Restaurant Purchase Approvals: How Multi-Location Operators Stop Vendor Overspend
If you operate even two restaurant locations, you already know the problem. Each GM has their own vendor relationships. Food cost varies wildly between locations selling the same menu. The wine guy at Location A is up-charging you 11% over what Location B pays. Last quarter's beverage spend is up 18% and nobody can explain why.
This is the multi-location restaurant operator's tax — and it's almost entirely avoidable with a real purchase approval workflow.
Why restaurants are uniquely exposed
Restaurants have a few characteristics that make spend control especially hard:
- Many small daily purchases — produce, proteins, dry goods, beverages, paper goods, cleaning supplies — that add up fast
- Decentralized buying — GMs, sous chefs, beverage directors, and floor managers all order from different vendors
- Time pressure — orders happen between services, often on phones, with no time for paperwork
- Vendor relationships built on trust — which means pricing creep is easy and rarely audited
- Multiple locations doubling, tripling the surface area
Add it all up and a 4-location restaurant group can leak 3–8% of revenue annually to uncontrolled spend variance.
The problem with how most restaurants try to fix this
Most operators try one of three approaches and hit the same wall:
1. The spreadsheet ledger. Operations puts together a master vendor list, sets target prices, and asks managers to log every order. It works for two weeks. Then a manager forgets, then another, then nobody updates it, then it's abandoned.
2. The "ask before you order" rule. GMs are told to text the owner before placing any order over $500. The owner is in service or at home or on a call. Orders happen anyway. The rule dies in 30 days.
3. The corporate card lockdown. Replace open accounts with a corporate card per manager and watch the statement. This catches what they spent but not whether they should have. By the time you see the charge, the food is already in the walk-in.
What works is something different: a lightweight, mobile-first approval workflow that takes 60 seconds per request and gives you visibility before the order is placed.
The 4-step workflow that works for multi-location restaurants
Step 1: Set per-location budgets and per-category thresholds
For each location, set monthly budgets per category. A typical 100-seat restaurant might look like:
- Food: $32,000/month
- Beverage: $12,000/month
- Paper/cleaning: $1,800/month
- Maintenance/repair: $1,200/month
Then set approval thresholds per category. Common pattern:
- Under $200: GM auto-approves (logged but no review)
- $200–$1,000: Director of Operations approves
- $1,000–$5,000: Owner approves
- Over $5,000: Owner approves + 24-hour cool-off
These numbers vary by concept and volume — adjust to your business.
Step 2: Mobile-first request submission
Every order request goes through a phone — because that's where managers actually are. The request includes:
- Vendor
- Category (food, beverage, supplies, etc.)
- Estimated total
- Brief justification ("running short on chicken thighs for weekend service")
- Optional: photo of inventory or order sheet
This takes 60 seconds. If it takes longer, managers will route around it.
Step 3: Approval in one tap
The approver — often the director of operations or the owner — gets a push notification, opens the app, and approves or rejects in a single tap. Most approvals happen in under 10 minutes. Auto-escalation kicks in if no response in 2 hours during business hours.
Step 4: Mandatory invoice capture at delivery
When the order arrives, the receiver (often a line cook or prep cook) takes a photo of the invoice and attaches it to the request. This:
- Confirms what was actually delivered
- Captures the real price (not the estimate)
- Creates an audit trail per location, per vendor, per category
- Makes month-end reconciliation effortless
What you'll see in the first 90 days
Multi-location restaurant operators who put this in place typically report:
- Vendor pricing variance shrinks 40–60%. When prices are visible across locations, GMs can compare and call out their reps. Reps stop pushing prices for fear of losing the account.
- Beverage spend drops 5–10%. This is consistently the biggest leak. Hard to explain why a $3 well pour costs different amounts at two locations selling the same drink — but the data shows it does.
- One-off "emergency" orders fall. When managers know "emergency" orders need to be flagged and justified, they plan ahead.
- Month-end close goes from 3 days to 1 day. Receipts are already attached, categorized, and tagged to a location.
Multi-location complexity: the real challenges
A few extra wrinkles for operators with multiple locations:
Per-location vendor accounts. Each location may have its own account with the same vendor (different account numbers, different reps, different pricing). Your approval system needs to track which location each request is for so reporting rolls up correctly.
Cross-location transfers. Sometimes Location A loans Location B a case of something. Your system should support an "internal transfer" type so it doesn't show up as a duplicate purchase.
Catering or events. Big one-off purchases for events should have their own approval flow with the event budget tied to revenue, not the regular monthly category budget.
New menu launches. When a new menu drops, prep purchases spike and approvers need context. A "menu launch" tag on requests helps approvers say yes faster without losing visibility.
What to look for in a tool
For multi-location restaurants, the must-haves:
- Mobile-first. Web-only tools die in this industry. Approvers and requesters live on phones.
- Per-location reporting. You need to see Location A vs Location B at a glance.
- Category and vendor breakdowns. Where the variance lives.
- Receipt photo capture. The cook receiving the order needs to snap a pic in 5 seconds.
- Push notifications. Email approvals get buried during service.
- Department/cost-center tags. Front-of-house vs back-of-house budgets are different.
Becision was built with exactly this workflow in mind — multi-location operators are one of our most common customers. Free for small teams; works for 1 location to 50. Set up an account →
The honest expectation-setting
A purchase approval workflow won't fix:
- A bad GM (they'll route around any system)
- Bad menu engineering (high-cost dishes that don't sell)
- Theft (you need POS controls and inventory counts for that)
- Vendor kickbacks (audits and rotation help)
What it will fix: invisible spending, vendor pricing creep, missing receipts at month-end, and you being a bottleneck for every $300 chicken order.
For multi-location operators, the ROI is usually obvious within the first quarter. If your spend variance across locations is more than 5%, the math is on your side.