Throughput strip — engineering spec

Throughput strip (PRESCRIPTIVE — Sales Analytics hero)

What it is: a row of four big clickable counts at the top of Sales Analytics — how many customers reached out, showed up, got desked, and bought inside the selected period. Every tile is a door into the list behind it.

1 · Anatomy — four tiles, no connectors

Desktop — one band, equal tiles. Values are a real mid-size store over 30 days.

Mobile — 2×2 grid, same four tiles. The band sits directly under the page header, above the hero tiles, and follows the dashboard's date range and location filter like everything else on the page. Four tiles, each a click-through. On the screens the strip and the keeping-up row compose as a single eight-tile grid with one uniform gutter — the strip is the top row. There are no chevrons or arrows between tiles — a deliberate refusal: connectors imply a funnel, and these numbers do not divide.

2 · The four contracts — tooltip copy and what it measures

Tile	Tooltip (what the user reads)	What it measures
Engaged	“Customers who reached out — a text, email or call in — during this period.”	Distinct customers with an inbound SMS, email or call in range, timeline-derived. Deliberately the raw inbound count with no deal filter — “customers who reached out” is the intuitive read and needs no asterisk.
Visited	“Sales appointments that actually happened in this period.”	Sales-type appointments in range with status completed or in_showroom, or a visit start recorded. Service and delivery appointments never count.
Desked	“Deals that hit the desk in this period — numbers were worked.”	Deals whose desking-lane entry falls in range, regardless of where the deal is now — a desked deal that sold still counts, a desked deal that died still counts. Never the raw desking-lane count, which accumulates forever (one store: 730 all-time vs 601 in 30 days).
Sold	“Deals sold in this period. Matches the DMS record when one exists.”	closeDate in range, any source, DMS-matched preferred. Never the closed-lane timestamp — at one store 85 DMS-confirmed sales in 30 days were never dragged to the Closed lane, so the lane number runs ~35% low, and a Sold tile that disagrees with the GM's DMS number kills trust in the whole strip.

3 · Click destinations — every tile lands somewhere you can act

Tile	Destination	What the user does there
Engaged	Inbox, filtered to recent inbound	Reads the threads and replies. Who is talking back is a conversation question, not a roster question.
Visited	Calendar, filtered to completed appointments for the period	Sees who came in, follows up on the ones who left without buying. (The Showroom tab has no route in and is today-scoped, so it is not the destination.)
Desked	Desk board, Working column	Works the deals where numbers exist — the closest thing to a live read of the desk.
Sold	The sold list	Reconciles against the DMS and works delivery.

Every destination inherits the same date range as the tile that opened it.

4 · Rules

These are period throughput counts — not live stage stocks, and not a funnel.

One deal that texted in, visited, was desked and bought inside the range appears in all four tiles.

The numbers therefore never divide into rates and must never be rendered with connectors or percentages between them.

There is deliberately no fifth Delivered tile: no trustworthy delivery signal exists (the delivery appointment type has 13 completions all-time at a busy store), and a tile that reads 1 because nobody logs it is a lie dressed as a metric.