test small` is a real option, not a failure.

## Output templates

**Deliverables table**

| Tier | Item | Type | Measurability | Expected ICP interactions (low / likely / high) | Assumption | Value basis | Question for organizer |
|---|---|---|---|---|---|---|---|

**Tier comparison**

| Tier | Fee | Hidden costs | Total cost | Expected ICP meetings (likely) | Cost per meeting (low / likely / high) | Expected qualified leads | Cost per qualified lead | Unpriced extras |
|---|---|---|---|---|---|---|---|---|

**Negotiation asks** (ranked)

| # | Ask | Why it matters | Cost to organizer | Fallback |
|---|---|---|---|---|

**Recommendation**

```
Recommendation: [Go | Renegotiate | Test small | Decline] on [tier]
Deciding numbers: [cost per meeting likely vs benchmark], [total cost vs budget]
What would change it: [e.g. organizer guarantees 15 ICP meetings]
Biggest risk: [one line]
```

Then the draft email to the organizer with the top 3 to 5 asks. Draft only.

## Validation checklist

- [ ] Every prospectus item appears once; items in equals rows out.
- [ ] Every number traces to the user's data, the organizer's data or a labeled assumption.
- [ ] Hidden costs included or shown as placeholders; fee alone is never called total cost.
- [ ] Vague items have a clarifying question.
- [ ] Unit costs shown as low, likely and high, not a single falsely precise number.
- [ ] Recommendation states what would change it.
- [ ] No vendor or organizer named as better than another beyond what the numbers show.

## Approval gates

- The organizer email is a draft. Never send it or share the analysis outside the chat without a clear yes.
- If writing results to a sheet, show the rows first.

## Privacy

- If the package includes an attendee list, ask whether it's opt-in. Flag EU and UK contacts: you'll need a lawful basis to email them, and many lists can't be used for cold email.

Benchmarks and sources: [references/benchmarks.md](references/benchmarks.md).
