Most restaurant BI dashboards have 30+ widgets and get glanced at twice a month. Which is worse than having no dashboard at all, because it creates the illusion of data-driven management while no one is actually reading the data. Here are the only seven numbers you need to review weekly.
1. Revenue (this week vs. same week last year)
The headline number, but only useful in comparison. Comparing this Tuesday to last Tuesday is noise — weather, events, and one-off factors dominate week-to-week. Comparing this week to the same week last year is signal.
Benchmark: a healthy independent grows revenue 5–15% year-over-year in its first three years. Flat or declining = something is wrong. 20%+ growth = you're scaling; start hiring ahead of demand.
2. Cover count (same framing)
Covers = the number of guests you served. Why separate from revenue? Because revenue growth driven by check size is different than revenue growth driven by more guests. Both are good; they mean different things.
- Revenue up, covers up: you're winning on volume. Expand hours, add capacity.
- Revenue up, covers flat: you're winning on check size. Menu pricing, upsells, kiosk, and loyalty are working.
- Revenue flat, covers up: guests are downtrading. Check menu psychology.
- Revenue down, covers down: demand problem. Marketing and guest experience.
3. Average check
Revenue / covers. Track weekly. The interesting signal is direction of travel:
- Trending up: menu, upsells, kiosk, loyalty are pulling upward.
- Trending down: guests are ordering more conservatively. Could be macro (recession), could be internal (team not upselling, menu feels dated).
4. Food cost % (COGS / revenue)
The classic. Target varies by concept:
- Pizza: 28–32%
- Fast-casual bowls/salads: 28–33%
- Burger/sandwich: 30–35%
- Full-service bistro: 30–35%
- Steakhouse: 35–40%
Weekly food cost % should move <1% up or down in a stable week. A 2%+ jump means: pricing changed on a key ingredient, waste increased, theft, or a bad count. Investigate.
5. Labor % (labor / revenue)
Targets:
- QSR/fast-casual: 22–28%
- Full-service casual: 28–33%
- Fine dining: 33–38%
This is the number that surfaces overstaffing fastest. Tuesday slow and labor % hits 38%? Cut shifts. If labor % creeps up over 3 weeks with flat revenue, you're overstaffed. If it creeps down with rising revenue, you may be understaffed and losing hospitality quality.
6. Top 5 selling items (and bottom 5)
Know your heroes and zeros. Every week.
Top 5
Your best-sellers are your menu's center of gravity. They tell you what guests are actually ordering (often different from what you think). Protect these items — consistent prep, never 86'd, featured on menu boards and online.
Bottom 5
Items that sell 5 or fewer units per week are candidates for removal. Menu bloat hurts: it slows prep, complicates inventory, confuses guests. A weekly review keeps the menu tight.
7. Repeat visit rate (unique guests who returned in the last 30 days)
The hardest of the seven, because you need a real guest database to measure it. But it's the single most predictive metric for long-term health.
Benchmark: a healthy independent restaurant has 35–50% of guests who visit at least twice in a 30-day window. Below 25% means you have a demand-gen problem (great first experience, no hook to return). 50%+ means you have loyal regulars and your word-of-mouth engine is working.
The weekly review
Pick a morning. 30 minutes, same day each week. Look at each of the seven numbers vs. last week and vs. same-week-last-year. Write three sentences:
- What went right?
- What went wrong?
- What are we changing this week?
That's it. Done religiously for a year, this habit beats any BI tool you can buy.
What to ignore (at this level)
- Social media follower counts (vanity metric).
- Google review rating (move it, don't track it weekly).
- Hourly breakdowns (too granular for weekly review).
- Third-party delivery channel breakdown (unless it's >15% of revenue).
Where these numbers live
In a stitched stack, these seven numbers live across 3–4 systems and never surface together. In a consolidated platform, they're on one dashboard — which is the difference between "I'll get to my numbers when I have time" and "I look at my numbers every Monday because they're right there."
Bottom line
You don't need more data. You need the same seven numbers, reviewed weekly, with honest comparisons. Restaurants that run this habit for a year outperform restaurants with fancy BI tools that nobody reads. Do the boring thing.