Case Study 15.2 🔬 Deep Dive: The Bar at Wildrye

📜 From the Field

The Reyes–Whitfield wedding is a Tier 3 illustrative composite. Every price, licence structure, and vendor below is constructed. The arithmetic reproduces from §15.2 and §15.3, and the raw-site ice finding in Act Four is a real and widely under-appreciated constraint.

The Constraint Set

Everything this bar has to live inside, assembled from six previous chapters:

And one fact worth stating before anything else: the $3,100 was allocated in Chapter 6 at **118 expected guests** — $26.27 a head. It has never been re-based. At 100 guests it is $31 a head, and this is one of the very few places in this project where the guest reduction produced genuine slack rather than a proportional cut.

It has not been re-cut, and it should not be. Chapter 14's rule applies: a re-cut is an allocation decision governed by the ranking, and the party is ranked second.

Act One: The Forecast

💰 Run the Numbers: the Reyes–Whitfield drink forecast

Drinkers: 100 − 6 children − 6 non-drinking or minimal = 88.

(The six: Hana Whitfield, who is pregnant; two guests who do not drink; and three who will have one and stop. The dietary matrix and the guest list between them already knew this.)

Duration: 5:10 p.m. to 10:30 p.m. = 5 hours 20 minutes = 5.33 hours.

First hour: 88 × 2.0	176
Remaining 4.33 hours: 88 × 1.0 × 4.33	381
Base	557

Adjustments, each applied to the base:

Factor
No shuttle; rural; 90 minutes from the city; guests are driving	−12%	−67
Warm September afternoon, outdoors, long cocktail hour on a lawn	+8%	+45
Dancing-heavy — the party is ranked second	+10%	+56
Mixed ages across both families	0	0
Adjusted		591
Shrinkage, +10%		+59
FORECAST		650 drinks

7.4 drinks per drinking guest across 5.33 hours — 1.4 an hour. That is a sane number for a wedding with a long dinner, wine on the table, and a hard 10:30 close.

🔍 Why Does This Work? §15.3's worked purchase order used 615 drinks for a generic 100-guest event. This one lands at 650, and the whole of the difference is in two lines that pull in opposite directions.

The −12% driving adjustment takes 67 drinks off. The +10% dancing adjustment and the longer, warmer cocktail hour put 101 back on.

That is the value of a decomposed forecast. A rule of thumb cannot tell you that this wedding's guests will drive home and dance hard, or that those two facts nearly cancel. Two lines on a worksheet can.

And note the uncomfortable thing the first line does, which Act Six comes back to: it reduces the alcohol budget because guests will be driving. Having written that down, you cannot then be relaxed about guests driving.

The mix

Both families, a family-style dinner with wine on the table, September, and a couple who have asked for a signature cocktail.

Note the non-alcoholic share is 8%, not the 5% default. §15.7's argument, applied: there are twelve people at this wedding who are not drinking alcohol at all and they are there for five and a half hours. Five per cent of the drinks would give them two each.

And the wine splits further, because the toast is a spike rather than a rate:

The 44 still-wine bottles are cross-checked against §15.3a's table figure: 13 rounds × 2 bottles at service, refreshed on about 8 tables = 34 bottles across dinner. That fits inside 38 with four left for the bar. Consistent.

Act Two: The Signature Cocktails

The couple wanted one. They are getting two, and both are batched.

📋 The Planner's Script: proposing the drinks

"Two thoughts, and they're both cheaper than a full bar rather than more expensive.

A paloma, batched — tequila, grapefruit, lime, salt on the rim. It's your family's drink, Alicia, it's the right drink for a warm afternoon on a lawn, and because we mix it in advance it pours in about eight seconds instead of forty-five. Which means the queue at half five doesn't happen.

And something with apples and bourbon for September, which is Sam's side of the room and is the drink people will want after dinner when it's cooler.

The paloma also comes without the tequila, same glass, same grapefruit, same salt, and I'm putting it on the same list as everything else with no label on it. Hana's not drinking and neither are five other people, and I'd like them to be handed a proper drink rather than a glass of juice.

That's about a hundred and eighty dollars of tequila and bourbon, total. It's the cheapest personal thing in the whole wedding."

Act Three: The Purchase Order

💰 Run the Numbers: the Reyes–Whitfield beverage order

Item	Basis	Qty	Unit	Total
Still wine	189 servings ÷ 5, +15%	44 bottles	$12.00 | $528.00
Sparkling	110 ÷ 6	20 bottles	$14.00 | $280.00
Beer, bottles	163, 7 cases of 24	168	$36.00/case | $252.00
Tequila blanco	~90 palomas ÷ 16	6 × 750 ml	$24.00 | $144.00
Bourbon, vodka, gin	~47 pours + Uncle Ray	3 × 750 ml	$26.00 | $78.00
Grapefruit soda and juice	~1 L per 9 drinks	20 L	$2.20 | $44.00
Mixers, tonic, soft drinks, cider	for 51 NA + children	—	—	$95.00
Garnish — limes, lemons, salt, rosemary		—	—	$48.00
Ice	see Act Four	400 lb	$0.34/lb | **$136.00**
Goods subtotal				$1,605.00
Sales tax, 7.75%				$124.39
BEVERAGE TOTAL				$1,729.39

$17.29 per guest for a 5⅓-hour hosted bar with two signature cocktails.

Three things about that order.

**The wine is $12 a bottle and this is deliberate.** §15.3: nobody tastes bottle four. The sparkling is $14 because the toast is the one pour of the night that everybody drinks at the same moment while looking at it.

44 bottles is 15% over the 38 the forecast requires, because the retailer has agreed in writing to accept returns on unopened, undamaged, labelled bottles. The beer and the sparkling are bought to forecast, not over it, because chilled beer and inexpensive sparkling generally cannot go back.

Nine bottles of spirits, not twenty-two. Six of them are tequila, because 90 of the 137 spirit drinks are palomas. A "complete bar" list would have added twelve more bottles at roughly $310, to serve perhaps two drinks each.

⚖️ The Clause: the corkage that did not apply

Sundown's catering proposal contains a line most planners skim:

"Beverage service not included. Corkage of $9.00 per bottle applies to alcohol served by Sundown staff."

At 64 bottles of wine that is $576 — nearly a fifth of the entire bar allocation.

It does not apply, because Wildrye's contract §9 requires a licensed bartending service and the bar is therefore not being poured by Sundown. But that is an inference, and inferences are not a defence against an invoice.

Confirmed in writing with Marguerite at Sundown: "Just so we're clear on the beverage line — the bar is being run by Ridgeline Bar Co. under their own licence and none of your staff will be pouring, so I'm reading the corkage clause as not engaged. Can you confirm?"

She did. The whole exchange took nine minutes and was worth $576.

Act Four: The Ice Problem

This is the part that only exists at a raw site, and it is the part almost nobody gets right.

🧩 Productive Struggle

§15.3 gives the ice rule as 2.5 lb per guest — 1.5 for drinks and 1 for chilling. At 100 guests that is 250 lb.

The order above buys 400 lb. Why?

Work it out before reading on. The relevant facts are all above.

The answer

Because the 2.5 lb rule assumes a venue with refrigeration, and Wildrye has none.

At a normal venue, wine and beer arrive cold or are chilled in a cold room, and the ice is for drinks plus topping up. At Wildrye there is nothing cold on the site at all.

Everything that has to be cold has to be made cold with ice:

Still white and rosé, to chill	30 bottles
Sparkling, to chill	20 bottles
Beer, to chill	168
Units requiring chilling	218
At ~1 lb of ice per unit in a tub of ice water	218 lb
Ice into drinks: 1.5 lb × 100	150 lb
Melt loss on a warm September afternoon, ~10%	37 lb
TOTAL	405 lb → order 400

The working figure for a raw site is 4 lb per guest, not 2.5.

And it cannot arrive in one delivery

🕐 From the Run Sheet: the ice schedule

Time		Owner
1:00 p.m.	First delivery: 220 lb. Wine, sparkling, and beer into tubs of ice water	Bar lead
1:00–5:00	Four hours of chilling. This is the constraint that sets the delivery time	—
4:30 p.m.	Second delivery: 180 lb, for drinks	Bar lead
5:10	Bar opens	Bar lead
8:00	Tubs topped from reserve; empties begin moving to bins	Bar lead

The four-hour chill is why the first delivery is at 1:00 p.m. and not at 4:00. A bar that opens at 5:10 with warm white wine has failed at something nobody will name but everybody will notice.

And the second delivery exists because the first one melts. 220 lb delivered at 1 p.m. and used for chilling is largely water by five. Ice delivered at 4:30 goes into drinks.

Cost of the second delivery: $40. It is the cheapest line in this case study and the one most likely to be cut by somebody trying to save money.

What holds it

Four insulated beverage tubs and two large coolers, which are not in the Chapter 13 rental order because nobody thought about where 400 lb of ice lives.

Added to the rental order at $85.

🚨 When It Goes Wrong: the gap between two vendors' assumptions

Chapter 13's rental order was built from the floor plan, the guest count, and the caterer's exclusions list. Ice tubs appear on none of those.

The bar service assumed the venue or the rental order would have them. The rental order assumed the bar service would bring them. Neither is wrong and the tubs do not exist.

This is the same failure as Chapter 13's glassware gap and Chapter 14's vendor meals, and it has the same fix: ask each vendor, in writing, what they are assuming somebody else is bringing. It is one question, it goes in every RFP from now on, and it has found something at every event where it has been asked.

Act Five: The Bar Service

Wildrye's contract §9 does not permit any other arrangement.

💰 Run the Numbers: Ridgeline Bar Co.

2 bartenders × 6.5 hours (4:00 p.m. setup – 10:30 p.m.) × $48 | $624.00
Travel surcharge — 90 minutes rural, each way	$120.00
Labour subtotal	$744.00
Service charge, 18%, distributed to staff	$133.92
BAR SERVICE TOTAL	$877.92

No separate gratuity — the 18% is distributed and Ridgeline states so in writing. §15.9: no tip jar on the bar.

Two bartenders for 100 guests is Chapter 11's ratio (one station per 75–100) at the top of its band. Act Six checks whether that is enough.

The file

✅ Best Practice: what was collected from Ridgeline, and when

Document	Status
Copy of licence	✅ Received −52 days
Written confirmation that their licence permits pouring client-supplied alcohol at a private event	✅ Obtained before any alcohol was bought
Liquor liability certificate, $1,000,000, Wildrye Farm LLC as additional insured	✅ Received −16 days (Ch.8 §7 deadline is −14)
General liability certificate	✅ Received −16 days
Written service plan: hours, last call, refusal protocol	✅ −21 days
Named refusal authority	✅ Devin Okonjo, Ridgeline bar lead. Alicia and Sam informed in writing

The county's temporary event alcohol permit is not on this list, because it is on Chapter 9's permit list with its own owner and its own filing date, and its cost sits in the $675–$2,800 permits line rather than in the bar allocation. Checked, not double-counted.

Act Six: The Cocktail-Hour Crunch

💰 Run the Numbers: 5:10 to 6:20, and whether two bartenders can do it

First-hour demand: 88 drinkers × 2.0	176 drinks
Of which in the first 20 minutes (~45%)	79
2 bartenders, mixed bar, 38/hr each	76/hr = 25 in 20 min
Backlog at minute 20	54
Demand across the 70-minute cocktail hour	205
Served across 70 minutes	89
Backlog at minute 70	116

The queue does not clear. It ends when everyone sits down at 6:32.

The fixes, applied

Cost of all three: two serving trays, a decision about the cocktails, and a tub that is already on the equipment order.

Zero additional bartenders. Chapter 11 said the queue is a design problem before it is a staffing problem, and this is that claim tested against a real event.

📐 On the Floor Plan: where the bar goes

Chapter 11 placed the bar on the north side of the tent, against the sidewall, with the queue running along the wall rather than into the room.

Two additions from this chapter:

1. The self-serve beer tub goes at the far end of the bar, not beside the bartenders — otherwise it joins their queue rather than bypassing it.
2. The passed-drink trays are loaded outside the tent, at the kitchen marquee, so servers do not cross the bar queue to restock. Chapter 11's three flows, still not crossing.

The Total

⚡ Quick Reference: the Reyes–Whitfield bar, complete

Beverage (goods + tax)	$1,729.39
Bar service (Ridgeline)	$877.92
Ice tubs and coolers (added to the Ch.13 rental order)	$85.00
Second ice delivery	$40.00
TOTAL	$2,732.31
Allocation	$3,100.00
UNDER BY	$367.69

This is the first line in the entire project to come in under allocation.

It is worth being precise about why, because it was not skill:

Three of those five are decisions. Two are accidents of the allocation. Say so to the couple; a planner who takes credit for a favourable base number will eventually have to take blame for an unfavourable one.

What the $367.69 Is For

🚨 When It Goes Wrong: the line you wrote and then ignored

Go back to the forecast. The largest single adjustment was:

No shuttle; rural; 90 minutes from the city; guests are driving. −12%.

That line reduced the alcohol budget by 67 drinks — roughly $170 — on the explicit assumption that a meaningful number of guests will drive home from a wedding.

You cannot write that down and then not act on it. Wildrye is 90 minutes from the city and fifteen from the nearest town. Rideshare coverage was tested on the site visit and is intermittent. There is one taxi firm.

The bar closes at 10:30 and a hundred people leave at once.

Three options, priced

📋 The Planner's Script: presenting the transport problem

"The bar came in three hundred and sixty-seven under, and before you spend it I want to show you what I'd spend it on.

When I forecast the drinks, I took twelve per cent off because people are driving. And then I read that back and thought — well, that's the problem, isn't it. Everyone leaves at half ten, we're ninety minutes from town, and the rideshare signal out there is patchy. I tested it.

Three options.

One — a minibus, two runs, ten-fifteen and eleven o'clock, into Cairnfield where the inn and the two B&Bs are. Four hundred and eighty dollars. That's a hundred and twelve more than we have, and I'd take it out of the design contingency.

Two — one run at eleven, two hundred and sixty. Covers the people staying in town, doesn't cover anyone leaving early.

Three — no bus, and a designated-driver plan instead. Free. You two pick the drivers off the guest list, they get a card at their place setting, the bar team know who they are and look after them, and I book the one taxi firm for eleven o'clock whether we need it or not.

Three is genuinely better than nothing and it is not as good as a bus. I'd do one. But it's your money and it's your guest list, and you know who's driving better than I do."

They took option one, funded by the $367.69 plus $112.31 from the design contingency.

🔍 Why Does This Work? Compare this with Case Study 15.1, where a planner reported an underspend and the couple gave it to the band.

The difference is not that Devon was careless. It is that he reported a number and this planner reported a number attached to a finding. An underspend presented on its own is money; an underspend presented next to the line that produced it is a decision.

The general practice: when a category comes in under, the first question is not "what else could this buy?" It is "did I make it cheap by assuming something I now need to pay for?"

What Is Still Open

Discussion Questions

1. The forecast's largest adjustment was −12% for guests driving, and the case argues you cannot write that and then not act on it. Is that a genuine logical obligation or a rhetorical move? Suppose the adjustment had been −3%. Does the obligation scale?

2. The bar came in under partly because the allocation was set at 118 guests and never re-based. Should it have been re-based? Chapter 14 restored a cut to catering on the grounds that the ranking governs. Does the same argument work in reverse — and if a category is over-allocated relative to need, who is entitled to the difference?

3. The corkage clause did not apply, and confirming that took nine minutes and was worth $576. Design the process that catches clauses like this systematically, rather than because a planner happened to read carefully. What document does it live in?

4. Ice at a raw site is 4 lb per guest, not 2.5. The chapter's own quick-reference says 2.5. Should the book's rule be changed, or is a rule with a stated exception better than a more complicated rule? Argue it, then write the version you would put in Appendix B.

5. The queue was solved with passed drinks, batched cocktails, and a self-serve beer tub — none of which the couple asked for and all of which change the feel of the cocktail hour. Whose decision are these? Chapter 3's decision-rights map put design decisions with the couple.

6. Option three — the designated-driver plan — is free and partial. Construct the argument that offering it at all is irresponsible, because it lets a couple choose the cheap option for a problem where the downside is somebody's life. Then respond.

7. Two bartenders serve 100 guests here, and the fix for the crunch was entirely demand-side. At what guest count does the demand-side toolkit run out and a third bartender become genuinely necessary? Compute it.

Mini-Project

Build a complete bar plan for a real event.

Pick a real venue and a plausible guest list. Then:

1. Establish the operating window — as a number, from a real timeline, not an estimate.
2. Forecast, showing drinkers, base, every adjustment with its direction, and shrinkage.
3. Split by mix, with the non-alcoholic share at 8% rather than 5%, and justify any change.
4. Convert to a purchase order at real local prices, with a written returns policy.
5. Compute the ice properly, including chilling load if the venue has no refrigeration.
6. Price the service under whatever arrangement the venue and your jurisdiction permit — and establish what they permit before you price it.
7. Run the cocktail-hour crunch and solve it demand-side first.
8. Write the responsible-service plan, with a name in it.
9. Then ask the question this case ends on: did I make anything cheap by assuming something I now have to pay for?

Deliverable: a two-page bar plan and a one-page purchase order. Steps 5, 6, and 9 are the ones that separate this from a shopping list.

Connections

• Chapter 5 — fully variable costs, and why a bar scales down where a tent does not
• Chapter 6 — the $3,100, allocated at 118 and never re-based
• Chapter 8 §9 — the clause that requires licensed service and ends service at 10:30
• Chapter 9 — the permit, the certificates, and the named refusal authority
• Chapter 11 — the bar's position, the queue's direction, and the three flows
• Chapter 13 — the glassware already ordered, and the tubs that were not
• Chapter 14 — the dinner block that sets the operating window, and the corkage clause in Sundown's proposal
• Chapter 16 — the bar as the largest piece of furniture in the room
• Chapter 25 — the four minutes when both bartenders are pouring the toast
• Chapter 30 — 232 empty bottles and a 10 a.m. Sunday deadline