EXPO Operating Framework

Everything that runs EXPO — the methodology we sell, the playbooks we deliver, how we go to market, and how we run the business — rendered for the web. Use the menu to move between documents. The two tools (ROI Model and Readiness Scorecard) are live calculators.

The framework at a glance

EXPO treats a restaurant group as four value streams, run on a fixed operating cadence. Every improvement, AI use case, and KPI is tagged to a stream.

What's inside

• Methodology — the sellable IP: principles, operating-system model, the SAFe/Agile→ops translation, the eight wastes, the maturity model, and the metrics framework.
• Consulting Playbooks — engagement SOPs, the SOW template, and two live tools (ROI Model, Readiness Scorecard).
• Go-To-Market — ICP, outreach, discovery, proposals, packaging, objections, KPIs.
• Business Operations — roles & RACI, onboarding, internal cadence, tooling, KPIs, revenue model.
• Case Studies — three end-to-end worked examples across the maturity range.

Confidential — internal & client-engagement use. © EXPO Restaurant Agile.

1. The Thesis

Restaurant operators run some of the hardest businesses in the economy: net margins of 3–5%, prime costs (food + labor) routinely at 55–65% of revenue, hourly turnover near 96% in full-service and 135% in limited-service, and a back office still run on spreadsheets, paper invoices, and email. The tools to fix this exist. Adoption — not technology — is the missing piece.

The single most important insight behind EXPO: latency is the enemy. Almost every operational and back-office problem reduces to the same shape — data is captured manually, reconciled monthly, and the problem is discovered too late to act on. A vendor price hike, a labor overrun, a food-cost variance, an at-risk employee: each is knowable days or weeks before it shows up in the P&L.

The unfair advantage. EXPO is built by the rare overlap of a 20-year software engineer and a 20-year restaurant operator. Most restaurant technology is built by engineers who have never run a Friday night, or by operators who cannot ship software. The operator ensures we build the right thing; the engineer ensures we build the thing right. That overlap is the entire moat — and it is exactly what an operator needs to adopt AI with confidence.

Why operational change usually fails — and how EXPO is designed against it

Roughly 70% of change initiatives and over 80% of AI projects fail. In restaurants the odds are worse because of conditions the framework must respect:

Design principle: AI won’t save a broken system — it will just make the mess move faster. EXPO installs the operating system first, then the intelligence.

2. The Seven Principles of Restaurant-Agile

These principles adapt the Agile Manifesto, Lean thinking, and SAFe’s Lean-Agile mindset to the realities of a restaurant. They are the values every EXPO engagement, product decision, and operating standard is measured against.

1. The guest defines value. Value is what the guest would pay for. Every step in an operation is value-adding, necessary-but-non-value-adding, or waste. We relentlessly remove the third and shrink the second.

2. Run on a cadence. Fixed rhythms beat heroics. A predictable beat — pre-shift, shift, weekly review, period close — creates alignment, surfaces problems early, and makes improvement routine rather than reactive.

3. Shrink the latency. The faster a signal reaches the person who can act, the cheaper the fix. Optimize for time-to-signal and time-to-decision, not just for reports.

4. Make the work visible. Targets, blockers, and flow are posted where the team can see them. Transparency builds trust and removes the guesswork that hides waste.

5. Improve relentlessly, in small steps. Kaizen over big-bang. Every period ends with a structured review and one or two changes carried into the next. Small, compounding improvements beat heroic reinventions.

6. Respect the people closest to the work. The line knows what breaks. The frontline does not resist change — it resists friction. Decisions and problem-solving authority move toward the people doing the job.

7. Build in quality and standards. Quality and food safety are not inspected in at the end; they are built into every station, every shift, with a clear Definition of Done. If a procedure can’t be inspected and scored, it won’t be followed.

3. The Restaurant-Agile Operating System

The framework treats a restaurant group as a set of value streams run on a cadence by clear roles using a small set of artifacts. This is the SAFe idea — align many teams to enterprise outcomes on a fixed heartbeat — translated to above-store operations.

3.1 Value streams

We map operations into value streams so improvement targets a flow, not a department. The four core operational value streams in any restaurant:

• Guest-to-Table — order → kitchen → expo → service → payment → feedback (speed, accuracy, experience).

• Plate-to-Profit — purchasing → receiving → inventory → recipe/portion → waste → food cost.

• People-to-Performance — hire → onboard → schedule → develop → retain (labor cost, turnover).

• Cash-to-Close — sales capture → reconciliation → AP/payables → payroll → period close (back-office accuracy & speed).

Every EXPO improvement, AI use case, and KPI is tagged to one of these streams. This keeps work focused and makes ROI legible.

3.2 The operating cadence

The heartbeat of the system. Each cadence has a fixed purpose, owner, timebox, and output — the restaurant translation of Agile ceremonies.

3.3 Artifacts

• Operations Backlog — the single, ROI-ranked queue of every improvement, fix, and AI initiative. Prioritized, not driven by who shouts loudest.

• The Board — a visible work-in-progress view (prep, expo, station flow; and the above-store improvement board) with WIP limits.

• Definition of Done — the shared quality and food-safety standard a shift, station, or task must meet to be “done.”

• The Scorecard — the standardized KPI pack every unit reports on, same metrics, same format, same cadence.

• The Operating Standards Library — inspectable SOPs; the institutional memory that survives turnover.

3.4 Roles & accountabilities

Adapted from Scrum’s accountabilities (2020) and restaurant above-store structure. One accountable owner per outcome.

4. The SAFe / Agile → Restaurant Operations Translation

The disciplines that made software teams dramatically more productive map cleanly onto restaurant operations. This translation is the heart of the EXPO IP — it is what makes a proven management system legible and credible to operators.

Credibility note. EXPO uses current terminology so the framework holds up with practitioners: Scrum (2020) “accountabilities,” and SAFe 6.0 core values — Alignment, Transparency, Respect for People, Relentless Improvement. We reference these as the lineage of the method, not as jargon to impose on operators.

5. The Eight Wastes, on the Line

Lean’s eight wastes (mnemonic DOWNTIME) are a practical lens for any current-state assessment. Here they are translated to a restaurant, with the signal EXPO uses to detect each.

6. The EXPO Operating Maturity Model

A five-level model that lets an operator see where they are and what “better” looks like. The levels are deliberately tied to the real breakpoints where multi-unit groups break — so the model doubles as a scaling diagnostic.

Most operators sit between Levels 2 and 3 and stall there. EXPO engagements are scoped to move a client up exactly one level at a time — never a leap — because skipped foundations are why transformations collapse.

How we score it

Each value stream is scored 1–5 across five dimensions: Standards (are SOPs inspectable?), Cadence (does the rhythm run?), Data (is it trusted and timely?), People (is the chain-of-help in place?), and Technology/AI (are the right tools integrated?). The full instrument is the Readiness Assessment Scorecard in the Consulting Playbooks.

7. The Metrics Framework

Everything anchors to prime cost. It is the metric spine of the framework: the sum of food, beverage, and labor cost, the largest controllable block of the P&L, and the fastest lever on a 3–5% net margin.

The standard EXPO scorecard, organized by value stream, with industry reference ranges (2024–2025). Reference ranges are targets, not guarantees; each client’s baseline is captured in the Readiness Sprint.

Profitability context: in 2025, an estimated 42% of operators were unprofitable (up from 29%), with food and labor costs roughly 35–38% above 2019 levels. Small, durable cost improvements are therefore disproportionately valuable.

8. How AI Plugs Into the System

AI is the intelligence layer on top of the operating system — never a replacement for it. We deploy it on the highest-volume, most repetitive, lowest-risk work first, where data is most available and ROI rests on verified industry economics rather than vendor hype.

8.1 The adoption wedge (sequence matters)

Roughly 95% of enterprise AI pilots deliver no measurable P&L impact — almost always because of integration and fit, not model quality. EXPO beats that by design: one use case, one site, human-in-the-loop, a go/no-go gate.

Deliberately late-wave and high-caution: full dynamic pricing (consumer backlash risk) and drive-thru voice AI (publicly documented failures, biometric/BIPA exposure). EXPO treats these as later, opt-in, and never as the pilot.

8.2 The data-readiness gate

Only about 7% of enterprises say their data is fully AI-ready; the #1 obstacle is siloed data. Restaurants are worse: ~69% run multiple, non-integrated systems. No AI use case ships until its data gate is met — for example, forecasting needs 12+ months of cleanly-named POS history; food-cost AI needs item-level invoice data tied to recipes. Closing this gate is itself a core EXPO deliverable.

8.3 Governance guardrails

• Human-in-the-loop with clear escalation on anything guest- or money-facing.

• Disclose AI use to guests; never label pricing “dynamic” or “surge.”

• Vendor due diligence on biometric/voice data handling (BIPA exposure is real and active).

• A lightweight written AI policy with employee acknowledgement.

• Position AI as augmenting staff — matching the ~70% operator belief that technology should augment, not replace, labor.

9. Glossary & Sources

Key terms

• Prime cost — Food + Beverage COGS + total labor, as a % of sales. EXPO’s north-star metric.

• Value stream — the end-to-end flow that delivers value to a guest or to the P&L.

• Cadence — a fixed-rhythm operating ritual with a purpose, owner, timebox, and output.

• Definition of Done — the shared quality/food-safety standard a shift or task must meet.

• Operations Backlog — the single, ROI-ranked queue of improvements and AI initiatives.

• Theoretical vs Actual (T vs A) — expected vs real usage/cost; the core food-cost variance signal.

Selected sources

• National Restaurant Association — State of the Industry 2025; Labor & Profitability 2024 (labor 36.5% / 31.7%).

• Black Box Intelligence — State of the Workforce 2024 (turnover 96% / 135%; cost-to-replace).

• ACSI Restaurant Study 2025 (satisfaction benchmarks).

• Deloitte — How AI Is Revolutionizing Restaurants, 2025 (82% increasing AI investment; top blockers).

• MIT NANDA / RAND / BCG / McKinsey, 2024–2025 (AI pilot failure rates; data readiness).

• Scaled Agile — SAFe 6.0 core values & configurations; Scrum Guide 2020; Agile Manifesto; Lean (Toyota Production System).

Full source URLs are maintained in the EXPO research file. Figures labeled as targets are industry rules-of-thumb; figures attributed to named studies are research-grade.

1. The Engagement Arc

Every EXPO relationship follows a four-phase arc. Each phase is sold as a separately-priced, standalone module with an explicit go/no-go gate between them. Selling phases independently — and letting the client stop cleanly after any one — is deliberate risk reversal: it lowers the client’s perceived commitment and raises close rates. Most clients continue once the first phase proves value.

2. Package A — Readiness Sprint

Duration 2–3 weeks, fixed scope. Objective Map current state across a representative slice of units and rank every AI & automation opportunity by ROI.

Week-by-week

Deliverables

• Readiness Scorecard — Maturity Model scores by value stream with the binding constraints flagged.

• Prioritized Opportunity Backlog — every AI/automation opportunity, RICE-scored.

• Baseline ROI Model — populated with the client’s actual numbers (the EXPO ROI Model workbook).

• Findings & Recommendations readout — the case for the Wave-1 pilot.

Exit criteria Validated scope, confirmed data access and stakeholder availability, a working delivery rhythm, and a client decision: continue / rescope / stop cleanly.

3. Package B — Transformation Roadmap

Duration 6–8 weeks, executive engagement. Objective Convert the prioritized backlog into a target operating model and a phased, costed rollout plan with a change-management playbook.

Workstreams

• Operating-model design — the cadence, roles, artifacts, and Definition of Done tailored to the client; the move from their current maturity level to the next.

• Tooling & data plan — integration map, the data-readiness gates, build-vs-buy (buy from specialists), vendor shortlist.

• Pilot design — the Wave-1 use case(s), pilot units, SMART objectives, exit criteria, and measurement plan.

• Change-management plan — stakeholder map, communication plan, sponsor roadmap, resistance plan.

• Business case — the costed roadmap and the ROI model, with a clear options paper.

Deliverables

• Target Operating Model document (cadence, roles, artifacts, standards).

• Options paper — typically 3 options with trade-offs, costs, and a recommendation.

• Phased rollout plan — milestones, owners, dependencies, KPIs.

• Change-management playbook — comms, sponsorship, training, resistance.

• Business case & ROI model — board-ready.

Exit criteria Client approves the roadmap, scope, and price; KPIs and acceptance criteria are locked before any implementation begins.

4. Package C — Embedded On-Ramp

Duration 90-day pilot, hands-on. Objective Implement 2–3 AI use cases in pilot units, stand up the operating cadence, train the teams, then measure and tune — and prove the numbers before any portfolio-wide commitment.

Pilot exit criteria (example, set per engagement)

• ≥ 40% reduction in manual hours on the targeted back-office task.

• ≥ 90% AP invoices auto-coded with human-in-the-loop review.

• Measurable movement on the targeted prime-cost lever vs baseline.

• Cadence adopted: huddles, flash, and weekly review running without EXPO present.

Then: a Fractional Chief AI & Ops Officer partnership sustains momentum and governs the rollout across the portfolio.

5. Discovery & Assessment Methods

5.1 Stakeholder interviews

Run early; use semi-structured 45-minute one-on-ones (a guide, not a script). Stay neutral, listen actively. Structure each guide around four areas: success metrics, priorities, history/expertise, and process/workflow. Always close with “Who else should I speak to?” to surface stakeholders and earn warm intros. Loop a synthesis summary back to confirm understanding.

5.2 Current-state assessment

Start with value-stream mapping (strategic, end-to-end) then process mapping (specific bottlenecks). Classify each step as value-adding, necessary-non-value-adding, or waste, using the eight wastes. Plot the org on the EXPO Maturity Model. Use real measurements, not estimates. The current-state vs target-state gap is the roadmap.

5.3 Opportunity prioritization — a tiered toolkit

EXPO standard: triage with Effort/Impact, then RICE-score the top contenders for the backlog. RICE Impact uses a fixed scale (Massive 3 / High 2 / Medium 1 / Low 0.5 / Minimal 0.25); Confidence (High 100% / Medium 80% / Low 50%); Effort in person-weeks.

5.4 Readiness dimensions

Pair the EXPO Maturity Model with a transformation-readiness lens. Data infrastructure & quality is almost always the binding constraint — score it honestly. External benchmarking is the step most often skipped; without it, scores are self-referential. The full instrument is the Readiness Assessment Scorecard workbook.

6. Change Management

Operating-model change succeeds or fails on adoption, not design. EXPO runs Kotter at the program level and ADKAR at the individual level.

• Preferred senders: executives deliver the “why”; immediate managers deliver the personal-impact “WIIFM.” Never let the project team be the only voice.

• Communicate key messages 5–7 times; start sooner than feels necessary; favor two-way channels.

• More than half of resistance is avoidable — prevent it with early change management, not reaction. Mid-level managers are typically the most resistant group.

• Active, visible sponsorship is the single biggest contributor to change success. Equip sponsors; don’t substitute for them.

7. Delivery Standards (Definition of Done)

EXPO eats its own cooking: our deliverables run on a cadence and a Definition of Done.

• The rule of 3 — every workstream defines three outcomes, three primary deliverables, and three KPIs. It keeps acceptance enforceable.

• Every deliverable has SMART, testable acceptance criteria agreed before work starts, and a named approver.

• Senior partner sign-off before any deliverable ships to a client; peer review on models and ROI math (zero formula errors).

• One issue log, weekly cadence, and a defined escalation path per engagement.

• Reusable assets (decks, models, templates) are captured into the EXPO knowledge library after every engagement, version-controlled.

EXPO · RESTAURANT AGILE

TEMPLATE

Statement of Work

Reusable EXPO engagement contract template. Replace every [bracketed] field. Delete guidance notes (in italics) before sending.

Document SOW Template

Owner EXPO — Restaurant Agile

Version 1.0

Use Duplicate per engagement

CONFIDENTIAL — Internal & client-engagement use. © EXPO Restaurant Agile.

This Statement of Work (“SOW”) is entered into as of [Date] by and between EXPO Restaurant Agile (“EXPO”) and [Client Legal Name] (“Client”), and is governed by the Master Services Agreement between the parties.

1. Background & Objectives

Client operates [#] restaurants across [concepts/regions]. EXPO will deliver the [Readiness Sprint / Transformation Roadmap / Embedded On-Ramp] engagement. The measurable business outcomes of this SOW are:

• [Outcome 1 — measurable]

• [Outcome 2 — measurable]

• [Outcome 3 — measurable]

Guidance: state 1–3 outcome-based results, not tasks. Outcomes drive acceptance.

2. Scope of Work (In Scope)

EXPO will perform the following workstreams:

• [Workstream 1]

• [Workstream 2]

• [Workstream 3]

3. Out of Scope

The following are expressly excluded and, if required, will be handled via a change order:

• [Excluded item 1]

• [Excluded item 2 — e.g., software development, hardware, third-party fees]

Guidance: an explicit out-of-scope list is the single most effective anti-scope-creep mechanism.

4. Assumptions & Client Dependencies

This SOW assumes the following Client-provided inputs. Delays may affect timeline and fees:

• Read access to POS, payroll, and accounting data within [X] business days of kickoff

• A named executive sponsor and a single point of contact

• Stakeholder availability for interviews and the weekly cadence

• Timely approvals within the defined review windows

5. Deliverables & Acceptance Criteria

Each deliverable is deemed accepted upon written sign-off by [Approver], or after [5] business days without written objection.

6. Key Performance Indicators

Success is measured against the following KPIs, each with a definition, baseline, and target captured at kickoff:

• [KPI 1 — definition / baseline / target / source]

• [KPI 2 …]

• [KPI 3 …]

7. Timeline & Milestones

Engagement runs [start] to [end], on the following milestones:

• [Milestone 1 — date]

• [Milestone 2 — date]

• [Go/no-go gate — date]

8. Governance

A weekly cadence call; a single shared issue log; decision rights per the RACI below; escalation to the executive sponsor within [24] hours of an unresolved blocker.

• EXPO lead: [name]

• Client sponsor: [name]

• Client point of contact: [name]

9. Change Control

Any change to scope, schedule, or fees is requested in writing, assessed by both parties for impact, and, if approved, documented as a signed change order before work proceeds.

10. Commercial Terms

11. Confidentiality & Data Handling

Each party protects the other’s confidential information; obligations survive [7] years. Client data is handled per the MSA and applicable privacy law; biometric/voice data, if any, is handled under a separate addendum.

12. Intellectual Property

Client owns the work product created specifically for Client. EXPO retains ownership of its pre-existing and background IP and methodology (the Restaurant-Agile Framework, templates, models, and tools), and grants Client a non-exclusive license to use them for its internal operations.

13. Term, Termination & Handover

Either party may terminate for convenience on [15] days’ written notice. On termination, EXPO delivers work-in-progress, returns or destroys Client materials as directed, and provides a clean handover. Fees for work performed through the termination date are due.

14. Acceptance

Agreed and accepted by the authorized representatives of the parties:

ROI Model — Interactive

Edit the inputs; results recalculate instantly. Mirrors the Excel tool — no spreadsheet required. Illustrative; replace defaults with a client's actuals during the Readiness Sprint.

Readiness Scorecard — Interactive

Score each value stream 1–5 across the five dimensions. The overall score and maturity level update live. Mirrors the Excel tool.

Scale: 1 Heroic · 2 Standardized · 3 Measured · 4 Connected · 5 Self-improving. Dimensions — Standards (inspectable SOPs) · Cadence (does the rhythm run?) · Data (trusted & timely?) · People (chain-of-help?) · Technology/AI (right tools integrated?).

1. Ideal Client Profile (ICP)

EXPO wins by going narrow. A tight, single-segment ICP produces higher win rates than chasing volume. The ICP defines which accounts we pursue; buyer personas (below) drive the message.

Firmographic

• Multi-unit restaurant operators and franchisees, ~5–150 units (sweet spot 8–60 — past the breakpoints where consistency, visibility, and reporting break).

• Full-service, fast-casual, and franchise QSR; single- or multi-concept groups.

• Regionally concentrated, family- or partner-owned, professionalizing their back office.

• $15M–$750M revenue; profitable but margin-squeezed; actively investing in technology.

Intent / trigger signals

• Recent acquisition or unit growth (integration pain), new CFO/COO/Director of Ops, or a stated efficiency/AI initiative.

• Multiple disconnected systems (POS + scheduling + accounting not integrated).

• Hiring above-store leadership or back-office roles; public commentary on labor/margin pressure.

2. Buyer Personas & Multithreading

B2B buying groups average ~11 stakeholders. Identify and message each; never email two contacts at one account simultaneously (they backchannel and it can flag spam).

3. Outreach

3.1 Deliverability (non-negotiable)

• Authenticate every sending domain: SPF + DKIM + DMARC with alignment; one-click unsubscribe.

• Use separate sending domains/inboxes; warm new domains at 5–10/day ramping over 4–6 weeks.

• Keep spam complaints < 0.10% and bounces < 2%; verify lists; consistent volume, no spikes.

• Optimize on reply rate and positive-reply rate — not opens (open tracking is unreliable).

3.2 Sequence structure

4–7 touchpoints over ~2–3 weeks (a 0 / 3 / 7 / 14-day pattern captures the majority of replies), mixing email + LinkedIn + a call. Email under 80 words; problem-first; one research-based personalized line; a single low-friction CTA (a binary question). Benchmark: average cold-email reply ~3.4%, good 5.5%+, elite 10%+. Step 1 drives ~58% of replies; follow-ups the rest — so always run the full sequence.

3.3 Sample templates

4. Discovery

Run SPIN to drive the conversation and MEDDIC to qualify the pipeline. The highest-leverage move is Implication questions — surfacing the cost of the status quo.

SPIN question bank

• Situation — How many units? What POS, scheduling, and accounting systems? Who owns above-store ops?

• Problem — Where does the month-end close break down? How current is your prime cost when you see it? How are invoices and schedules built today?

• Implication — What does two-week-late variance cost you across [#] units? What does turnover at ~96% cost in re-training? What happens to consistency as you add units?

• Need-payoff — If you saw prime cost weekly and AP was 90% automated, what would that be worth? What would one point of margin across the group fund?

MEDDIC qualification checklist

5. Proposals

A proposal is a summation of prior conversations, not a selling document. Structure (after Alan Weiss): situation appraisal; outcome-based objectives; measures of success; value (so the fee feels small); methodology with three escalating options; timing; joint accountabilities; terms (fees appear here, after the value); acceptance.

• Always offer three options (good-better-best) — a “choice of yeses,” not a yes/no. Option 2 = Option 1 + one item, ~20–30% higher.

• Lead with an executive summary that proves you understood the problem; frame the problem before pricing.

• Present the proposal live — it drives materially higher win rates than emailing pricing. Target a 60–80% win rate.

6. Packaging & Pricing Logic

EXPO sells productized services: standardized scope, a fixed price (a number, not a range), and crystal-clear inclusions/exclusions. This lets the business scale beyond the founders and protects margin.

Consulting ladder (land → expand)

SaaS tiers (per location / month)

• Line · Foundation — AI back-office essentials.

• Service · Growth — optimization + agile ops workflows (the recommended tier for mid-size multi-unit).

• Command · Franchise — multi-concept command center, volume terms for 100+ units.

Pricing principles

• Value-based, not hourly — anchor the fee to the client’s upside (points of prime cost, turnover saved), established with the economic buyer before quoting.

• Negotiate scope, not rate — a 10% rate cut erases ~33% of profit; remove deliverables or phase the work instead.

• Land for comfort, expand for revenue — most lifetime value is in the retainer + SaaS recurring, not the first project.

7. Objection Handling

8. Sales KPIs

1. Operating Model & Roles

EXPO is a boutique consulting + SaaS firm run by two partners — a software engineer and a restaurant operator — who both carry business-development and delivery responsibility. The model scales through productized services and the platform, not through headcount. We keep overhead minimal and add roles only when utilization or pipeline demands it.

Engagement RACI

R = Responsible · A = Accountable · C = Consulted · I = Informed. One Accountable per row.

2. Client Onboarding

A clean, fast onboarding sets the cadence for the whole engagement. Target: kickoff within 5 business days of signature.

Onboarding checklist

1. Countersigned SOW + deposit invoice issued (30–50% on signature).

2. Welcome email: named EXPO lead, sponsor, and single point of contact confirmed.

3. Data-access request sent: read access to POS, payroll, accounting; secure transfer method.

4. Kickoff call scheduled; stakeholder interview list and calendar holds set.

5. Shared workspace created (issue log, document folder, KPI scorecard).

6. Baseline capture: pull the client’s real numbers into the ROI Model and Scorecard.

7. Cadence set: weekly status call, escalation path, review windows agreed in writing.

3. Internal Operating Cadence

EXPO runs its own agile cadence — we practice the methodology we sell.

4. Delivery Operations

Utilization & capacity

• Target billable utilization 70–80% for delivery staff; partners run lower (60–75%) to leave room for BD.

• Forecast utilization a quarter ahead; protect a backlog of 3–6 months of contracted work.

• Cap concurrent engagements per consultant to protect delivery quality.

Quality control

• Quality gates with predefined pass/fail criteria at each engagement stage.

• Senior partner sign-off before any deliverable ships; peer review on ROI models (zero formula errors).

• Standardized templates and the brand style applied to every external artifact.

Knowledge management

• After every engagement, capture decks, models, and reusable assets into the EXPO knowledge library, version-controlled and tagged by package/concept/issue.

• Treat reusable IP as a core asset — it is what lets a non-founder deliver and what compounds margin over time.

5. Tooling Stack

6. Business KPIs

The metrics EXPO manages to. Targets are directional benchmarks for a healthy boutique professional-services + SaaS business.

7. Revenue Model

EXPO compounds three revenue streams. The strategic goal is to shift the mix toward recurring (retainer + SaaS) over time, which raises both stability and enterprise value.

• Project consulting — Readiness Sprint, Transformation Roadmap, Embedded On-Ramp (fixed-fee, milestone-billed).

• Recurring advisory — Fractional Chief AI & Ops Officer retainer.

• SaaS platform — per-location/month subscriptions across three tiers.

Executive Summary

Marlowe runs two casual-dining concepts and a wine bar across 14 units. Strong hospitality, but the back office was drowning: invoices keyed by hand, food cost discovered three weeks late on a monthly P&L, and no common scorecard across units. EXPO landed on AP/invoice automation and a food-cost variance engine, installed a weekly prime-cost cadence, and stood up a portfolio dashboard.

1. Client Snapshot

The core problem. Food cost had crept to ~33.5% and nobody saw it until the monthly close — three weeks too late to react. Managers spent nights keying invoices instead of running the floor, and there was no consistent scorecard across the 14 units.

2. How We Landed Them (GTM)

Trigger signal. A newly hired Director of Operations posted about “getting our arms around food cost across all units” — a classic above-store consistency signal. EXPO referenced their multi-concept structure in the opener.

The cold email that opened the conversation:

Discovery call — SPIN highlights

• Situation — Toast at unit level, QuickBooks for accounting, no integration; AP keyed by GMs.

• Problem — Food cost only visible monthly; vendor price hikes undetected; no common scorecard.

• Implication — At ~$42M, even one point of food cost is ~$420K leaking annually, found too late to fix.

• Need-payoff — Weekly prime cost + automated AP would free manager nights and stop the leak at the source.

MEDDIC qualification

3. Readiness Sprint (2–3 weeks)

3.1 Current-state findings

• AP fully manual: ~1,100 invoices/month keyed by hand; no line-item price tracking.

• Food cost reviewed monthly; theoretical-vs-actual not tracked.

• No common scorecard; each GM reported differently.

• Pre-shift huddles inconsistent; line checks not logged.

• Strong culture and tenured GMs — high adoption potential once friction is removed.

3.2 Maturity scorecard (filled)

Scores 1–5 by value stream × dimension. Overall: 1.8 -> Level 2 — Standardized. (Live workbook: the filled Readiness Scorecard in this folder.)

3.3 Prioritized opportunity backlog (RICE)

3.4 ROI model (filled)

Built with the client's actual numbers (live workbook: the filled ROI Model in this folder). Conservative base case:

4. Statement of Work (key terms, filled)

Full contract built from the EXPO SOW Template; fees shown are illustrative for this worked example.

5. Transformation Roadmap

Chosen option. Option 2 (recommended): back-office automation + the weekly prime-cost operating cadence, piloted in 4 units before portfolio rollout.

Target operating-model moves

• Install the weekly prime-cost review and standardized scorecard across pilot units.

• Make pre-shift huddles and logged line checks the Definition of Done for every shift.

• Integrate Toast + QuickBooks; AP invoices captured and auto-coded.

• Stand up the food-cost variance engine (theoretical vs actual).

Phased plan

6. 90-Day Embedded On-Ramp (Pilot)

Pilot scope. AP/invoice automation and the food-cost variance engine in 4 units, plus the weekly prime-cost cadence and portfolio scorecard. Human-in-the-loop on all invoice coding.

Exit criteria

• ≥ 90% of invoices auto-coded with review.

• ≥ 40% reduction in manager hours on AP.

• Food-cost variance visible weekly in all 4 pilot units.

• Weekly prime-cost review running without EXPO present.

Run log (highlights)

7. Results

Payback. Modeled annual impact of $1,166,172 against ~$131K of illustrative first-year consulting fees implies roughly a 9× first-year return — before the platform’s ongoing contribution.

8. Expansion & Why It Worked

The expand. Marlowe rolled the platform to all 14 units on the Service · Growth tier and retained EXPO as Fractional Chief AI & Ops Officer to run the portfolio cadence and sequence Wave-2 (forecasting + scheduling).

Why it worked (framework principles in action)

• Latency attacked first — weekly prime cost replaced the monthly blind spot.

• Landed on the lowest-risk, highest-volume work (AP) and proved it before expanding.

• Standards lived in the system and cadence, not in any one GM’s head.

• Respected tenured operators by removing friction, not adding a program.

Executive Summary

Sunbelt is a professionalized 62-unit QSR franchisee with solid reporting but a labor problem: labor running hot, hourly turnover near 130%, and District Managers building schedules by gut. EXPO landed on AI demand-matched scheduling and a retention/attrition signal, integrated into the existing stack, with a district-level accountability cadence.

1. Client Snapshot

The core problem. Labor was the margin killer: ~33% of sales, hourly turnover near 130%, and schedules built on intuition. Drive-thru speed varied widely by district with no shared accountability rhythm.

2. How We Landed Them (GTM)

Trigger signal. A job posting for multiple District Manager roles plus a new VP of Operations signaled scaling pain and above-store reorganization — a strong intent signal for an operating-cadence + labor engagement.

The cold email that opened the conversation:

Discovery call — SPIN highlights

• Situation — R365 in place, NCR POS, a scheduling tool nobody fully uses; 7 districts.

• Problem — Labor ~33%, turnover ~130%, schedules by gut, drive-thru speed inconsistent.

• Implication — Turnover at this rate costs well over $500K/yr in replacement alone; labor points are millions.

• Need-payoff — AI scheduling + retention signal + a district cadence would move labor and stabilize crews.

MEDDIC qualification

3. Readiness Sprint (2–3 weeks)

3.1 Current-state findings

• Labor at ~33% with wide unit variance; OT poorly controlled.

• Turnover ~130%; no early-warning on at-risk crew.

• Scheduling tool present but schedules still built manually by DMs.

• Good data foundation (R365) — ready for AI; the gap is People and Tech utilization.

• No shared district accountability cadence; speed varied 40+ seconds across districts.

3.2 Maturity scorecard (filled)

Scores 1–5 by value stream × dimension. Overall: 2.9 -> Level 3 — Measured. (Live workbook: the filled Readiness Scorecard in this folder.)

3.3 Prioritized opportunity backlog (RICE)

3.4 ROI model (filled)

Built with the client's actual numbers (live workbook: the filled ROI Model in this folder). Conservative base case:

4. Statement of Work (key terms, filled)

Full contract built from the EXPO SOW Template; fees shown are illustrative for this worked example.

5. Transformation Roadmap

Chosen option. Option 3 (recommended): AI scheduling + retention layer on top of R365, plus a district accountability cadence — piloted in 2 districts.

Target operating-model moves

• Deploy AI demand-matched scheduling with OT/compliance guardrails.

• Stand up the retention/attrition signal with a manager intervention playbook.

• Install a weekly district business review (the above-store cadence).

• Make schedule adherence and speed part of the district scorecard.

Phased plan

6. 90-Day Embedded On-Ramp (Pilot)

Pilot scope. AI demand-matched scheduling and the retention signal across 8 units in 2 districts, with the weekly district business review and OT guardrails. Human-in-the-loop on schedule approval.

Exit criteria

• ≥ 1.5 pts labor reduction in pilot units vs baseline.

• At-risk crew flagged weekly with manager playbook in use.

• District business review running without EXPO present.

• No degradation in drive-thru speed or guest scores.

Run log (highlights)

7. Results

Payback. Modeled annual impact of $2,987,222 against ~$165K of illustrative first-year consulting fees implies roughly an 18× first-year return — with turnover and labor being the dominant levers.

8. Expansion & Why It Worked

The expand. Sunbelt rolled out to all 7 districts on the Command · Franchise tier and retained EXPO on a fractional basis to govern the rollout and sequence predictive P&L and speed analytics.

Why it worked (framework principles in action)

• Targeted the biggest dollar levers verified by data (labor + turnover), not hype use cases.

• Leveraged an existing clean data foundation (R365) so AI could land fast.

• Added a district chain-of-help cadence so the gains stuck above store.

• Piloted in 2 districts with explicit exit criteria before a 62-unit commitment.

Executive Summary

Verde is a beloved 6-unit fast-casual concept whose founder still is the operating system — nothing documented, quality traveling with his presence. Growth was diluting consistency. EXPO’s job here was foundational: install inspectable standards and a basic operating cadence, with light AP automation as the only Wave-1 AI. The win was a repeatable operation that survives the founder not being in the building.

1. Client Snapshot

The core problem. The founder was the system. No documented SOPs, quality varied by which unit he visited that week, and onboarding new managers took months. The 4th, 5th, and 6th units exposed the cracks: checklist drift, inconsistent ops, and a founder stretched thin.

2. How We Landed Them (GTM)

Trigger signal. An inbound referral — the founder told a peer he was “drowning trying to be everywhere.” The opener led with the 2nd-location-test framing and the staged breakpoints at 3–7 units.

The cold email that opened the conversation:

Discovery call — SPIN highlights

• Situation — Square POS, Gusto, spreadsheets; founder + a part-time bookkeeper.

• Problem — No SOPs; quality varies by unit; manager onboarding is slow; founder overloaded.

• Implication — Every new unit dilutes the concept and pulls the founder thinner; growth stalls.

• Need-payoff — Inspectable standards + a daily cadence would let Verde grow without the founder in every store.

MEDDIC qualification

3. Readiness Sprint (2–3 weeks)

3.1 Current-state findings

• No documented SOPs; standards live in the founder’s head.

• Pre-shift and line checks done inconsistently and never logged.

• AP and reconciliation fully manual; the founder approves every invoice.

• No scorecard; the founder ‘knows’ how units are doing by visiting.

• Genuine culture and loyalty — the foundation is people-strong, system-weak.

3.2 Maturity scorecard (filled)

Scores 1–5 by value stream × dimension. Overall: 1.3 -> Level 1 — Heroic. (Live workbook: the filled Readiness Scorecard in this folder.)

3.3 Prioritized opportunity backlog (RICE)

3.4 ROI model (filled)

Built with the client's actual numbers (live workbook: the filled ROI Model in this folder). Conservative base case:

4. Statement of Work (key terms, filled)

Full contract built from the EXPO SOW Template; fees shown are illustrative for this worked example.

5. Transformation Roadmap

Chosen option. Option 1 (recommended): foundations first — standards + daily cadence + light AP automation. No heavy AI until Level 2 is solid.

Target operating-model moves

• Build the operating-standards library: inspectable SOPs for open, close, prep, and line check.

• Install the daily cadence: pre-shift huddle + logged line check as the Definition of Done.

• Automate AP so the founder stops approving every invoice by hand.

• Stand up a one-page unit scorecard so performance is visible without a store visit.

Phased plan

6. 90-Day Embedded On-Ramp (Pilot)

Pilot scope. Operating-standards library, the daily huddle + logged line check, light AP automation, and a one-page scorecard in 2 units — designed to prove the concept runs without the founder in the building.

Exit criteria

• Inspectable SOPs in use; line checks logged every shift.

• Daily huddle running without the founder present.

• AP invoices automated; founder approves by exception only.

• Manager onboarding time measurably reduced.

Run log (highlights)

7. Results

Payback. Modeled annual impact of $252,426 against ~$64K of illustrative first-year consulting fees implies roughly a 4× first-year return — but the real prize is a concept that can finally scale without diluting quality.

8. Expansion & Why It Worked

The expand. Verde rolled standards to all 6 units on the Line · Foundation tier and kept EXPO on a light retainer to certify managers and prepare the operation for units 7–10.

Why it worked (framework principles in action)

• Met the operator where they were — foundations (standards + cadence) before any heavy AI.

• Moved exactly one maturity level; skipped foundations are why scaling fails.

• Made standards inspectable so they survive the founder’s absence and turnover.

• Respected a people-strong culture by reducing friction, not importing complexity.