Menu Engineering Breaks on a Food Truck. Here's the Fix.

Standard menu engineering asks which items are popular and which are profitable, then sorts them into four boxes — stars, plow-horses, puzzles, dogs. It's genuinely useful advice, and it was built for a dining room.

In a dining room the scarce thing is a seat. Every analysis quietly assumes that: your ceiling is how many people can sit down, so you want the highest margin per cover.

On a truck, nobody sits down. Your scarce resource is the service window, and there's exactly one of it. Your ceiling is how many orders you can push through that window during the ninety minutes that actually matter. Which means the number you need isn't margin per item, or margin per cover.

It's profit per minute of line time.

What that means

For every item on your menu, two numbers:

1. The profit it makes — price minus what the ingredients cost you.

2. The window time it consumes — how many seconds from the customer starting to order to that order leaving your hands.

Divide the first by the second. That's profit per minute of line time, and it will reorder your menu in ways the four-box grid never will.

The reason is straightforward. A high-margin item that takes four minutes to assemble is actively costing you money during a rush, because while you're building it, the six people behind that customer are waiting — and some of them leave. A modest-margin item you can hand over in forty seconds might be your best performer at peak and look mediocre on a spreadsheet.

Standard menu engineering would tell you to promote the first one. On a truck, at peak, that advice is backwards.

How to actually measure it

You need real numbers, and you can get them this week with a phone.

Step 1 — get your profit per item. Cost out your recipe: every ingredient, actual portions, including the sauce and the container. Subtract from your price. Most operators discover two or three items here they'd been mispricing for a year.

Step 2 — time your items. Pick your busiest hour. For each item, time from "customer begins ordering" to "food leaves your hands." Do at least five of each and take the median, not the average — one disaster order will skew a small sample.

Time the whole interaction, not just the cooking. Ordering, the modifier conversation, payment, assembly, handoff. All of it consumes window.

Step 3 — divide. Profit divided by minutes. Rank the menu.

Step 4 — look at the bottom five. For each, one of four things is true, and each has a different fix:

• Slow and low margin — Cut it

• Slow and high margin — Prep it further ahead, or make it off-peak only

• Fast and low margin — Raise the price — speed is why it survives

• Slow because of the modifiers, not the cooking — Simplify the options, not the recipe

That last row is worth dwelling on. A surprising amount of window time isn't cooking at all — it's the conversation. "What comes on that?" and a six-option build-your-own turn a forty-second transaction into two minutes. Reducing choices is often the single fastest thing you can do to your throughput, and it costs nothing.

What this reframe changes in practice

Once you're ranking by profit per minute, decisions that were arguments become obvious:

• Your peak menu should be shorter than your off-peak menu. Not a smaller version of the same thing — a deliberately different set, weighted to your fastest earners. Plenty of trucks already do this by instinct. This gives you the number to defend it.

• Combos earn their keep through speed, not margin. One decision instead of three is less window time, even at a discount.

• Prep moves become measurable. Portioning ahead of service converts window minutes into commissary minutes, which are cheaper. Now you can see exactly how much you bought.

• The "signature item" gets audited. Some signature items are worth their time because they're why people came. Some are just slow. The number tells you which, and it's often uncomfortable.

Where the honest caveat goes

Two things this doesn't capture, and I'd rather say so.

It ignores why people showed up. If your slowest item is the thing you're known for, cutting it can cost you the line itself. Profit per minute measures throughput, not demand. Use it to decide what to speed up, prep ahead, or price differently — not as the only input to a cut.

Off-peak, it barely matters. When there's no line, window time isn't scarce and standard margin thinking is fine. This frame is specifically a peak-hours tool. That's also when most of your money is made.

Where the manual version breaks down

You can do all of this with a stopwatch and a spreadsheet, and you should — go do it Saturday, whether or not you ever buy anything from us. The insight is worth more than any software.

The wall is that it's a snapshot. You timed things one Saturday. Then your menu changed, your prices moved, your ingredient costs went up, you hired someone new, and your rankings are quietly wrong again. Redoing it properly means re-timing everything and rebuilding the sheet, so nobody does it twice.

Half of that problem we solve: on Avocado your sales and margins live in one system, so "what did this item actually sell and at what margin, this month versus last" is a question you ask instead of a spreadsheet you rebuild. You can ask which items never sell, which have the lowest margin, and what sells together — from your own data, in plain language.

The timing half is still yours with a stopwatch. I'd rather tell you that than imply we do something we don't.

Name the metric anyway. Once your staff knows that profit per minute of line time is the thing you're optimizing, decisions at the window start making themselves.

Avocado is a POS built for independent restaurants and food trucks. We're operators first, and we'd rather tell you the honest version than the version that closes faster.