Order Count vs Like Rate: Which Metric Actually Matters

Two numbers get thrown around constantly in restaurant analytics, and they get treated as interchangeable far more often than they should be. Order count tells you how many times a dish sold. Like rate tells you what fraction of guests who ordered it actually enjoyed it. A dish can score high on one and low on the other, and mistaking one for the other leads to some genuinely backwards decisions about what to promote.

What each number is actually telling you

Order count is shaped by everything except pure guest preference: menu placement, price relative to neighbors, name recognition, how often a server mentions it. A dish can rack up a high order count purely because it sits at the top of a popular category, regardless of whether it's actually the best thing in that category. Like rate strips away all of that noise and asks a more direct question: of the people who tried this, how many were genuinely glad they did? That's a much closer proxy for the dish's actual quality than order count will ever be on its own.

Neither number alone tells the full story. A high order count with a low like rate means a lot of guests are trying a dish and coming away unimpressed, which is a recipe problem wearing a popularity disguise. A low order count with a high like rate means a genuinely great dish that isn't getting enough visibility, a placement problem rather than a quality one.

• High order count, high like rate: a genuine star, worth featuring prominently
• High order count, low like rate: a workhorse with a hidden quality problem worth investigating
• Low order count, high like rate: an underrated dish that needs better visibility, not a recipe change
• Low order count, low like rate: a clear candidate to cut or substantially rework

Why treating them as one number causes real mistakes

The most common mistake is using order count alone as a proxy for quality, since it's the number most POS systems surface easily while like rate has traditionally required a separate feedback mechanism most restaurants never built. That default toward order count means restaurants often keep promoting dishes that sell well for reasons unrelated to guest satisfaction, while genuinely excellent dishes stay buried because nobody had the second number to notice they deserved better.

The fix isn't picking one metric over the other. It's looking at both together, for every dish, before deciding what to promote, reprice, or cut. A promotion decision made on order count alone risks amplifying a dish guests don't actually love. A promotion decision made on like rate alone risks ignoring how much untapped demand might exist for a dish that's rarely been tried.

A worked example of the four combinations

Say a biryani sells forty orders a week with a like rate of 92 percent. That's the genuine star, worth featuring wherever you can. A pasta dish next to it also sells forty orders a week but with a like rate of 61 percent, a workhorse that's selling on habit or placement while quietly disappointing a third of the people who order it. A dessert sells only eight orders a week but earns a 95 percent like rate among those who tried it, a clear candidate for better placement or a mention from the server. And a side dish sells six orders a week at a 50 percent like rate, which is about as close to a clean cut candidate as the data ever gets. Four dishes, four completely different stories, and order count alone would have made the biryani and the pasta look identical.

Why this matters most at menu revision time

The gap between order count and like rate matters every week, but it matters most when you're deciding what survives a menu revision. Cutting purely on order count risks removing an underrated dish that just needed better visibility. Keeping purely on order count risks protecting a dish that's selling on inertia while actively disappointing a meaningful share of the people who try it. Looking at both numbers together before a revision is the difference between trimming the menu intelligently and trimming it by habit.

TasteGraph's dashboard tracks both numbers side by side for every dish, and its Promote category specifically flags the combination that matters most for growth: high like rate paired with low order count, the exact profile of a dish that deserves more visibility right now.