A multi-state, multi-brand furniture chain had grown via acquisitions and was operating four different POS stacks across 11 showrooms and two distribution centers. The result:

• Checkout drag: End-to-end “commit to pay → receipt printed” routinely ran 40-50 minutes at peak (financing, protection plans, delivery windows, tax/shipping, promo logic, signatures).
• Dual entry + swivel-chairing: Associates re-keyed customer data across financing portals, delivery schedulers, and inventory allocations—22+ manual fields per order on average.
• Promo + pricing inconsistencies: Store A honored a promo Store B couldn’t see; price overrides and manager approvals spiked.
• Inventory blind spots: Siloed databases created oversells and costly substitutions when DC availability lagged store promises.
• Training friction: Transfers between acquired banners meant learning a new POS each time; onboarding was slow, confidence was low.
• Leadership visibility: No single source of truth on attach rates, financing approvals, or SKU velocity across banners.

Mandate to RBA Global: Standardize on one enterprise-grade POS, cut checkout to a 12–15 minute band, and give leadership real-time control—without choking the floor during rollout.

We executed a non-disruptive, phased unification across tech, data, workflow, and people. No big-bang rip-and-replace; we sequenced risk.

A. Diagnostic & Design (4 weeks)

• Time-and-motion studies on 180 live checkouts; mapped step count, handoffs, and re-entry by order type.
• Built a cross-banner promo rules catalog to codify exceptions before platforming.
• Security review to set a clean baseline.

B. Platform & Integration Architecture (6 weeks)

• Selected a single enterprise POS with native APIs for financing, delivery, and protection plan catalogs.
• Implemented one-screen finance to pre-fill customer data, run soft pulls, and return approvals without leaving the POS.
• Delivery module embedded with DC calendars, truck capacity rules, and zip-level lead-time logic.

C. Data Spine & Migration (8 weeks, concurrent)

• Consolidated to a cloud data layer:
• 176k customer records deduped and normalized.
• 19k active SKUs harmonized with full price/discount history.
• Implemented near-real-time inventory sync across store/DC/e-com.

D. Workflow Re-Engineering (3 weeks)

• Reduced checkout from 8–10 steps to 4–5, depending on order type: cash/CC, financed, delivery vs pickup, custom order.
• Customer identification (driver’s license barcode scan for name/address, or lookup if existing).
• Cart finalize (SKU scan, promo auto-apply, tax auto-calc).
• Finance inline (if used) + contextual protection plan prompt.
• Delivery/PU scheduling with truck/ZIP capacity guardrails.
• Signature + receipt.

E. Governance, Security, and Reliability (2 weeks)

• Role-based permissions with manager-light overrides.
• PCI scope reduction, MFA for back-office, encrypted P2PE terminals.
• Blue/green rollout for instant failover during launch.

F. Training & Change Management (8 weeks total)

• Cohort model: 4 groups × ~30 staff each (sales, cash office, service desk).
• POS Champions per store; shadowed floor for two weekends post-go-live.
• Curriculum: 1.5 days classroom + 2 days live-floor lab per associate.
• Micro-videos for edge cases; competency checks before solo use.

All results measured across a 90-day stabilization window, apples-to-apples (same stores, same seasonal mix).

Speed & Flow

• Median checkout time: from ~42 minutes → 12–15 minutes (−62–71%).
• 90th-percentile orders: from ~62 minutes → ~22 minutes (−40 minutes).
• Financing segment: inline approvals cut ~14 minutes → ~6–8 minutes on average (fewer re-keys, fewer timeouts).
• Manual fields per order: from 22+ → 9 (−59%).
• Peak queue length: typical peak reduced from 12–18 waiting → 4–6; visible line anxiety largely removed.

Accuracy, Risk & Cost

• Price overrides: down 71% after promo rules were codified; exception quality went up (cleaner audits).
• Order error rate (wrong item/price/delivery): 2.3% → 0.6% (−74%).
• Oversell incidents: 1.8% → 0.3% (−83%) with near-real-time inventory.
• Chargeback exposure: −28% (cleaner charge slips, consistent tax/shipping).
• IT incident tickets per 1k orders: −46% (fewer brittle handoffs).

Revenue & Attach (right-sized expectations)

• Average ticket: +5–8%, driven by time returned to selling and fewer mid-checkout derails.
• Protection plan attach: from 9–10% → 12–14% (roughly +3–4 pts, a 30–40% relative lift) from contextual prompts—not hard scripting.
• Financing utilization: +11% relative (more approvals completed because it stayed on one screen).

People & Adoption

• Training coverage: ~120 associates trained across 8 weeks; 92% passed competency checks on first attempt.
• Time-to-first-solo-checkout (new hire): from ~3 weeks to ~8–10 days with micro-learning and sandbox mode.
• Manager time reclaimed: −35% time spent on overrides/voids.

Executive Control

• Live dashboards: store-level throughput, attach by category, approval rates by lender, promo performance, & exceptions heat-map—finally a single version of truth.

Holiday stress test (realistic): Black Friday weekend maintained sub-20-minute 90th-percentile checkouts; no store exceeded 6-person peak queues; finance timeouts dropped to near-zero.

How We Measured

• Checkout time = from “Ready To Pay” to “Receipt Issued” (printed or emailed).
• Segments analyzed separately: cash/CC, financed, delivery vs pickup, custom order.
• Sampling = all transactions across three pre-selected weeks both pre and post; outliers >99.5th trimmed.
• Attribution: any marketing changes were controlled by using same-store, same-period comps and isolating checkout-stage metrics.