Theoretical food cost depends on R365 understanding exactly what was sold and applying the correct recipe to that sale.

A critical integration issue has been identified in which modifiers can lose their association with the parent menu item when the sales mix is imported from Toast into R365.

This may sound like a relatively minor technical problem, but it can have a significant impact on theoretical food cost.

Consider a simple example.

A restaurant sells three quesadillas. One is a small quesadilla with chicken, while another is a large quesadilla with steak.

In Toast, the transaction contains the relationship between the parent menu item, the size, and the protein modifier. Toast knows that the small portion belongs with the chicken selection and the large portion belongs with the steak selection.

If that relationship is not preserved when the sales information is imported into R365, R365 can lose the context necessary to determine which recipe should be applied.

R365 may effectively receive information indicating that a quesadilla was sold, a small modifier was used, a large modifier was used, chicken was selected, and steak was selected – but not reliably retain the relationship between small + chicken and large + steak.

That creates a fundamental problem:

The theoretical food-cost calculation may be mathematically correct based on the information R365 receives while still being operationally incorrect because the relationship between the menu item and its modifiers has been lost.

This is an important distinction for management. The problem should not automatically be characterized as an R365 AvT calculation problem. It is more accurately described as a data-integrity problem at the Toast-to-R365 integration layer that directly affects the AvT calculation.

The recovery project must therefore address the entire data flow rather than simply attempting to rebuild recipes or adjust food-cost percentages.

The Core Problem: The Multi-Brand Data Chain Is Not Reliable

Accurate theoretical food cost depends on a continuous chain of information:

Toast Menu Item → Modifier/Portion → R365 Recipe → Ingredient → Purchase Item → Unit of Measure → Cost → Inventory Depletion → AvT

Every link must work correctly.

For a two-brand restaurant management group, there is an additional dimension:

Brand → Menu → Recipe → Ingredient → Purchase Item → Vendor → Cost

Some ingredients may legitimately be shared between the brands. Others may look similar but represent different products, specifications, brands, vendors, or recipes.

The system must therefore distinguish between: 

• Shared products that should intentionally use common items 
• Brand-specific products 
• Brand-specific recipes 
• Brand-specific menu items 
• Brand-specific vendors or specifications 
• Common ingredients with different costs 
• Similar products that should remain separate 

Without clear governance, a multi-brand R365 environment can become difficult for restaurant managers and corporate users to navigate. 

The current review has identified issues throughout this chain, including incomplete or incorrectly linked recipes, cross-brand recipe relationships, obsolete recipes, duplicate items, inconsistent naming conventions, Toast modifier and portion limitations, inaccurate units of measure, invoice-processing issues, inventory-count errors, late invoices, and inconsistent inventory approval practices. 

More than 4,000 obsolete or duplicate recipes have already been removed. That is a significant accomplishment, but it also demonstrates the level of complexity that accumulated in the system. 

The objective is not simply to make the next AvT report look better. 

The objective is to create a controlled multi-brand data environment where the numbers can be trusted every period. 

The first phase should focus on the most visible and actionable problems while the broader remediation effort begins.

For the significant theoretical food-cost change identified in a recent period, the project team should perform a Period 7 versus Period 8 comparison at the item and recipe level.

The objective is to identify:

1. Which items changed?
2. Which recipes changed
3. Which modifiers or portions changed
4. Which costs changed
5. Which inventory movements changed
6. Whether the change represents actual operational performance or a system/data issue.

For example, one brand historically operating in an approximately 18 -20% theoretical food-cost range declined to approximately 16.8% in Period 8. The other brand has reported theoretical food cost near 13%.

These results should not automatically be interpreted as operational improvement.

They should first be treated as exceptions requiring validation.

Several fundamental controls should be implemented immediately:

• Confirm that AvT reports are configured consistently.
• Confirm that Net Sales is being used.
• Ensure weekly inventory counts are completed and approved.
• Identify inventory counts left in “In Progress” status.
• Review invoice timing and period posting.
• Identify unusual cost movements.
• Validate recipe changes before accepting significant variance.
• Standardize the report configuration used by Operations, Accounting, and executive management.
• Require an explanation for significant or unusual food-cost variances before reporting them to senior leadership.

These controls provide an immediate protective layer while the larger project is underway..

Rebuild the R365 Multi-Brand Foundation

The Importance of Standard Operating Procedures

The next phase should address the underlying R365 configuration.

Every recipe should ultimately be reviewed to confirm:

• Correct brand
• Correct menu item
• Correct ingredients
• Correct portions
• Correct purchase items
• Correct units of measure
• Correct costing
• Correct Toast mapping
• Correct recipe application

The review should cover 100% of the recipes for both brands, rather than relying solely on sampling.

Although the initial investigation has found relatively few cross-brand recipe issues compared with the total population, the management group needs complete confidence that one brand is not inadvertently using ingredients, recipes, or costs associated with the other.

The distinction between shared and brand-specific items should also be formalized.

Shared Items

Where both brands intentionally purchase and use the same product, the R365 item should be standardized and governed centrally.

Brand-Specific Items

Where the brands use different specifications, brands, vendors, pack sizes, or recipes, those items should remain clearly differentiated.

A standardized naming convention should make this distinction obvious to users.

This is particularly important because the restaurant manager should not have to determine whether two similarly named products are actually interchangeable.

Recipes should no longer be treated as static records created during implementation.

They should have a controlled lifecycle:

Create → Validate → Approve → Publish → Monitor → Retire

A new menu item should not be considered production-ready until both the Toast configuration and the R365 recipe have been validated.

Similarly, when a menu item changes, the change must be evaluated across both platforms.

The process should identify:

• Who creates the recipe
• Who validates the recipe
• Who approves the recipe
• Who changes the Toast item
• Who validates the integration
• When the change becomes effective
• How the old recipe is retired
• How the change is communicated to the restaurants

Obsolete and duplicate recipes should be archived rather than remaining available to restaurant users.

The removal of more than 4,000 obsolete recipes is an important first step toward creating this cleaner environment. 

Address the Toast Modifier and Portion Problem

The Toast integration issue deserves its own formal workstream.

The project team should document how Toast represents:

Parent Item → Modifier Group → Modifier → Portion → Recipe

and compare that structure with how R365 receives and interprets the same information.

Representative transactions should be tested from beginning to end:

Toast Transaction → Toast Sales Mix → R365 Imported Sales → R365 Recipe Application → Theoretical Depletion

Testing should deliberately include complex examples rather than only simple menu items.

For example:

• Small vs. large portions
• Multiple proteins
• Multiple modifiers
• Modifier combinations
• Add-ons
• Substitutions
• Shared ingredients
• Brand-specific recipes
• Items with different portion costs

The team should also evaluate R365’s forthcoming rules-based Toast import capability before performing significant manual remediation of modifier issues.

If the new functionality can preserve or reconstruct the required parent-item/modifier relationship, it could eliminate a significant amount of manual work.

The project team should therefore establish a decision point:

Do not manually remediate large populations of Toast modifier issues until the new R365 integration capability has been evaluated.

Temporary corrections should be limited to known high-impact exceptions.

Unit-of-measure problems represent another significant threat to food-cost accuracy.

An invoice can contain the correct product and price but still generate an incorrect cost if R365 interprets a case as an individual unit or an individual unit as a case.

The review should cover:

• Purchase items
• Vendor items
• Pack sizes
• Units of measure
• Conversion factors
• Catch-weight products
• Recipe units
• Vendor pricing
• Invoice pricing
• Brand-specific products

This is also an area where the management group should evaluate R365 Capture AI.

If Capture AI can validate invoices, improve unit-of-measure accuracy, and correctly link invoice items to R365 purchase items, it could reduce both data errors and the significant amount of manual work currently being performed by restaurant managers, Accounting, and corporate personnel.

Barcode-based receiving and inventory-count capabilities should also be evaluated.

The objective is straightforward:

Reduce manual data entry wherever possible and introduce system controls where human error is predictable.

Establish Inventory Accountability

Inventory accuracy ultimately depends on disciplined execution in the restaurants.

The future-state process should clearly define responsibility for:

• Counting
• Reviewing
• Correcting
• Approving
• Investigating
• Escalating

Weekly inventory counts must be completed and formally approved.

A count left in “In Progress” should be treated as an exception rather than an acceptable status.

Count sheets should also be redesigned to minimize user confusion.

For example, units should be logically presented from smaller to larger quantities, such as:

Each → Dozen → Case

This reduces the likelihood that a manager will enter the correct physical quantity against the wrong R365 unit.

More importantly, restaurant managers should not be responsible for navigating thousands of potential items to determine what they are supposed to count.

The system should control which items appear on each brand’s count sheet.

A restaurant operating Brand A should not routinely see Brand B products simply because both brands exist in the same R365 database.

This is a critical future-state principle:

Do not rely on training alone to compensate for poor system design.

The system should make the correct process the easiest process.

The project team should document how Toast represents:

Parent Item → Modifier Group → Modifier → Portion → Recipe

and compare that structure with how R365 receives and interprets the same information.

Create Formal Reporting Controls

Once the underlying data is corrected, formal reporting governance must be established.

A standardized AvT report should be created and saved with approved settings so that Operations, Accounting, and executive management are reviewing the same information.

The organization should define:

• Report name
• Report version
• Required filters
• Sales basis
• Period selection
• Inventory assumptions
• Variance thresholds
• Review responsibility

Executive food-cost reporting should include an operational review before distribution.

No significant variance should simply be passed through to senior management because “that is what R365 says.”

The review should answer three questions:

1. What changed?

Identify the specific item, recipe, ingredient, cost, or inventory movement.

2. Why did it change?

Determine whether the cause was operational, purchasing-related, inventory-related, recipe-related, POS-related, or a system/integration issue.

3. Can the number be trusted?

If the variance cannot be explained, the report should be flagged for further investigation before being used for executive decision-making.

This creates an important control point between system-generated information and executive management.

Add Independent R365 Expertise and Oversight

The complexity of the problem makes it impractical for restaurant operators and internal staff to perform the complete remediation independently.

An experienced external R365 resource should be engaged to provide independent oversight and technical expertise.

The ideal resource should have demonstrated experience with:

• R365 Operations
• R365 Inventory
• R365 Recipes
• R365 AvT
• Purchase items
• Units of measure
• POS integrations
• Toast
• Multi-brand restaurant environments

The external resource should conduct a formal technical audit and lead the remediation under a defined scope of work.

The project should include:

• Defined milestones
• Deliverables
• Completion criteria
• Weekly status reporting
• Issue tracking
• Testing requirements
• Final validation

The resource should not simply be used as a consultant to answer individual questions.

They should provide oversight of the recovery effort from beginning to end.

This provides the management group with an independent checkpoint that the system is actually being corrected rather than simply appearing cleaner.

Each SOP should identify:

Who does it → What they do → When they do it → What system they use → Who approves it → What happens when something goes wrong

This converts the recovery effort from a one-time cleanup into a sustainable operating process.

Recommended Project Structure

The recovery should be managed as a formal cross-functional technology and Operations project with executive sponsorship and weekly governance.

Phase 1 – Stabilize

Identify immediate reporting anomalies, standardize reporting, ensure inventory approvals, correct known high-impact issues, and establish temporary executive reporting controls.

Phase 2 – Audit

Engage the external R365 expert and conduct a complete audit of the Toast/R365 data flow across both brands.

Phase 3 – Remediate

Correct:

• Recipes
• Recipe applications
• Toast mappings
• Modifiers
• Portions
• Purchase items
• Vendor items
• Units of measure
• Inventory templates
• Invoice processes
• Brand separation
• Shared-item governance

Phase 4 – Validate

Conduct controlled testing from Toast through R365.

Validation should include representative transactions from both brands and complex modifier combinations.

The team should prove that:

What was sold in Toast = What R365 understands was sold = What recipe R365 applies = What inventory R365 theoretically depletes.

Phase 5 – Operationalize

Implement SOPs, training, accountability, approval workflows, standardized reporting, and executive review.

Phase 6 – Sustain

Establish recurring quality checks to ensure the environment remains accurate as menus, vendors, products, recipes, prices, and POS configurations change.

The most important lesson from this project is that food-cost accuracy cannot be treated as an implementation deliverable that is completed once and forgotten.

A multi-brand restaurant management group requires ongoing governance.

A permanent governance model should include:

Technology

Owns platform configuration, integration health, system changes, and technical escalation.

Operations

Owns restaurant execution, inventory counting, recipe usage, variance investigation, and operational compliance.

Accounting

Owns invoice processing, costing, financial-period integrity, and reconciliation.

R365 Administrator / Business Systems

Owns day-to-day R365 configuration, controlled changes, user support, and coordination with R365.

Executive Leadership

Owns the accountability framework and ensures significant unexplained variances receive appropriate attention.

Changes to menus, recipes, products, vendors, units of measure, or POS configurations should follow a controlled change-management process.

This is particularly important in a two-brand environment where a change that is appropriate for one brand may be completely inappropriate for the other.

Conclusion: From System Implementation to Trusted Management Platform

The current food-cost challenge is recoverable, but it requires a change in mindset.

The management group should stop treating unexpected AvT results as isolated reporting anomalies and instead address the underlying Toast-to-R365 data architecture, multi-brand configuration, operational processes, and governance.

Toast remains a strong operational platform for the restaurants. Its flexibility, intuitive interface, and ease of training have created meaningful value at the restaurant level. R365 has also proven effective as the accounting and back-office platform.

The challenge is ensuring that the integration between these two platforms preserves the information required to make operational calculations reliable.

The modifier-association problem illustrates the issue particularly well. If Toast knows that a small chicken quesadilla and a large steak quesadilla were sold, but the relationship between the modifiers and their parent items is lost during the integration into R365, the theoretical food-cost calculation can no longer be trusted – even if every recipe and ingredient cost in R365 is otherwise correct.

That is why the recovery cannot focus solely on recipes.

It must address the entire ecosystem:

Toast → Sales Mix → Modifiers/Portions → R365 Recipes → Ingredients → Purchase Items → Units of Measure → Inventory → AvT

The immediate goal is to restore confidence in the numbers.

The longer-term goal is to create a controlled operating environment in which accurate food-cost information is produced consistently, exceptions are identified quickly, and restaurant managers are not burdened with unnecessary manual processes.

The recovery plan should therefore combine independent R365 expertise, Toast integration validation, recipe and item governance, operational discipline, system controls, standardized reporting, executive oversight, and documented SOPs.

Success should ultimately be measured by more than achieving a reasonable food-cost percentage.

The real measure of success is when restaurant managers, Operations, Accounting, Technology, and executive management can look at the same AvT report and confidently answer:

• What did we sell?
• What should it have cost?
• What did it actually cost?
• And if there is a variance, do we know why?

That is the point at which Toast and R365 become more than connected applications. They become a trusted technology platform for controlling food cost, improving operational performance, and protecting the profitability of both brands.