Your selling analytics
What each listing measures, how to read hotter and colder, and how the AI turns the signals into listing, pricing, and sell-or-hold recommendations.
TL;DRAI summary
This chapter explains how to read your vehicle's selling analytics, like views and offers, to help you decide whether to adjust its price, promote it, or accept an offer. It also shows how the system can recommend actions based on these signals.
For: Sellers — this is your instrument panel, per vehicle and per promotion. The platform-wide dashboard is the CM's chapter; different tool, different objective.
Every listing measures its own demand. The seller's job isn't to admire the charts — it's to decide: cut, promote, hold, or take the offer. This chapter is about reading the signals well enough to do that, and where the AI does the reading for you.
What gets collected
Per vehicle, per day, per channel:
- Views — a buyer opened the listing (deduplicated; a refresh isn't interest).
- Saves — the strongest pre-offer signal: someone wants to watch it. Unsaves count too, honestly.
- Offers — demand with a number on it, including Buy Now.
- Where it came from — every event attributes to its door: Public marketplace, Promotion (down to the campaign email), Direct link (down to which link), Storefront.
- On promotions: the full funnel — sent → delivered → opened → clicked → viewed → saved → offered — plus geography, device, and per-visitor drill-ins, so "who exactly is looking" has an answer.
What's deliberately not collected: anything from logged-out browsing that could identify a person. Signals are aggregated demand, not surveillance.
Reading hotter and colder
The temperature badge (HOT / WARM / COLD, with a trend arrow) blends recent views, saves, and offer velocity — recency-weighted, so last week counts and last month fades. The nuance is in the history, not the word:
- Cooling with traffic (views steady, saves falling) — the market looked and moved on: the price is the suspect.
- Cooling without traffic — the listing isn't being seen: reach is the suspect (channels, promotion).
- Warming with saves piling up — watchers waiting for a move; a small step converts them.
- The event markers are the sharp tool: every price change, photo update, promotion entry, and share-link creation lands on the timeline with its measured before/after — "the campaign moved views +180%" is read off the chart, not guessed.
The Market History report
The vehicle panel's Market History tab assembles the whole biography — launch snapshot, buyer reach, the engagement series with event impacts, valuation context, days on market — and exports as a PDF: the document for the conversation about a van that isn't moving. What we did, what each move changed, what the market is saying.
The decision playbook
| Signal | First move | Where |
|---|---|---|
| Cold + falling, low traffic | Reach: add a promotion, check channels | Channels tab |
| Cold + falling, decent traffic | Price: check the Sequoia delta, consider the markdown ladder | Pricing |
| Warm, saves accumulating | Nudge: a visible small cut converts watchers | Pricing |
| Hot, no offers | The ask is testing the market's ceiling — hold or trim | Sequoia's read |
| Hot + offers in the band | Decide the offer, not the listing | Offers |
What Sakura recommends here
Listing and offering recommendations, on demand. Ask on any vehicle — "how should I list this?" — and Sakura answers with readiness (photos, condition, valuation), a buyer-pool estimate, channel advice, and timing. On live listings it reads this chapter's signals for you: the Market History summary regenerates every time you open the tab, and its last sentence is always one concrete next step.
What Sequoia sees here
Pricing and sell-or-hold. Sequoia owns the market read: the valuation in context, comparables cited, the trend, and a plain call — "SELL — vans in this band trending down 2%/month" or "HOLD — segment tightening, revisit in three weeks." The ±% deltas under your floor and ask are Sequoia's distance-from-market, everywhere prices appear.
How Cedar can help here
From signal to action in one sentence. The playbook above is Cedar's to execute: "this van is cold — fix it" returns a diagnosis (reach vs price) and a proposal — add to a promotion, or step the ask down floor-clamped — as cards you confirm. Reading the analytics is optional; acting on them never waits on a dashboard.
Worked example
The Transit ran WARM for ten days at ~30 views/week, saves flat — decent traffic, no conversion: price suspect, but Sequoia read HOLD (the ask was +3.5% over a firm market). Then the campaign email: +180% views in 48 hours, measured on the event marker; badge flips HOT; Dan's save and offer follow. No price action was ever taken — the analytics said reach, the promotion delivered it, and the deal came from the funnel the charts predicted.
Under the hood
Engagement rolls up daily per vehicle-channel; the report reconstructs launch-day state rather than storing it; event impacts correlate the notice stream against the engagement series. Sellers see their vehicles' signals — cross-org aggregates belong to the platform's dashboard.