The Items dashboard answers which items sell, which don't, and which ones moved recently. It is your menu's report card.

Open it when you are pricing, deciding what to cut, planning a menu refresh, or trying to understand whether a new item has legs.

Sum of order_items.quantity across confirmed items on closed orders for the selected range.

Number of distinct menuItemId values that appeared at least once. A higher number means more breadth on what is selling.

What share of your menu actually sold in the range. The denominator is your outlet's own menu — the base menu assigned to the outlet (with sensible fallbacks for outlets migrated from older setups). The result is capped at 100%, since the range can include items that sold and were later removed from the menu

Reading it: if this is below 60%, a big chunk of your menu isn't earning its keep. That is the kind of thing the Movers list and the bottom-20 table help you confirm.

Bars are each item's revenue, sorted descending. The line on top is the cumulative share of total sales as you walk left-to-right across the items.

Reading it: if the cumulative line hits 80% after only 5 items, you are extremely dependent on a tiny number of dishes; if it takes 40, the menu is broad and balanced. Either pattern has implications for prep planning.

The 20 best sellers by revenue in the range, with quantity sold and revenue.

The 20 worst sellers by revenue (with at least one sale). Candidates for cutting, repricing, or photographing better.

The biggest absolute movers in quantity between two windows: the last 4 weeks vs the 4 weeks before that. Items must have had at least 10 units in the prior window to qualify (so the % change is meaningful).

Columns: prior 4-week quantity, recent 4-week quantity, absolute change, percentage change. Sorted by absolute change so the rows at the top are the ones that moved the most in either direction.

How it's calculated: a rolling 8-week window ending today. Recent = last 4 weeks. Prior = the 4 weeks before that. The threshold of 10 in the prior window filters out items that swung from "tiny to small" — those would have huge % changes for the wrong reason.

This tab starts with a picker table at the top. Click any item row to filter every chart below to that single item; click again to clear. The screenshots below show the charts with no item selected, so they aggregate across the whole menu — once you click an item, each chart will show just that item.

For the filtered item: how many units sold on each channel and how much revenue each channel produced.

Reading it: useful for spotting a dish that's a dine-in star but a delivery dud (or vice versa). A heavy partner skew might suggest the dish travels well; a heavy dine-in skew means you might lose it if you push it on delivery.

A 24×7 heatmap showing the average quantity of this item sold per hour, by day-of-week.

How it's calculated: total quantity sold in each (hour, weekday) cell, divided by the number of distinct dates that contain that weekday in the selected range. So a value of "2.3" in the Tuesday 13:00 cell means an average of 2.3 units of that item between 1pm and 2pm on the Tuesdays in the range — not a single big day.

Reading it: identifies time-of-day patterns. A dish that's only ordered Friday and Saturday evenings might be worth featuring only on those dayparts; one that sells flat through the week is a daily staple.

The top 10 other items that appear in the same cheque as the filtered item, sorted by number of co-occurrences.

How it's calculated: we find every order that contained the filtered item, then count how often each other item appears in those same orders. A drink that shows up next to most main courses is your "natural pairing"; an unexpected pairing might be a marketing opportunity.

For the filtered item: the option groups attached to it (e.g. "Spice level", "Add-ons") and the most-chosen options inside each group.

Same picker at the top. The four charts below show how the filtered item's revenue and quantity have moved over time at four scales. The screenshots show the unfiltered (all-items) view.

Every menu item whose first-ever sale at this outlet was in the last 30 days. For each one we show total quantity to date and the quantity sold in each of the first four weeks since launch (W1, W2, W3, W4).

Reading it: the W1 → W4 trajectory tells you whether a new item has stickiness. A spike-and-fade pattern (big W1, falling after) suggests novelty-driven and worth a rethink; a steady or growing curve is a sign the item is finding its audience.

How it's calculated: for every distinct menuItemId, we take the minimum reportingDate. If that first date is within the last 30 days, the item qualifies. The W1–W4 buckets are calendar windows from that first-sale date forward.

Menu item groups are combos / set menus — one POS line item that contains several sub-items (a burger combo with a side and a drink, for example). The charts on this tab show combined quantities: the same item sold individually plus the same item sold as part of any combo. This is the most accurate count of "how many units of the item actually left the kitchen".

Total quantity per item, across the whole range, combining standalone and combo sales.

How it's calculated: for line items that are not combos (subOrderItems empty), we take the line's quantity. For combo line items (subOrderItems non-empty), we expand the JSON sub-items, multiply each sub-item's quantity by the parent line's quantity, and credit the sub-item's name. The two streams are summed.

The 40 most-chosen options across the whole menu. Each bar is one option (e.g. "No ice", "Extra cheese", "Spicy"), grouped by the option group it belongs to.

How it's calculated: we flatten the options JSON on each confirmed line item, sum the option quantities, and join back to the option groups (looking up both option_groups_v2 and the legacy option_groups table so historic data is included).

The same 40 most-chosen options ranked by the revenue they brought in (option price × quantity chosen).

Reading it: an option that's chosen often but adds no revenue is a "free modifier" — useful info but no price lever. An option with high revenue per choice is your upsell hero (extra cheese, premium upgrade).