How Restaurants Track Their Best-Selling Menu Items

How Restaurants Track Their Best-Selling Menu Items

Most Restaurants Are Guessing

Ask a restaurant owner which dish is their most profitable and most will answer confidently. But if you ask them to back it up with data, many can't. They know what sells — what the servers push, what guests ask about, what runs out first on a Friday. But "sells a lot" and "most profitable" are rarely the same thing.

This gap — between popular items and profitable items — is where restaurant margins get lost.

Tracking your best-selling menu items isn't just a data exercise. It's how you decide what to keep, what to promote, what to price differently, and what to quietly retire.


What You're Actually Trying to Track

There are four numbers worth knowing about every item on your menu:

Volume — how many times it's ordered in a period. Easy to pull from your POS.

Revenue contribution — total revenue that item generates. Volume times price.

Contribution margin — revenue minus direct food cost. This is the number that actually matters. A $40 steak with $28 food cost contributes $12. A $14 pasta with $3 food cost contributes $11. They're nearly equal in profitability despite a $26 price difference.

Interaction rate — on a digital menu, how often guests view an item versus how often they order it. A high view rate with a low order rate is a signal — usually pricing, description, or photo quality is the problem.


The Classic Method: POS Reports

Every POS system generates item-level sales reports. Most restaurant owners look at them occasionally but don't act on them systematically.

The useful data from your POS:

  • Items sorted by quantity sold in a date range
  • Revenue per item in a date range
  • Item performance by day of week or daypart (lunch vs dinner)
  • Modifier attach rates (how often guests add extras)

The limitation: POS data tells you what was ordered. It doesn't tell you what was looked at but not ordered, what guests asked staff about before deciding, or how menu placement affected the decision.


The Better Method: Digital Menu Analytics

A well-built digital menu adds a layer of data your POS will never have — guest behaviour before the order.

When guests browse a digital menu, every interaction creates a data point:

  • Which categories they opened
  • Which items they viewed and for how long
  • Which items they scrolled past without clicking
  • What questions they asked the AI assistant
  • What language they were browsing in

This is the difference between sales data (what they bought) and intent data (what they considered).

Example: Your salmon dish has the lowest order volume on the menu. POS data says pull it. But digital menu analytics show it's the third-most-viewed item, and guests frequently ask the AI "is the salmon sustainable?" — a question your current description doesn't answer. The fix isn't removing it. It's adding one line about sourcing. Order volume doubles.

That's a conclusion you could never reach from POS data alone.


Menu Engineering: Turning Data Into Decisions

Menu engineering is the practice of categorising every item based on two variables — popularity and profitability — and treating each category differently.

The four categories:

Stars — high popularity, high margin. These are your core items. Protect them, feature them prominently, use photos, and never discount them.

Plowhorses — high popularity, low margin. Guests love them but they're bleeding your margins. Options: raise the price slightly, reduce the portion, or pair them with a high-margin modifier (upgrade the side, add a sauce).

Puzzles — low popularity, high margin. These are your best opportunity. Nobody orders them but they'd make you the most money if they did. Fix: better description, better photo, better placement on the menu, or AI recommendation.

Dogs — low popularity, low margin. Remove them. Every item you keep has a maintenance cost — it has to be stocked, prepped, and listed. Dogs drain your kitchen without contributing.

Running this analysis quarterly is enough for most restaurants. Monthly if you change your menu frequently or have high food cost volatility.


How to Run This Analysis Without a Spreadsheet

Most independent restaurants don't have a data team. The practical approach:

Step 1: Pull your item-level sales report from your POS for the last 90 days. Sort by quantity sold.

Step 2: Add a column for food cost per item. Your kitchen staff know these numbers. If not, calculate: (cost of ingredients per serving / menu price) x 100 = food cost percentage. Aim for 28-35% for most categories.

Step 3: Calculate contribution margin: menu price minus food cost per item.

Step 4: Plot items on the four-quadrant matrix above. Anything above median volume is "high popularity." Anything above median contribution margin is "high margin."

Step 5: Act on each quadrant as described above.

If you're running MenuClips, the analytics dashboard surfaces item view counts alongside your menu data, so you can cross-reference what guests are browsing versus what they're ordering without building a separate spreadsheet.


Common Patterns to Watch For

High volume, low margin items that guests complain about when you raise prices.

These are your plowhorses. The right fix is rarely removing them — it's adding a profitable modifier (premium protein upgrade, premium side) and letting guests self-upgrade. Frame it as an enhancement, not an upsell.

Items that AI conversations mention frequently but that have low order rates.

This means guests are curious but not convinced. The description isn't doing its job. Rewrite it to address the question being asked. Add a photo if there isn't one.

Items that consistently 86 before end of service.

Either increase your prep volume or price them higher. If they sell out every night, guests are willing to pay more than you're charging.

Items that have never been ordered in 90 days.

Remove them without sentiment. Every item on the menu costs you in inventory, prep, and cognitive load for guests reading it.


The Frequency Question

How often should you do this analysis?

For most restaurants: once per quarter for full menu engineering, weekly for operational 86ing and special scheduling.

Seasonal menu changes should always be preceded by a performance review of the items being replaced — otherwise you risk removing a quietly profitable item because it didn't get enough marketing push.


Frequently Asked Questions

How do I find out which menu items sell the most?

Your POS system generates item-level sales reports sorted by quantity. If you use a digital menu platform, you also get view and interaction data that shows what guests looked at before ordering — which is often more useful than raw sales volume.

What is menu engineering?

Menu engineering is a methodology for optimising restaurant menus based on each item's popularity and profitability. Items are categorised as Stars (high on both), Plowhorses (popular but low margin), Puzzles (high margin but underordered), or Dogs (low on both). Each category requires a different management strategy.

How often should restaurants review their menu performance?

Most restaurants benefit from a full menu engineering review every quarter. Operational decisions — 86ing items, scheduling specials — should happen as frequently as daily based on inventory and prep data.

Can a digital menu help track best-selling items?

Yes. Digital menu platforms like MenuClips provide analytics on which items guests are viewing, interacting with, and how those interactions correlate with orders. This gives you data that your POS alone cannot — the gap between what guests consider and what they actually order.

What is a good food cost percentage?

Most restaurant categories target 28-35% food cost. Fine dining often runs 30-35% because of premium ingredients. Fast casual targets 25-30%. Bars typically run 18-24% on beverages. What matters most is the contribution margin per item, not just the food cost percentage.


*Related reading: Digital Menu Software for Restaurants — Buyer's Guide | The Best Digital Menu Software Compared*

Ready to Boost Your Restaurant Revenue?

Join hundreds of restaurants using MenuClips to serve smarter. Get started free today.

No credit card required · Free for 30 days · Cancel anytime