The CFO's expensive problem
Multi-unit operators run five or six core systems that don't talk to each other. The POS lives in one place. Payroll in another. Accounting in a third. Scheduling, inventory, and labor compliance each have their own homes. Every week the finance team rebuilds the same picture by hand.
Fragmented systems, manual reassembly
Five to seven systems that don't share data. Exports become spreadsheets. Spreadsheets become reports. Reports go stale before they're read.
10 to 15 day close cycles
Peers at the same scale close in five days. The gap is data plumbing, not effort.
Decisions made on gut feel
The data exists. It just isn't reachable in the moment a decision needs to be made.
SaaS reporting tools rent your data back to you
Aggregation vendors pull your operational data into their system. You pay forever to access reports built on top of it. Cancel and you lose everything.
Your data, in your cloud, from day one
We build a unified data foundation that lives in your cloud account, on your subscription, under your IAM. Microsoft Fabric if your team is on Microsoft. Google BigQuery if your team is on Google. Either way, the warehouse, the dashboards, and the code all belong to you.
You can fire us tomorrow and keep everything.
Connect your systems
POS, payroll, accounting, scheduling, inventory. API-based where available, automated extracts where not.
Land in your warehouse
Modeled, documented, reconciled. Microsoft Fabric or BigQuery. Your subscription, your billing, your access controls.
Reach for the answer
A weekly Excel report your CFO can pivot. A daily flash your operators can act on. A Power BI dashboard that loads in seconds.
First working deliverable in weeks, not quarters.
Microsoft Fabric or Google BigQuery. Your subscription, your billing, your IAM.
A weekly Excel report your CFO can pivot. Not a vendor-locked dashboard.
Cancel anytime. Keep all the data and all the code.
What you actually get
A weekly P&L by location your CFO can open in Excel, pivot, drill, and send to the operator team. Variance flags surface where the operator should look first. Numbers below are illustrative.
| Location | Net sales | Labor % | Food % | Prime cost | vs. target |
|---|---|---|---|---|---|
| Location 042 | $42,180 | 24.1% | 28.8% | 52.9% | ▼ 1.4 pts |
| Location 087 | $38,940 | 26.7% | 29.2% | 55.9% | ▲ 0.8 pts |
| Location 113 | $51,202 | 23.5% | 28.1% | 51.6% | ▼ 2.1 pts |
| Location 156 | $33,615 | 28.4% | 30.1% | 58.5% | ▲ 3.4 pts |
| Group average | $44,107 | 24.8% | 28.7% | 53.5% | ▼ 0.9 pts |
Who you're working with
Three decades in financial operations and controllership before this. We've closed the books, lived through the close grind, run FP&A and reporting inside multi-unit operators, and partnered with CFOs to surface the numbers they actually needed. We know what it feels like when the data exists but you can't reach it on a Tuesday morning.
The same lived experience now shapes the architecture. We build the reports we wish we'd had, in the form a finance team actually uses, on infrastructure a CFO would sign off on. AI-leveraged delivery is what compresses the timeline. Restaurant-finance and multi-unit operations is what shapes the output.
How we shape the engagement
Every group is different. Some operators want a foundation first and decide on advanced capabilities later. Others want the full picture from day one. We size each engagement to your scope, your systems, and your timeline.
Foundation
A unified warehouse with the reports your finance and ops teams need first.
- •Source-system integrations (POS, payroll, accounting, scheduling, inventory)
- •Weekly P&L by location with prime-cost waterfall
- •Daily operational flash report
- •Theoretical vs. actual food cost, labor analytics, vendor and AP analytics
- •Reconciled data model your team can extend or query directly
- •Documentation, runbooks, and a knowledge transfer to your team
What changes for your team:
- Monday-morning P&L instead of a mid-month spreadsheet rebuild.
- Operators look at a dashboard, not at email attachments.
- Variance flags surface before close, not during it.
- Finance spends time on analysis, not on data prep.
- GMs see the same numbers as the home office, in the same place.
- Same-day store-level visibility, not week-old PDFs.
Advanced capabilities
Layered on top of a working foundation, where the data quality earns the right to predict.
- •Predictive sales and labor forecasting
- •Anomaly detection (shrink, variance, compliance flags)
- •Natural-language Q&A on your operational data
- •Auto-generated executive summaries and GM coaching outputs
- •Scenario modeling and what-if analysis on labor, pricing, and menu mix
- •Cross-location pattern matching to identify what's working at top-quartile stores
Questions your team could ask:
- "Which locations are tracking under labor target this week?"
- "What's driving the food-cost spike at Location 156?"
- "Show me the top five days of comp-store sales in the last 90 days, and what they had in common."
- "Flag any shift with a labor compliance risk this week."
Ongoing support, enhancements, and change requests are available after launch, sized per engagement. No minimum commitment.
Investment depends on scope, systems, and timeline. We size it together in the discovery call.
How it compares
You have other options. Here is how this approach differs.
| Approach | Data ownership | Time to first value | Typical model |
|---|---|---|---|
| Big 4 consultancy | You own it, but layered overhead | 12 to 18 months | Large fixed engagement, multiple staff |
| Boutique data shop | You own it | 4 to 8 months | Mid-size fixed engagement |
| Restaurant analytics SaaS | Vendor owns it. Cancel and you lose access. | Days to weeks (off the shelf) | Annual subscription, perpetual lock-in |
| Internal BI hire | You own it | 12 months ramp before output | Fully loaded salary, recruiting risk |
| DIY in-house | You own it | 12 to 18 months, high failure rate | Existing team plus contractor time |
| Forge RPA Recommended | You own it. Code and data in your cloud from day one. | Weeks for first deliverable | Solo and AI-leveraged. Hourly support optional after launch. |
Big 4 consultancy
- Data ownership
- You own it, but layered overhead
- Time to first value
- 12 to 18 months
- Typical model
- Large fixed engagement, multiple staff
Boutique data shop
- Data ownership
- You own it
- Time to first value
- 4 to 8 months
- Typical model
- Mid-size fixed engagement
Restaurant analytics SaaS
- Data ownership
- Vendor owns it. Cancel and you lose access.
- Time to first value
- Days to weeks (off the shelf)
- Typical model
- Annual subscription, perpetual lock-in
Internal BI hire
- Data ownership
- You own it
- Time to first value
- 12 months ramp before output
- Typical model
- Fully loaded salary, recruiting risk
DIY in-house
- Data ownership
- You own it
- Time to first value
- 12 to 18 months, high failure rate
- Typical model
- Existing team plus contractor time
Forge RPA Recommended
- Data ownership
- You own it. Code and data in your cloud from day one.
- Time to first value
- Weeks for first deliverable
- Typical model
- Solo and AI-leveraged. Hourly support optional after launch.
Who this is for
Multi-unit operators with five to several hundred locations and the operational complexity that comes with running across systems and brands.
Restaurant franchisee groups
Taco Bell, McDonald's, Wendy's, Burger King, KFC, Pizza Hut, Dunkin', and multi-brand operators.
Multi-unit fitness operators
Planet Fitness, Anytime Fitness, F45, Orangetheory, and similar franchise gym groups.
Home services franchisees
Servpro, Mr. Rooter, Maid Brigade, and other multi-territory service brands.
Dental support organizations
Multi-practice DSO groups consolidating clinical, billing, and operational reporting.
Hospitality groups
Small and mid-size hotel franchisee groups with property management, revenue, and labor data spread across systems.
Auto service multi-units
Midas, Jiffy Lube, Valvoline, and similar multi-location automotive operators.
The value at your scale
Below are the value drivers most multi-unit operators see compound once unified reporting is in place. At your scale, even one or two of these compounds quickly.
- → Labor efficiency. Visibility into labor-to-sales by daypart, by location, with variance flags before the shift ends.
- → Food-cost reduction. Theoretical-vs-actual surfaced weekly, with hotspot identification at the SKU and location level.
- → Shrink and vendor price-creep. Patterns caught early, before they compound across the year.
- → Labor-compliance protection. Early-warning flags that prevent the violations that hit audits and lawsuits.
- → A faster close cycle. Days instead of weeks, freeing finance to advise the business instead of reformatting spreadsheets.
What a single percentage point can be worth
Illustrative annual impact of a one-percentage-point improvement on labor or food cost, against an average unit volume of $2 million per location. Your actuals depend on your starting baseline, your mix, and your execution.
| Locations | 1 pt on labor | 1 pt on food cost | Combined |
|---|---|---|---|
| 30 | $600K / year | $600K / year | $1.2M / year |
| 80 | $1.6M / year | $1.6M / year | $3.2M / year |
| 150 | $3.0M / year | $3.0M / year | $6.0M / year |
These are conservative one-point examples. Most operators see multi-point swings once the variance flags become a weekly habit. Shrink, compliance, and close-cycle savings compound on top of these numbers.
We work the math together in the discovery call, with your actual revenue per location, your current baseline, and the priorities your team would tackle first.
Common questions
How is this different from a SaaS reporting tool? +
SaaS reporting vendors pull your data into their system and rent it back to you. We build the warehouse in your cloud account, with your billing and IAM. The schemas, the models, and the dashboards all belong to you. If you ever decide to stop working with us, you keep everything.
Do I have to be on Microsoft? +
No. We pick the stack that matches what your team already uses. If you live in Microsoft 365 and Power BI, we build on Microsoft Fabric. If your team is on Google Workspace, we build on Google BigQuery. Either way, the architecture and the outcomes are the same.
How fast can you really build this? +
The first working deliverable usually lands within a few weeks. The full foundation comes online over the following months. We use an AI-leveraged delivery approach that compresses what used to be a year of work into a fraction of the time. The savings show up on your time-to-value, not on your invoice.
What if I want to cancel partway through? +
You can stop at any natural milestone. You keep all the data, all the code, and the running infrastructure in your cloud. There is no perpetual lock-in. Specific notice terms are written into each engagement and discussed in the discovery call.
Who owns the data? +
You do. 100 percent. From day one. Every byte lives in your cloud account, on your subscription, under your access controls. The code lives in your GitHub. There is no shared environment, no vendor-side copy, no extraction layer that holds your data hostage.
You may also be considering
These services often pair with a data-lake build, or stand on their own depending on what your team needs most right now. Many engagements include one or more of them.
Automation Assessment
Data-driven process scoring and ROI projections. Works before, alongside, or after a data-lake build to identify the highest-ROI automations across finance and ops.
Learn more →ERP Data Migration
Clean, reconciled data conversion across Workday, SAP, NetSuite, Ariba, and other systems. A natural complement when source systems are also being refreshed.
Learn more →Process Mining
Discover how your processes actually flow through event data. A complementary lens on where the largest improvement opportunities live across operations.
Learn more →Let's size it together
Book a 30-minute discovery call. We will walk through your systems, your reporting pain, and what a foundation could look like in your environment. You leave with a clear picture even if we never work together.