The Cost Index · week of 2026-06-05 · 5 min read · By Don Goldstein

Where the basket stands this week. What's flashing.

The restaurant cost index for the week of 2026-06-05: the weighted basket reads +9.1% against its baseline, 9 of 13 tracked ingredients above their own window. These are public wholesale levels, never your delivered price — a read on the market, so you can tell a real move from a vendor markup.

Here is the read I run on a Tuesday between the produce drop and the pre-shift, and it is the same read this dispatch carries. The cost index for the week of 2026-06-05 has the weighted basket sitting at +9.1% against its baseline, at high confidence across 16 contributing ingredients. You already watch your own invoices — this is the wholesale market underneath them, so a delivered-price jump can be checked against whether the market actually moved or your vendor did.

One honesty line before the numbers, because it changes how you read every one of them. Each ingredient's percentage here is its read against its own tracked baseline window — a state-of-play "what's flashing this week," never "moved +9.1% since last week." The panel does not archive weekly snapshots yet, so I will not pretend it measures a week-over-week delta it cannot see. And every figure is a public wholesale level, never your delivered price: this is a read on the cost index, not a line for your food cost sheet. The point is direction and gap, not a number to paste into a cost sheet.

Above baseline3 below · 1 flat

Weighted baskethigh confidence · 16 ingredients

Where the panel sits this week: the spread of reads above baseline, and the weighted basket's own reading. Both are reads versus each baseline window, not a week-over-week move.

What's flashing this week

The panel sorts into a short action list: 1 re-price signal, 1 on watch. A re-price flag means the move looks structural — elevated and sustained against the baseline. A watch flag means a real move that has not persisted long enough to act on yet. Neither is advice; both are calibrated, low-regret reads off the measured index.

Re-price — Romaine lettuce. It reads +168.9% against its baseline; elevated and sustained — the increase looks real. This one is seasonal, so it typically eases when the season turns.
Watch — Pork loin. It reads +0.5% against its baseline; a real move that has not persisted yet — watch the next read.

If nothing here matches a line on your own menu, that is fine — only act where the flashing item is something you actually buy. The whole panel is filtered to the index's shippable set, so every name above is an ingredient the hub can show a live reading for.

The widest gaps from baseline

Beyond the action flags, here is the full spread of movement. Reading above baseline this week: Romaine lettuce +168.9% · Ribeye +30.3% · Salmon fillet +17.6% · Onions +16.7%. Reading below: Whole chicken −31.8% · Butter (AA, bulk) −11.0% · Tomatoes (round) −8.9%. The bars below scale to the largest mover so the gaps are legible — rust where cost is building, teal where it is easing.

Widest gaps from baseline this week (bars scaled to the largest mover; rust is building cost, teal is easing)

Romaine lettuce

+168.9%

Whole chicken

−31.8%

Ribeye

+30.3%

Salmon fillet

+17.6%

Onions

+16.7%

Butter (AA, bulk)

−11.0%

Each bar is a read versus that ingredient's own tracked baseline window — a state-of-play snapshot of what's flashing, not a move since last week.

The widest gaps from each ingredient's tracked baseline this week. Rust bars are building cost; teal bars are easing.

Read these as gaps, not verdicts. A wide rust bar on a seasonal item often unwinds when the season turns; a wide teal bar can be a vendor clearing inventory rather than a durable easing. The bar tells you where to look; your delivered invoice tells you whether it reached your back door.

The feed context behind the proteins

Proteins do not move on their own — the feed market underneath them sets a floor that flows through on a lag. This week, Corn (feed) reads −10.5% against its baseline, and Soybeans (feed) reads −7.2% against its baseline. A feed read that is easing is the context behind a protein reading softer than its baseline; a feed read that is building is the early-warning on one heading the other way.

1

Corn (feed) reads −10.5%

A tracked feed input, read against its own baseline window this week.
2

It sits behind 7 proteins on the panel

Including Chicken breast (boneless), Whole chicken, Pork loin, Pork shoulder — the items whose cost the feed market helps set.
3

The read flows through on a lag

A easing feed market is the context behind those proteins' own reads — directional, not a forecast.

The feed-to-protein chain behind this week's reads. Directional context from the measured index, never a forecast.

The feed-to-protein chain is directional context, drawn from the same measured index — never a forecast and never a delivered price. It tells you which way the wind is blowing on the proteins you carry, so a quote that moves the other way is worth a question to your vendor.

How to read this, and what it is not

Three rules keep this honest. First, every number is a public wholesale level, never your delivered price — freight, contract, and pack size all sit between this panel and your invoice. Second, each percentage is a read versus that ingredient's own tracked baseline window, a state-of-play snapshot of what is flashing, not a week-over-week move. Third, the panel is drawn from public USDA, BLS, and FRED data; when an input cannot earn a credible reading, it stays off the page rather than showing you a guess. Watch your own delivered invoices against these reads — the gap between the two is where a vendor conversation lives, and it is the first place a moving prime cost shows up.