How we curate

Automation keeps the catalog fresh, but humans set the bar. Every guide and product detail page goes through the workflow below before it ships.

Where AI supports the newsroom

We lean on focused AI workflows to surface patterns faster while keeping editors accountable for the final word.

• Signal triage: Overnight monitors cluster sentiment, price swings, and restock data so editors wake up with a prioritized queue.
• Quality nudges: Lightweight models flag anomalies—like mismatched specs or sudden review shifts—so humans can re-check the evidence.
• Knowledge recall: AI summaries point back to archived testing notes and sourcing receipts, helping the team reuse hard-won research.

1. Signal-first sourcing inside our Trendwatch Airtable

   We track price histories, restock alerts, reader requests, and social chatter in a shared Airtable base. When the Velocity Score column crosses 18 and at least two editors confirm the "Reader pull" checkbox, the row promotes from "Ideas" to "Fact-check" for deeper review.

2. Hands-on verification in Notion's "Guide QA" board

   Editors inspect imagery, packaging, and seller credibility directly in the Guide QA kanban. If photos or specs are missing, we request better assets or replace the listing with something we trust before moving the card to "Ready to publish."

   Guide QA checklist excerpt

   Every approval is logged in Notion with these required fields before a card can leave the "QA" column:

   • Hero image status: Confirmed 2400px source uploaded and alt text drafted.
   • Affiliate link audit: Sponsored / nofollow toggles matched to contract and click-path spot check recorded.
   • Price guarantee timestamp: Latest price observation noted with retailer, currency, and screenshot URL.
   • Backup pick ready: Secondary product URL and copy snippet pasted for emergency swaps.
   • Approver signature: QA editor initials auto-stamped with time and Slack thread link.

   Screenshot reference: Notion → Guides workspace → Guide QA board, card view "Checklist".

3. Story-driven copy

   We draft blurbs that explain why the gift hits, how to stage the surprise, and what to pair it with. A second editor reads the copy aloud to keep the tone conversational and to catch jargon.

4. Compliance by default

   Disclosures, schema markup, and tracking parameters update alongside the copy. Links launch with sponsored, nofollow, and noopener attributes, and someone manually checks the click path.

5. Lifecycle tracking

   Once a pick publishes, automation checks stock and pricing thresholds. If something drifts, we pause the link, slot in a backup, and note the change in our changelog so readers see what shifted.

What qualifies as a win

Every guide needs a balance of wow-factor and practicality. We look for gifts that photograph well, arrive on time, and include thoughtful packaging that elevates the moment. If we can’t explain the story in a sentence—or if the price-to-delight ratio feels off—it doesn’t make the cut.

Tools that keep us accountable

• Catalog ledger: Documents when each product was last checked, who approved it, and which backup options are ready to go.
• Quality heatmaps: Surface guides that need a refresh based on click-through rate, return feedback, and stock volatility.
• Reader feedback loops: Surveys and inbox replies feed back into the sourcing queue so we can chase what you actually want.

Transparency at every step

We publish a live changelog and annotate every affiliate link with disclosures. When automation swaps an item, the changelog records the timestamp, the reason, and the editor who double-checked the replacement.

Review checklists by channel

Different categories demand different proof. We tailor the verification checklist based on where an item comes from so every pick clears the right hurdles.

• Marketplace finds: Confirm seller history, warranty coverage, and packaging condition with a fresh photo trail before a listing enters rotation.
• Direct-to-consumer drops: Interview founders when possible, audit shipping SLAs, and test customer support response times during off-hours.
• Indie maker submissions: Verify production capacity, sustainable sourcing claims, and whether the maker can support restock surges.
• Reader tips: Trace the original buyer experience, collect unfiltered feedback, and only move forward once we can reproduce the delight ourselves.

When we pass on a product

Declining an item leaves a paper trail too. We log the reason, the evidence we reviewed, and what would need to change before we reconsider it.

• Inconsistent pricing or vague fulfillment timelines that could sour a gifting moment.
• Packaging, instructions, or onboarding flows that leave our testers confused.
• Returns feedback that signals structural issues, not one-off delivery hiccups.
• Claims we can’t verify through receipts, lab data, or hands-on time.

Care and feeding of the catalog

The work doesn’t stop at publish. Automation pushes alerts into Slack when stock dips, competitors undercut pricing, or retailer terms shift. Editors triage every ping, refresh copy where needed, and pull backup picks from a bench we update weekly.

All of that rigor points at a single goal: dependable gifting experiences. By pairing transparent documentation with human judgment, we help readers celebrate the people they love without worrying whether a surprise will land.