Blog

Margin in Plain Sight: What Prime Cost Doesn't Show Heading Into 2027

Written by Mary Hill Amason | November 20, 2026

Most margin conversations start and end with prime cost. At RFDC 2026 in Las Vegas, Craftable CEO David Cantu made the case that the next few points of margin are hiding somewhere else.

David opened Margins on the Table, a lunch for restaurant operators and finance leaders at Fogo de Chão before the conference. His talk, "Margin in Plain Sight: Keep More of What's Yours," walked through the cost pressures facing restaurants in 2027, the bills that rarely get a line in the weekly review, and the KPIs that bring them into view.

Here are the main takeaways. You can also download the full deck.

Four forces shaping 2027, from most to least certain

David ranked the pressures on 2027 costs by how sure we can be about them:

  1. Weather (high confidence): NOAA puts the odds of a historic El Niño at 83%.
  2. Fuel (known today): diesel is at $6.35 a gallon, up 74%. Every delivery carries that cost.
  3. Food costs (shifting): 78% of operators expect food costs to rise from tariffs and other pressures.
  4. Traffic and rates (least certain): same-store traffic is down 2.0%, and interest rates remain a risk.

On commodities, the picture is mixed. USDA expects beef to rise more slowly (4.7% at retail, after 8.7% in 2026), while pork, chicken and dairy look steady. Produce is the wildcard: fresh vegetables rose 5.8% in 2026 and the 2027 range is wide. USDA projects menu prices up 2.8% in 2027.

David's question for the room: are these in your P&L, and does your plan assume relief that may not come?

The quiet bills outrunning menu prices

Food and labor get the attention, but the bills nobody reviews line by line are rising fastest. Every one of these is climbing faster than menu prices (+2.8%):

Cost Change year over year
Foil, film and wrap +11.8%
Trash liners +10.7%
Water and sewer +8.2%
Waste removal +6.6%
Electricity +5.9%
Pest control +5.5%
Linen and laundry +5.0%
Hourly pay (BLS) +4.9%
Menu prices +2.8%

These "other operating costs" are only about 3% of sales, so they're easy to wave off. But a 6% rise costs about 0.2 points of margin, and roughly 1 in 5 controllable dollars sits outside prime cost. At Chipotle it's 16.0 of 71.2 points of controllable cost (Q3 2026); at Texas Roadhouse, 14.2 of 82.1 (Q2 2026).

Five numbers you may not be seeing

The core of the talk was five numbers that rarely show up in a P&L review but quietly move margin:

  1. Forecast error. Only 35% of operators forecast sales within 5%. Top-tier operators hit that 68% of the time; bottom-tier operators, 7%. Only 14% of brands with 100+ locations made the top tier.
  2. Quiet bills. Other operating costs run about 3% of sales and are rising roughly twice as fast as menu prices.
  3. Recipe costs. In Craftable's data, 2.4% of recipe lines are off by 3x or more, so theoretical food cost is wrong before the week starts.
  4. Over-buying. Across 22 restaurants, an average of 6¢ of every perishable dollar is bought beyond need.
  5. Traffic. Sales rose 1.8% while traffic fell 2.0%. Price increases can hide a shrinking guest count.

David also pointed to the opposite of waste: running short. 86'd items, emergency grocery runs (a median of 10 a year, about $1,167 in logged runs alone) and hot-shot deliveries (1 in 13 arrives off schedule) all cost money. Waste and shortages share a root cause: orders that aren't built from the forecast.

That's why the forecast's inputs matter. A useful forecast combines sales history by channel, real-time third-party and catering orders, weather and local events, and purchasing data from invoices. It should produce one forecast by location, channel and menu item, with a range, and drive ordering and prep. David's advice to operators evaluating any vendor: ask what goes into the forecast, and what happens to your pricing data.

New KPIs, from the store to the board

David closed with a set of suggested KPIs (not industry benchmarks) that put these hidden costs in front of the people who can act on them:

  • Store, daily: forecast error by daypart; actual vs. theoretical for the top 5 food and liquor, beer and wine items sold; sales and guests per labor hour (SPLH/GPLH).
  • Region, weekly: forecast bias by store and daypart; 86s, store runs and hot shots; price-adjusted traffic; SPLH/GPLH.
  • Board, monthly: year-over-year store-level EBITDA; forecast accuracy by store; food waste as a % of purchases; GM turnover.

His final question was simple: can you report these today? If not, that's where to start. What gets measured gets managed.

Get the full deck

The full presentation includes the data sources behind every number above. Download "Margin in Plain Sight" (PDF).

Want to see how forecast-driven ordering and prep work in your own stores? Book a demo with the Craftable team.

Sources cited in the talk: NOAA, AAA, EIA, Restaurant365, Black Box Intelligence, USDA Economic Research Service, Bureau of Labor Statistics, Fourth and QSR Magazine (2026), and Craftable internal data.