Cover and turn analysis for Indian restaurants — measuring table efficiency

About this piece. Most Indian dine-in operators talk about "footfall" and "ABV" but never put a number on the one metric that ties seat capacity to revenue: turns per table per day-part. This piece walks through the cover and turn maths, day-part benchmarks for Indian formats, and the four staffing and menu calls that the numbers should drive.

Definitions, in 60 seconds

Two numbers, often confused:

• Cover — one diner served. A four-top with three guests = three covers, not one.
• Turn — one full seating cycle of a table. A four-top that seats two parties in a service = two turns.

Together they answer the only question that matters for a dine-in P&L during peak hour: did we move enough people through the seats we paid rent for?

The two formulas you actually use

Covers per day-part = Σ (party size) over the day-part
Turns per table     = Covers per day-part ÷ (seats × occupancy factor)
Turn time (avg)     = Day-part minutes ÷ Turns per table

The occupancy factor is the share of seats actually usable in a service — almost never 100%, because two-tops next to a service station get blocked, four-tops with a party of two leave two seats empty, and a row of joined tables for a group blocks turns elsewhere. A reasonable default is 0.85 for casual dining, 0.75 for fine-dine, 0.92 for QSR with shared seating.

Benchmark turns by format (India, weeknight dinner service)

The numbers below are weeknight dinner ranges from operator interviews, NRAI India Food Services Report category notes, and Crisil F&B SME bulletins. Treat them as the band you should land in, not as targets.

The mistake we see most often is comparing turns across formats. A 1.4 turn in fine-dine is healthy. A 1.4 turn in QSR is a problem. Pick the right band, then chase the band.

A worked weekly example

A 60-seater casual-dine in a tier-1 city, dinner service 7pm–10pm (180 minutes):

Mon: 78 covers   Tue: 84   Wed: 92
Thu: 110         Fri: 168  Sat: 196  Sun: 152

Apply the formula at 0.85 occupancy:

Two readings jump out:

1. Mon–Tue runs below 1.6 turns. Either trim FOH headcount (one floor + one runner, not two of each) or run a Mon–Tue value menu to lift covers without discounting the weekend.
2. Fri–Sat is past 3 turns in a 180-minute window. That's a 47–60-minute table cycle in a format meant to seat for 90+. Service quality is being consumed for revenue. Either shorten menu items that take 25+ min to plate, add a hostess to manage waitlist, or accept the cap and stop overbooking.

The four decisions cover-and-turn data should drive

If the data isn't driving any of these four decisions, you don't need it yet — you need a simpler tally.

1. Staffing roster by day-part. Mon-Tue trims, Fri-Sat adds. The roster should match the turn band, not "one team for every shift".
2. Menu engineering for speed. A high-turn day-part needs a menu where 70% of orders plate in under 12 minutes. The Stars-Plowhorses-Puzzles-Dogs matrix for menu engineering overlays neatly on top of turn data.
3. Reservation and waitlist policy. Once a service hits 2.4+ turns in a CDR, a no-walk-in policy past 8pm starts paying for itself in cover quality.
4. Capex on seats vs throughput. Adding seats only helps if you're hitting the top of the turn band on weekends. Adding seats when Mon-Wed runs at 1.5 turns just spreads the same demand thinner.

A weekly cover-and-turn template (free)

The template captures, per day-part:

• Open seats × occupancy factor
• Covers (manual tally or POS export)
• Turns/table (auto-calculated)
• Avg turn time (auto-calculated)
• Service-quality flag (auto: green if in band, amber if approaching cap, red if over)

A printable PDF of the same grid sits inside the restaurant till close checklist so the closing manager fills it in once per shift, not as a separate ritual.

Common measurement mistakes

• Counting reservations, not arrivals. A 6-cover booking that walks in as 4 is 4 covers, not 6. Update the count at seating.
• Ignoring no-shows in occupancy. A 30-minute no-show on a Friday at 8pm is a missed turn. Tighten the hold-time policy before adding seats.
• Mixing dine-in and takeaway covers. Takeaway has zero seat cost — track it on a separate line. Mixing them inflates the turn number and hides a real seat-utilisation problem.
• Using monthly averages. Turns are a peak-hour metric. A monthly average masks the Mon-Tue underuse and the Fri-Sat overload that need different fixes.

What to do this week

If covers and turns aren't tracked yet, start with three days of manual tally — Wed, Fri, Sat — at 30-minute intervals through dinner service. Plot them against the band table above. By the second weekend you'll see whether the binding constraint is demand (raise covers via menu / promo / hours) or seats (manage turns via menu speed and waitlist). The fix follows the constraint.

Related on Restaurant Daily

• Menu engineering matrix for Indian restaurants
• Restaurant labour cost percentage — Indian benchmarks
• Restaurant unit economics in India — full guide