How Multi-Unit Restaurant Operators Should Read Consolidated P&Ls

Written by The Pragmatic CFO. 15+ years running restaurant P&Ls.

Last updated July 31, 2026.

Multi-unit consolidated P&Ls hide as much as they reveal. Read them in three passes: total company view, unit-by-unit view, and a variance-to-comp view. The consolidated number is the least interesting of the three. What matters is which units are subsidizing which, and why.

The three-view rule

Never look at the consolidated P&L in isolation. Always read it against two other views:

1. Unit-by-unit: same line items, each column is a unit, plus a corporate overhead column.
2. Variance-to-comp: each unit’s actual vs. its own budget, and vs. comparable units in the portfolio.

The consolidated view gives you the shareholder answer (are we making money?). The unit view gives you the operator answer (which locations are healthy?). The variance view gives you the manager answer (who is off pattern and why?).

What the consolidated P&L hides

Consolidated P&L averages disguise dispersion. A 6% blended 4-wall EBITDA might come from 8 units all running 5 to 7% (healthy portfolio) or from 4 units running 12% and 4 units running -1% (broken portfolio). Same headline number, completely different reality.

The math that matters: standard deviation of unit EBITDA. A portfolio with a wide spread has a manager problem (or a real estate problem, or a market problem) that consolidated numbers will not show for another two quarters.

Add a “portfolio dispersion” line to your monthly close package: mean 4-wall EBITDA and standard deviation across units. If SD is more than 40% of the mean, you have a dispersion problem that needs attention.

Corporate overhead allocation is a decision, not a report

How you allocate corporate G&A to units affects which units look profitable. Common allocation methods:

Method	How it works	Distortion
Pro-rata sales	Corporate G&A allocated by % of total sales	Penalizes high-sales units unfairly, ignores actual support consumed
Flat per-unit	Total G&A / unit count	Overloads small units, undercharges large units
Support hours consumed	Time tracking of area coach and support visits	Most accurate but requires tracking discipline
Blended (50/50 sales + flat)	Compromise of first two	Reasonable default for most 5-25 unit operators

Pick a method, document it, apply it consistently. Do not change methods mid-year to make a struggling unit look better. That is how boards lose trust.

The four lines to read first on any multi-unit P&L

1. Same-store sales growth

Comp sales isolate volume trend from unit count growth. New unit revenue can mask same-store declines for 12 to 18 months. If same-store is negative and total revenue is up, you are growing the wrong direction.

2. Prime cost by unit

The unit with the highest prime cost in the portfolio is usually the one that will close first. Not always, but the correlation is strong. See our post on prime cost.

3. Occupancy percentage by unit

Occupancy variance across units in the same market tells you which leases were signed at what point in the cycle. Units above 12% occupancy in a low-sales quarter need renegotiation or a closure decision.

4. New unit performance vs. underwriting

Every unit less than 24 months old should have a variance report to underwriting: sales vs. plan, prime cost vs. plan, EBITDA vs. plan. Below 85% of underwriting is a signal. Below 70% is an escalation.

The consolidation view that boards actually want

Line	Current period	Prior year	Var $	Var %
Company sales	$14.2M	$13.4M	$800K	+6.0%
Same-store sales	$12.6M	$12.9M	($300K)	-2.3%
New unit contribution	$1.6M	$500K	$1.1M	n/m
Prime cost %	61.4%	59.8%	160 bps unfav
4-wall EBITDA	$1.72M (12.1%)	$1.85M (13.8%)	($130K)	-7.0%
Corporate G&A	$820K	$740K	$80K	+10.8%
Company EBITDA	$900K	$1.11M	($210K)	-18.9%

The story that jumps out: 6% headline growth is really -2.3% comp masked by new unit contribution. Prime cost drift of 160 bps is eating margin. The board should be asking why comp is declining, not celebrating the top-line growth.

Same-store sales math done right

Only include units open for the full comparable period on both sides. A unit that opened in month 3 of prior year and is open all 12 months of current year should be excluded from comp. Rolling in partial-year units inflates comp growth by 1 to 3 points.

The other common error: comp against a period that includes closed units. If you closed 2 units in Q2 last year and they were losing money, your comp base is artificially low. Restate the prior year excluding closed units to get honest comp.

Weekly cash burn versus GAAP EBITDA

GAAP EBITDA does not equal cash flow. For multi-unit operators, the gap widens as unit count grows because of new unit capex, deferred maintenance, and working capital tied up in growth.

Report both weekly. Cash from operations divided by weeks in period should track within 5% of EBITDA divided by same weeks for a stable business. Bigger gaps signal that either the P&L is missing something or working capital is deteriorating. See why profitable restaurants still run out of money.

Unit closure decision framework

When a unit runs negative 4-wall EBITDA for 4+ consecutive quarters, the closure math needs to be run explicitly:

1. Compute the 12-month forward loss at current run rate
2. Compute the closure cost: lease buyout, severance, asset disposal, brand refunds
3. Compute the alternative: sublease value, sale to another operator, conversion to a different concept
4. If closure cost < 12-month loss and no viable alternative, close it

Most operators wait too long. The cost of holding a losing unit for one extra year is usually 2 to 4x the closure cost. See our related piece on why your restaurant is losing money.

FAQ

How often should I review unit-level P&Ls?

Monthly at minimum. Weekly for units under stress. Corporate should not close the books without seeing every unit variance report.

What is a healthy corporate G&A ratio?

5-10 units: 10 to 14% of sales. 10-25 units: 7 to 10%. 25+ units: 5 to 8%. Anything above these ranges is corporate bloat that eats into 4-wall margin.

Should I allocate corporate G&A to units on the internal P&L?

Yes, but keep an unallocated view too. The allocated view helps unit managers understand true cost. The unallocated view helps corporate see the raw 4-wall performance without allocation methodology arguments.

What is the right number of units for a portfolio’s optimal G&A use?

Efficiency curves out around 20 to 40 units for most concepts. Below 20, corporate is expensive per unit. Above 40, complexity starts to add cost back that you thought was already scaled.

How do I benchmark my unit dispersion?

Standard deviation of 4-wall EBITDA as a percentage of the mean. Under 25% is tight and healthy. 25 to 40% is normal. Above 40% signals real problems with operational consistency or real estate strategy.

For the full 2026 benchmarks, see our State of Restaurant Finance report.