What the first hours of a new Marketplace search look like

The first hours after you activate a Facebook Marketplace search usually look like a short flood of fitting listings, then a quieter trickle. That shape is normal. The first sweep collects stock already live on Marketplace. Later hours are mostly genuinely new posts. A lull after the fill is not proof the search broke.

See that shape on a no-card Day Pass. One search, 24 hours. Pass facts: Facebook Marketplace Day Pass.

What happens in the first sweep

The first sweep surfaces listings that already pass your filters and were already on Marketplace when you activated.

Crawlbench runs a first check when a search goes live. From then on, it checks on your plan's schedule and sends an alert when a new listing passes your filters, in every city on the search. Those early rows land in your results so you can triage. They are not a private firehose of only "just posted" inventory. Facebook's feed is not a clean stream of brand-new posts.

If you expected zero rows until the next seller hit Publish, your results will feel too busy. If you expected a phone storm for every baseline row, the channels will feel too quiet. Both reactions miss the same mechanism: fill your results, keep outbound noise for fresher posts.

Why alerts stay quieter than your results

Alerts for new listings only go out for listings Facebook posted recently. Older baseline rows stay in your results without flooding email, Telegram, Slack, Discord, or webhooks.

That split is intentional. A new search can pass dozens of cars or rentals that have been sitting for a day. Sending every one of them to your phone trains you to ignore the channel. Price-cut alerts on listings you already have are not limited to recent posts.

Set channels under alert preferences. Turn off one noisy search if it is still too chatty after the settle.

Posted vs Seen in the interface

Posted is Facebook's post time when Crawlbench has it. Seen is when Crawlbench first recorded the listing.

Listing rows prefer Posted. When post time is missing, the row shows Seen until enrichment backfills a real Facebook time (or gives up). That is why two timestamps can disagree, and why a listing can appear in your results before alerts treat it as fresh enough to send.

Do not treat Seen as "the seller just listed." Treat it as "this search noticed the listing."

What a healthy first day looks like

A healthy first day usually has a noticeable fill of results early, then a steady flow of new posts, each alerted on the next check after it passes your filters.

Use this checklist before you rewrite filters or open a support chat:

1. Coverage on the dashboard looks healthy (checks are landing), not stuck catching up
2. Early results include older posts that still pass your filters
3. Phone and chat alerts are reserved for fresher new listings, not every baseline row
4. After the settle, volume tracks the market and your filters, not the first-hour spike

If your results stay empty while coverage is healthy, open filter pressure in the search details or the empty feed. It shows seen versus passed today and which dimensions fail most. Loosen one dominant fail before you change everything. Setup steps live in create your first search. More on timing: what alert timing means.

Steady state is the product

Steady state is scheduled checks plus a pass/fail gate: every alert is a listing that passed your filters.

Facebook does not publish an SLA for how feeds or native notifications behave, and its timing varies by account and day. Crawlbench checks busy cities every 10 minutes on Starter, every 5 on Growth, every 2 on Pro and every minute on Scale, as pricing shows on 18 October 2026. The alert goes out on the check that finds the listing, with filters you can defend and alerts that stay worth opening after the first fill. You message sellers on Facebook yourself, so your account stays safe.

If you want to see that loop on your own filters, start from features or activate one tight search and follow the first day without panicking at the shape.