# Ldg Brief The Vendor

> Writes a requirements brief for an outside training provider or a learning tool, producing the outcomes wanted, must-haves against nice-to-haves, the same questions and a demo script on your own content for every provider, selection criteria with weights the team sets and a decision record. Use for "run ldg-brief-the-vendor", "training rfp", "lms rfp", "vendor selection for training", "choose a training provider", "request for proposal training", "compare learning tools", "demo script for a vendor", part of The Complete Claude Guide for L&D Pack by 90n.

- Skill: `polar-bear-org/ldg-brief-the-vendor` (Agent Skill)
- Install (CLI): `npx get-skill add polar-bear-org/ldg-brief-the-vendor`
- Raw SKILL.md: https://api.skillmd.com/api/skills/polar-bear-org/ldg-brief-the-vendor/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: polar-bear-org (https://skillmd.com/u/polar-bear-org)
- Updated: 2026-11-14
- Page: https://skillmd.com/skills/polar-bear-org/ldg-brief-the-vendor

---


# Brief the Vendor

## When To Use
The tool was chosen from a feature list and a polished demo, and the regret arrives after the contract. Run this once the route is to buy or to partner and before anyone books a demo, for a provider, a bought course or a tool. It answers: what do we need this to do for which learners, how will we compare the offers, and who signs the choice?

## When Not To Use
If nobody has decided whether to spend or whether to build in-house, run Make the Business Case first; this skill chooses who or what, not whether. For a small purchase from a provider you already use, a short email with the need and the dates is enough.

## Inputs
- The need in your words, the learners by role, and what should change at work.
- The constraints you know: your data and privacy rules, the accessibility level, your LMS and your authoring tool, languages, budget range, dates.
- One piece of your own content and one real task for the demo, and who will score the proposals, by role.
If you have none of this, I start from the need and the learners by role and mark the output as a first draft.

## Approach
The method is a brief that specifies outcomes rather than features, with selection criteria and weights fixed before any proposal arrives, proposals scored independently by at least two people and then discussed to a consensus with reasons recorded. That sequence follows a UK Government guidance note on bid evaluation. It was written for public buying, so use the structure, not the rules. The judgement is in what you refuse to leave open until after the demos. The failure it prevents: three providers each demo their best screen on their own sample course, the team picks the one it enjoyed, and the first upload of real content shows what nobody asked.

## Workflow
1. Ask at most three questions: what must people be able to do, or what should change, once this is in place; which constraints would end a conversation with a provider; and who signs the choice and by when?
2. Write the need in one line, the learners by role, and the outcomes wanted. Turn each feature wish into the outcome behind it ("reporting" becomes "the L&D lead can see, by group, which items most people miss").
3. Sort the requirements. A must-have is something whose absence ends the conversation; keep that list short, and everything else is a nice-to-have with a reason. Then list the constraints: data and privacy, accessibility level, fit with your LMS and your authoring tool, languages, budget range, dates. Mark the contract, procurement-rule and data protection points: check with a qualified adviser.
4. Write the questions every provider answers, the same for all, and a demo script on YOUR content and one real task, with what to watch for at each step. A provider who will not demo on your content has told you something.
5. Set the selection criteria and their weights with the team and write them down BEFORE any proposal or demo. The balance of price and quality is the team's choice; I lay out the grid, as a spreadsheet file if file creation is on, and you fill it.
6. Plan the scoring: at least two people score each proposal alone, then a moderated discussion reaches one score per criterion, with the reason for any change recorded. Prepare the decision record: choice, runner-up, reasons, open risks.

## Output Format
```markdown
# Vendor Brief
Need: [one line] | Learners: [roles] | Route: [buy or partner] | Decision owner: [role]
## Outcomes and requirements
| Outcome wanted (what people will do, or what should change) | How we will know | Must-have or nice-to-have | Why |
|---|---|---|---|
| [outcome] | [evidence] | [type] | [reason] |

Constraints: [data and privacy, accessibility level, fit with your LMS and your authoring tool, languages, budget range, dates]; for a qualified adviser: [points]
## Questions and demo script, the same for every provider
Questions: [question 1]; [question 2]; [question 3]
| Step | On our content | What we watch for |
|---|---|---|
| [n] | [task with our file] | [what a good result looks like] |
## Selection criteria (fixed on [date], before proposals)
| Criterion | Weight | What a high score means |
|---|---|---|
| [criterion] | [weight the team sets] | [description] |
## Decision record
| Proposal | Agreed score by criterion | Reason for any change after discussion | Open risks |
|---|---|---|---|
| [proposal] | [scores] | [reason] | [risk] |
## Decision
[Named decider] signs the choice and the runner-up, with reasons, by [date]; [role] takes the contract points to a qualified adviser before signing.
```

## Done When
- Every requirement is an outcome or is marked as a feature wish with its reason.
- The criteria and weights carry a date earlier than the first proposal or demo.
- Every provider gets the same questions and demo script, and at least two scorers by role are named.

## Quality Bar
- No product or provider is named, recommended or ranked by Claude, and no price or market rate is invented; scores come from the people who read the proposals.
- Weights are never changed after a demo without a written reason and the decider's agreement.
- Contract, procurement-rule and data protection points are listed, never answered: check with a qualified adviser.
- Proposals are scored, never people; the team sets the weights before the demos, and a named person signs the decision.

## Next
Run ldg-plan-the-build (Plan the Build) to plan the work, with or without the provider.

## About the makers

This pack is made by 90n ([90n.ai](https://90n.ai)): AI-native ways of working and tools for Product, Marketing and HR teams, built by ex-McKinsey founders. We map where time and money leak, redesign the work around AI and stay until your people run it themselves, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).

