# Support & Research Ops Sample

This is a sanitized sample of the kind of remote support/research workflow I can run for a client or team. It avoids private inboxes, accounts, customer data, and platform-specific confidential material.

## Task

Turn a messy queue of leads, emails, research items, or support requests into a clear working tracker that a human can review quickly.

## Intake Fields

| Field | Purpose |
| --- | --- |
| Source | Where the item came from: email, form, job board, Reddit, CRM, spreadsheet, or website. |
| URL / thread ID | Makes every claim traceable. |
| Pay / value | Incentive, budget, revenue signal, or urgency. |
| Fit | High, medium, low, or unknown. |
| Risk | Scam, privacy, payment, policy, or missing-info risk. |
| Status | Lead, screened, staged, submitted, saved, blocked, dead. |
| Next action | The smallest concrete next step. |
| Human-needed field | Anything that needs approval or personal info. |

## Workflow

1. Collect the raw queue without changing external state.
2. Deduplicate repeated leads or duplicate messages.
3. Classify each item by type, urgency, pay/value, and risk.
4. Summarize only facts supported by the source.
5. Draft next actions, but keep outbound messages unsent until approved.
6. Record decisions so similar future items get faster.

## Example Output

| Item | Status | Next Action | Notes |
| --- | --- | --- | --- |
| Paid study invite | staged | Open screener and answer only confirmed facts | Stop on demographics, health, income, or schedule questions. |
| Remote support role | saved | Use support/ops resume and verify direct employer | Reject if staffing/scam signals appear. |
| Reddit research gig | draft-ready | Check subreddit rules, write short reply, require payment scope | No unpaid full sample. |
| Customer support backlog | ready | Create category labels and response macros | Human approval before any customer-facing send. |

## Why This Helps

The value is not just speed. The value is preserving source links, not losing state, making blockers explicit, and giving a human a queue they can trust.