How to Control Alcohol Inventory in a Restaurant (2026)

May 2, 2026 · Brian Clow

Updated April 2026 · Affiliate disclosure: we may earn a commission if you purchase through our links. This does not affect our rankings.

Quick Answer: Why Alcohol Inventory Control Matters

Alcohol is the highest-margin category in most full-service restaurants and bars. Spirits typically carry a gross margin of 70 to 80 percent. Draft beer runs 75 to 80 percent. Wine by the glass runs 60 to 70 percent. These margins are what make a bar program financially valuable to the overall operation.

They also make alcohol the category where uncontrolled loss is most expensive.

A restaurant losing five percent of alcohol cost through over-pouring, untracked comps, and shrinkage on $400,000 in annual beverage sales is losing $20,000 per year — but because of the margin structure, that $20,000 represents $60,000 to $80,000 in retail revenue that should have been generated and was not. Every ounce of alcohol lost or given away without a sale carries the full retail value of the lost margin, not just the cost of goods.

If you want to understand how losses like this impact your overall profitability, see How Much Are High Food Costs Really Costing Your Restaurant.

The compounding problem: most alcohol loss does not appear on any standard report by name. It shows up as an unexplained gap between what should have been sold based on the pour model and what was actually counted. Without formal tracking, that gap is attributed to spillage, breakage, or measurement error — explanations that are sometimes true and sometimes a convenient substitute for a real investigation.

For a full comparison of inventory and food cost tools, see Best Restaurant Inventory Management Software.

How Alcohol Inventory Works in Real Restaurants

Alcohol inventory is more complex than food inventory because three different product categories — spirits, beer, and wine — each have different units, different tracking methods, and different loss mechanics.

Spirits:

Measured by the bottle and by the pour, with partial bottles requiring a consistent method for estimating remaining volume
Standard pour sizes vary by drink type — typically 1 to 1.5 ounces for spirits, 4 to 6 ounces for wine, 12 to 16 ounces for draft beer — and each is the basis for theoretical usage calculations
Price per ounce varies dramatically by product, which means a consistent two-ounce over-pour on a premium whiskey is a much larger financial event than the same over-pour on a well spirit

Draft beer:

Tracked by keg, with usage calculated by volume poured versus volume in the keg
Foam, line waste, and temperature-related issues create legitimate loss that needs to be separated from actual over-pouring or theft
Keg weight scales or flow meters give more accurate tracking than visual estimation of remaining volume

Wine:

Sold by the glass and by the bottle, which creates two separate tracking needs
Bottle counts are relatively straightforward — by-the-glass tracking requires connecting pour volume to POS sales to identify variance
Wine by the glass is often where the first bottle of a new pour gets opened casually before service, creating a partial that is not logged

What makes variance harder to detect than food:

A missing case of produce is a visible, physical absence
An over-pour is a fraction of a second longer on a trigger or free-pour — it produces no physical evidence and no record
The only way to detect it consistently is to compare theoretical pour volume against actual physical counts, which requires both a working recipe pour model and a consistent counting process

Where Alcohol Loss Actually Happens

Over-Pouring

Over-pouring is the single largest source of measurable alcohol loss in most bar operations, and it is usually not intentional.

Most bartenders free-pour. A bartender who has poured thousands of drinks has a highly calibrated sense of what one ounce feels like — but that calibration drifts. A study published in the journal Substance Use and Misuse found that even experienced bartenders over-poured by an average of 26 percent compared to standard measures when free-pouring without measuring tools.

What that means in practice:

A standard cocktail built on 1.5 ounces routinely being poured at 1.75 to 2 ounces represents 17 to 33 percent more product per drink
On a bar doing 200 spirit pours per night at an average cost of $1.50 per ounce, a 0.25 ounce over-pour per drink is $75 per night in product cost overrun — $27,375 per year from over-pouring alone
The bartender doing this is not stealing. They are simply pouring the way they always have, and no one has measured it

Untracked Comps

A comp that is rung into the POS is a traceable cost. A comp that is given without being rung in is simultaneously a product cost and a revenue loss.

Most restaurants have a comp policy. Most also have a significant gap between that policy and what actually happens during a busy service
A drink given to a regular at the bar that is never entered into the POS represents the full retail value of that drink — the product is consumed, no revenue is generated, and the only trace it leaves is an unexplained variance in the beverage cost report
Servers and bartenders who comp items informally to drive tips or maintain guest relationships are making a business decision with someone else’s money and no record of it

Staff Giveaways

Staff giveaways are related to comps but distinct — they are product consumed by or given to employees rather than guests.

Shift drinks that exceed what the policy permits, drinks poured for off-duty staff who come in for a visit, or product consumed during prep or service without logging all fall into this category
The per-instance cost is small. The aggregate annual cost can be significant in a high-volume operation with a large staff
This category is the hardest to address because it requires a direct conversation about behavior, but it is addressable through consistent counting and a clear policy enforced by management

Internal Consumption and Undocumented Drinking

This is one of the most common sources of alcohol loss in restaurants and one of the least discussed. It is not primarily a theft problem — it is a culture and accountability problem, and the distinction matters for how you address it.

In most full-service restaurants and bars, some level of informal staff drinking is normalized. A bartender tastes a cocktail before service. A server finishes a glass that came back to the bar. A closing crew shares a bottle at the end of a late shift. None of these events feel significant in the moment. Most of them are not done with any intent to steal. But they share a common feature: none of them get recorded anywhere, and the product cost shows up as unexplained beverage variance at the end of the period.

How it typically happens:

Bartenders and kitchen staff consume product informally during or after service without it being rung in
Shift drinks are poured generously, without a consistent standard for what the policy actually allows
Tastings that are legitimate — learning a new spirit, evaluating a batch cocktail — happen without documentation and blur into informal consumption
Closing staff treat leftover open bottles as accessible in ways that are not formally authorized
The behavior is often inconsistent — enforced by some managers and ignored by others — which creates ambiguity about what the actual policy is

Why it is hard to detect:

No single event is large enough to trigger an alert
The behavior is intermittent and informal, which makes it invisible in any single shift report
It often happens during the periods of lowest oversight — late service, closing shifts, or transitions between management
Staff who are otherwise performing well and genuinely valued are often the ones involved, which makes the conversation feel disproportionate to the behavior

How it shows up in the data:

Persistent elevated beverage cost variance that does not improve after portioning and comp controls have been tightened
Variance patterns that correlate with specific shifts, closing teams, or high-trust staff members rather than distributing evenly across all periods
Beverage theoretical versus actual gaps that are explained partially but not fully by documented loss — the numbers are closer but never quite right
A general sense among management that the bar is “running a little heavy” without a specific attributable cause

Practical ways to reduce it:

Write down what the shift drink policy actually is and communicate it clearly — ambiguity creates the conditions for informal interpretation
Require all staff drinks, including shift drinks and tastings, to be rung into the POS as comps or staff consumption — the act of ringing it in changes the behavior because it makes it visible
Separate legitimate product education and tasting from informal consumption — if bartenders are learning new spirits, structure it with a manager present and log it
Review beverage variance by shift rather than only by period — if one closing team consistently produces more variance than others, the data will show it before you need to guess
Avoid approaching this as a disciplinary issue unless the evidence points clearly in that direction — most operators get better results by tightening the system and making expectations explicit than by assuming bad intent and creating a confrontational environment

The goal is not to eliminate the hospitality culture that makes restaurants good places to work. It is to make sure that culture operates within a framework where the costs are visible and the policy is clear. Most staff will comply with a consistent, communicated standard. What they will not self-regulate is an environment where the standard is unclear and enforcement depends on which manager happens to be closing.

Bottle Shrinkage

Bottle shrinkage includes any product that leaves inventory without a corresponding sale or logged comp. This includes:

Bottles broken during service or stocking that are not logged
Product consumed in cocktail prep and tasting that is not accounted for in recipe costs
Bottles removed from inventory entirely — which requires either a recording failure or something more concerning
Low-level theft that accumulates slowly enough that no single instance is obvious

Shrinkage is distinct from over-pouring in that it typically involves whole units or significant volumes. It is caught through formal inventory counts compared against purchasing records, not through pour analysis.

Counting Errors

Counting errors are not loss — they are measurement failures that make loss invisible or make non-existent loss appear real.

A partial bottle estimated at 60 percent full when it is actually 40 percent full overstates ending inventory and understates cost of goods, making beverage cost look better than it is
Inconsistent methodology — one bartender counting partials by weight, another by visual estimate — produces counts that are not comparable across periods
Items counted in the wrong location, double-counted, or missed entirely shift the inventory value and distort every variance calculation built on it

Consistent methodology is not about precision for its own sake — it is about producing counts that can be compared reliably from week to week, which is the only way to make variance data meaningful.

The Math (Why Alcohol Loss Is So Expensive)

The financial impact of alcohol loss is disproportionate to its volume because of the margin structure of beverage sales.

Base example:

Annual alcohol sales: $350,000
Beverage cost percentage at target: 22 percent
Expected cost of goods: $77,000
Actual cost of goods at 27 percent: $94,500
Cost of the five-point variance: $17,500 per year

The revenue impact:

Spirits sold at $12 retail with a $2.40 cost (20 percent beverage cost) produce $9.60 of gross margin per drink
A drink given away at no charge costs $12 in lost revenue plus $2.40 in product — $14.40 total impact per untracked comp
Five untracked comps per shift across a Friday and Saturday is 520 drinks per year at $14.40 each: $7,488 per year from that source alone

Why this matters more than food waste of the same dollar amount:

A restaurant losing $17,500 per year in food waste is losing product that carries a 30 to 35 percent cost — the retail value of that waste is $50,000 to $58,000
A restaurant losing $17,500 per year in alcohol cost at a 22 percent beverage cost is losing product with a retail value of nearly $80,000
The same dollar loss in alcohol represents significantly more foregone revenue because the margins are higher

Alcohol loss is often just one part of a broader waste problem across the operation. Use our Food Cost & Profit Calculator to see what a few percentage points of alcohol is actually costing your restaurant.

Why Most Restaurants Miss Alcohol Loss

The reasons alcohol loss goes undetected are structural, not personal. Most operators are not ignoring it deliberately — they simply do not have a system that makes it visible.

No consistent counting:

Bar inventory in most restaurants happens less frequently than food inventory, if it happens at all on a formal basis
When it does happen, the methodology is often inconsistent — different people, different methods, different assumptions about how to measure partials
Inconsistent counts produce results that cannot be meaningfully compared across periods

No standard pour tracking:

Most bars do not track theoretical pour volume — the number of drinks sold multiplied by the recipe pour size — and compare it against actual product used
Without this comparison, over-pouring is invisible regardless of how significant it is
The POS tracks revenue. It does not track whether the product that generated that revenue was poured to recipe spec

No separation of beverage cost:

Operators who calculate a combined food and beverage cost percentage and manage to that single number cannot isolate alcohol variance
A strong food cost week can mask a bad beverage week in a combined calculation
Alcohol loss that is absorbed into a combined percentage stays invisible until it is large enough to move the combined number by a full point or more

Reliance on feel:

“The bar feels like it’s doing well” is not a measurement
Busy nights, high bar tabs, and strong tip percentages create a sense of performance that does not correspond to beverage cost percentage
Revenue and margins are not the same thing, and high-volume nights can coexist with high beverage cost if over-pouring and untracked comps are running in proportion to volume

This same lack of visibility is what drives inventory-related losses across the business. See How Much Is Poor Restaurant Inventory Management Costing You for a deeper look.

How to Track Alcohol Inventory Properly

Tracking alcohol inventory accurately requires method, consistency, and a comparison against a theoretical model.

Counting bottles:

Count every spirit bottle in every storage location — back bar, speed rail, back stock, service wells
Count full bottles as full units — one 750ml, one liter, etc.
Measure partials using a consistent method — a scale (grams converted to ounces) is the most accurate; visual tenths are acceptable if applied consistently by the same person every count
Count wine by the bottle for unopened stock and by pour count for partially opened bottles connected to by-the-glass tracking
Count draft beer by remaining keg weight or volume

Measuring partials:

Partial bottles are where the most counting error occurs
A kitchen scale that measures bottle weight (minus the weight of the empty bottle, which should be documented for each bottle type) converts directly to remaining ounces
Tenths estimates — 0.1 through 0.9 of a full bottle — are a faster alternative that sacrifices some precision for speed
Whatever method is used, it must be used the same way every count by the same person or team

Setting par levels:

Par levels for alcohol should be set based on actual usage data from the previous two to four weeks, not on gut feel or historical ordering habits
A bar that routinely over-orders creates excess stock that becomes a tracking and accountability problem — more bottles in more locations means more opportunities for variance to hide

Tracking by category:

Spirits, beer, and wine should be counted and tracked as separate categories
Within spirits, tracking by well, call, and premium separately gives you a more precise picture of where variance is occurring
Draft and packaged beer are counted separately because their tracking methodology and loss drivers differ

Weekly vs Monthly Alcohol Tracking

Monthly alcohol tracking is the standard in most restaurants. It is not adequate for managing beverage cost.

Why monthly is too slow:

A bartender over-pouring at a consistent rate does so across every service in the period
If counted monthly, that pattern runs for 30 days before it is detected
At $75 per day in over-pour cost on a busy bar, a month-long detection lag costs $2,250 before anyone looks at a number

What weekly tracking gives you:

A corrective window — if beverage cost is running high in week two, you have two weeks left in the period to investigate and respond
Pattern identification — consistent variance in the same week that happens to fall on the same days each month points toward a specific shift or bartender
Closer alignment between what was counted and what was actually used during a shorter, more manageable window — partials are easier to estimate accurately over seven days than over thirty

Recommended counting schedule:

Full bar count every week, on the same day and time
Same counter or counting team each week for consistency
Theoretical versus actual comparison run within 24 hours of the count while the period is still recent enough to investigate

This is also why many operators move beyond manual tracking and use systems that monitor cost during the period rather than after it closes. See Do You Really Need Multiple Systems to Manage Restaurant Ordering, Inventory and Food Cost.

What a Healthy Beverage Cost Looks Like

Beverage cost percentage benchmarks vary by concept type, product mix, and market positioning. These ranges reflect commonly cited industry norms.

Spirits and cocktails:

Well-managed bar programs typically target 18 to 22 percent beverage cost on spirits
Craft cocktail programs with premium ingredients may run 22 to 26 percent as an intentional choice related to ingredient quality
Above 26 percent on spirits consistently is a signal worth investigating

Draft beer:

Draft beer cost typically runs 20 to 25 percent of draft beer revenue when properly managed
Foam, line waste, and temperature losses are legitimate — typically accounting for two to four percent of keg volume — and should be factored into the target
Packaged beer (bottles and cans) typically runs 22 to 28 percent depending on product mix and retail pricing

Wine by the glass:

Well-managed by-the-glass wine programs typically run 25 to 35 percent
Wine is more difficult to control than spirits because partial bottles are harder to track and portion sizes vary more in practice than in theory

Wine by the bottle:

Bottle wine typically runs 28 to 38 percent depending on the wine list and pricing strategy

Overall blended beverage cost (all categories combined):

Full-service restaurants with a bar program typically target 18 to 24 percent blended beverage cost
Bars and nightclubs with a high spirits mix and simple draft beer program often target 18 to 22 percent
Concepts with a significant wine focus may run 22 to 28 percent blended

What these ranges assume: consistent weekly counting, a functioning pour standard, and a POS that accurately captures sales by category.

How Much Alcohol Loss Is Typical (And What Is Acceptable)

Loss is not the same as beverage cost percentage — it is the gap between theoretical and actual usage. The distinction matters because some operations run high beverage cost intentionally through product mix and pricing, while others have high beverage cost because product is walking out the door.

Normal and acceptable variance:

A well-managed bar with consistent counting, trained staff, and regular monitoring typically runs one to two percent variance between theoretical and actual usage
This accounts for legitimate measurement imprecision, minor spillage, and normal pour variation
One to two percent is not a red flag — it is the realistic floor for a bar operating with good controls

Elevated but manageable variance:

Three to four percent variance is worth monitoring and investigating by category
At this level, the variance may be driven by a single product, a specific shift, or a measurement issue in the counting process
It should not be accepted as normal without understanding what is causing it

Concerning variance:

Five percent or above is a significant problem regardless of concept
At this level, the cause is almost certainly not measurement error — it is operational, behavioral, or a combination of both
A bar doing $15,000 per week in beverage sales at five percent variance is losing $750 per week — $39,000 per year

What creates the difference between controlled and uncontrolled loss:

Controlled loss is logged — spillage documented, breakage recorded, legitimate comp entries in the POS
Uncontrolled loss is unexplained — the count is short, the theoretical says X should have been used, Y was actually counted, and no one knows where the gap came from
Uncontrolled loss that is consistently unexplained is the most important signal that something deliberate may be contributing to the variance

How to Reduce Alcohol Loss

Tighter pour control:

Measured pours using jiggers are the most effective mechanical control for over-pouring — research consistently shows that jigger use reduces pour variance by 15 to 30 percent compared to free-pouring
Speed pourers with measured triggers are a middle ground that maintains service speed while limiting pour size
Periodic weighed pour tests — weighing a bartender’s poured drink on a scale against the recipe spec — create accountability without requiring a full count
Training new bartenders to the pour standard before they go on the floor is more effective than correcting drift after it has established itself

Comp tracking:

Every comp should be entered into the POS before the product leaves the bar — not after service, not at the end of the night
Manager authorization for comps above a defined value creates a record and a control without eliminating the practice
Weekly comp reports compared against beverage cost variance identify whether untracked comps are a significant driver of the gap

Manager accountability:

Beverage cost percentage should be assigned to the bar manager as a named target, reviewed weekly, and connected to their performance evaluation
A bar manager who knows their beverage cost is measured weekly and discussed as part of their performance manages the bar differently than one who sees a monthly report with no consequence attached

Standardized counting:

The same counter, the same sequence, the same methodology for partials, every week — consistency is what makes variance data meaningful
Rotating who does the count without standardizing the method produces results that cannot be compared reliably
Investing two hours per week in a properly done count is worth significantly more than the same two hours spent on any other operational task in a bar with a loss problem

What Tools Help Control Alcohol Inventory

BlueCart

Role in alcohol control: BlueCart addresses alcohol loss at the ordering level. Centralized purchasing management, delivery tracking, and price comparison across suppliers reduce over-ordering, catch delivery discrepancies, and create a record of what was purchased. For bars dealing with receiving errors — being invoiced for product that did not arrive or at prices that do not match the agreement — BlueCart provides the documentation to dispute them. Does not track beverage cost percentage, variance, or pour analysis.

MarginEdge

Role in alcohol control: MarginEdge processes every invoice including beverage purchases and connects cost data to POS revenue for a daily P&L that includes beverage cost. Tracking beverage cost separately from food cost through MarginEdge’s invoice processing gives operators daily visibility into whether alcohol cost is trending above target — catching problems during the period rather than after it closes. Does not provide pour-level variance analysis or bottle-count comparisons.

MarketMan

Role in alcohol control: MarketMan supports beverage inventory counting alongside food inventory. Weekly counts entered into the platform generate theoretical versus actual usage comparisons by category, including beverage. Recipe pour costs connected to live invoice pricing update automatically when supplier prices change. The variance report identifies which beverage categories are over-running theoretical usage, giving operators a specific starting point for investigation rather than an aggregate number to interpret. For operations with a bar program that is not their primary revenue driver, MarketMan handles both food and beverage inventory adequately.

Restaurant365

Role in alcohol control: Restaurant365 includes beverage cost in its consolidated financial reporting. For multi-location groups where bar performance varies by unit, location-level beverage cost reporting identifies which units are the outliers and where investigation should be prioritized. As with food cost, the strength of Restaurant365 in this area is consolidation and financial visibility across multiple locations rather than bar-specific operational detail.

Note on dedicated beverage platforms: For high-volume bars, full-service restaurants where beverage represents 40 percent or more of revenue, or operations with persistent alcohol cost problems, a dedicated beverage management tool like Craftable adds bar-specific capabilities — pour calculators, bottle sliders, keg credit tracking, and spirit-by-spirit cost analysis — that food-first platforms do not provide at the same depth.

The Simplest Way to Start

If alcohol inventory is not currently tracked formally, here is the most practical starting sequence.

Count weekly:

Pick a consistent time — Sunday before opening, Monday morning before the first delivery — and count every bottle in every location
Use a consistent methodology for partials from the first count forward
Do not skip weeks — two consecutive consistent counts are more valuable than ten inconsistent ones

Track variance:

Calculate what was used: beginning count plus purchases minus ending count
Calculate what should have been used: number of each drink sold multiplied by the recipe pour size, totaled by spirit category
The gap between the two is your variance — identify whether it is within the one to two percent range or above it

Separate beverage cost from food cost:

Track beverage invoices separately from food invoices
Pull beverage revenue from your POS separately from food revenue
Calculate beverage cost percentage independently and manage it against a separate target from food cost

Review consistently and require explanations:

A beverage cost or variance number that is reviewed and filed without explanation does not improve anything
The review question is not just what the number is but why it moved — which category, which shift, which week
Variance that is explained is manageable. Variance that is attributed to “it was a busy week” is a problem that will recur.

Final Takeaway

Alcohol loss is real, often significant, and almost always more invisible than food loss because the tracking processes most restaurants apply to beverage are less rigorous than those applied to food. The financial consequence of that gap is not small — on a bar doing $400,000 in annual alcohol sales, the difference between one percent variance and five percent variance is $16,000 per year in avoidable cost, representing $48,000 to $64,000 in foregone high-margin revenue.

The tools to fix it are not complicated: weekly counts done consistently, theoretical pour usage compared against actual, beverage cost tracked separately from food cost, and a named manager accountable for the number. None of these require sophisticated software to start — though software makes them faster, more reliable, and more actionable once the counting habit is established.

The operations that run tight beverage cost are not running a different bar from everyone else. They are measuring the same things more consistently, investigating variance more specifically, and holding their teams accountable to numbers rather than feelings.

Start counting. Compare to theory. Investigate what does not match. The rest follows.

If you want to compare the cost of the tools mentioned here against the losses outlined above, see Restaurant Inventory Software Pricing Comparison.

Visit MarketMan — Get a Demo

Visit MarginEdge — See Pricing

Visit Restaurant365 — Request a Demo

Visit BlueCart — Start Free

FAQ

How do you track alcohol inventory?

Count every bottle in every storage location on a consistent weekly schedule, using a consistent methodology for partials. Compare the physical count against what should have been used based on POS sales data and recipe pour sizes. The gap between theoretical and actual is your variance — the starting point for identifying where loss is occurring. Dedicated tools like MarketMan automate this comparison once counts are entered. For high-volume bars, a dedicated beverage platform like Craftable adds pour-level analysis that food-first platforms do not provide.

What is a good beverage cost percentage?

Spirits: 18 to 22 percent is a common target in well-managed operations. Craft cocktail programs may run 22 to 26 percent intentionally. Draft beer: 20 to 25 percent accounting for legitimate line loss. Wine by the glass: 25 to 35 percent. Blended overall beverage cost for full-service restaurants with a bar program: 18 to 24 percent. These ranges assume consistent counting, a working pour standard, and POS sales data that accurately captures beverage categories separately from food.

How much alcohol loss is normal?

One to two percent variance between theoretical and actual usage is typical in well-managed bar operations with consistent counting and trained staff. Three to four percent is elevated and worth investigating by category. Five percent or above is a significant operational problem regardless of concept — at that level, the cause is almost certainly not measurement error alone. The distinction between normal and concerning loss is whether the variance is consistent and explainable or inconsistent and unexplained.

How often should I count bar inventory?

Weekly. Monthly counts catch problems a month after they start, by which time a significant portion of the damage is already done. Weekly counts produce a corrective window — if variance is elevated in week two, you have two weeks left in the period to investigate, identify the source, and address it before month-end. Weekly counting also produces more accurate partial estimates because shorter intervals mean partials change less between counts, making them easier to measure consistently.

Can software help control alcohol inventory?

Yes, with the right expectations. Software makes counting faster, automates the theoretical versus actual comparison, and produces variance reports that identify categories and locations driving the problem. What software cannot do is stop a bartender from over-pouring, catch a comp that was never entered into the POS, or replace the management conversation that follows when variance is identified. Software exposes the problem. Accountability, training, and operational response are what reduce it.