CEDX
Retail · United States
Harbor Retail Group
Harbor put POS, CRM, inventory, and desk on one login so store close and support share the same customer.
40 specialty stores · 380 employees
- Industry
- Retail
- Region
- United States
- Footprint
- 40 specialty stores · 380 employees
- Products live
- CRM, POS, Inventory, Desk, Payments
- Primary metric
- 38% faster store close
- Scenario type
- Illustrative reference
Executive summary
At a glance
- 40-store specialty retailer reconciling POS, inventory, and support on separate systems.
- Pilot: 8 stores on CRM + POS + Inventory + Desk in four weeks.
- Target outcome: same-day store close and tickets with purchase history attached.
The team opens the live app before every standup — no deck, no lag, just the work.
The challenge
What was broken
Deals lived in email. Stock lived in a WMS no one outside ops could read. Support tickets had no purchase history.
Every Monday, revops rebuilt a forecast from three exports. Store managers did not trust HQ numbers; HQ did not trust store counts.
Holiday peaks broke the stack. Support volume spiked while inventory truth lagged by a day.
Stack change
Before and after
What the team stopped using - and what runs on CEDX.
The change
What CEDX replaced
CRM, POS, Inventory, Payments, and Desk on one estate. The same customer graph feeds the register, the ticket, and the forecast.
Store managers see live stock and open tickets. HQ sees the same deal stages the floor uses.
Training used live product pages and demo tenants — the same build prospects open from the site.
Outcomes
What changed on the floor
Results used as pilot targets - the shape of success, not vanity dashboards.
- Store close moved from multi-day reconciliation to same-day numbers leadership will defend.
- Support tickets open with full purchase history attached.
- Forecast meetings dropped the spreadsheet rebuild; the live pipeline is the source.
Team impact
Who feels the change
Store manager
Sees stock and open tickets without calling HQ.
Revops
Forecast from live stages, not three exports.
Support lead
Ticket opens with purchase history already on the record.
Why an estate
Why not another point tool
One customer graph
Sales, service, and finance stop arguing about whose record is authoritative.
One login
Operators do not maintain six passwords and six partial truths.
Live product demos
What you evaluate on the website is the build you implement.
Wave-ready architecture
Start with one motion; add products without a second platform.
Rollout
How they got live
Typical shape - your pilot may compress or expand by product footprint.
Week 1–2
Discovery + estate map
Identity, store hierarchy, catalog cutover plan.
Week 3–5
POS + Inventory pilot
Eight stores live; CRM and Desk wired for HQ.
Week 6–8
Full rollout
Remaining stores, payments, support training.
Week 9+
Hypercare
SLA review, weekly metrics, partner SE on call.
Risks
What this design avoids
- Shadow spreadsheets reappear when systems disagree — the estate is designed so the live number is the board number.
- Tool sprawl after the pilot — success criteria and hypercare keep the team on one stack.
- Training debt — enablement uses the live product, not a parallel slide deck of screenshots.
Your next 90 days
Playbook you can copy
Same skeleton Harbor, Northline, and Atlas-shaped programs use.
Days 1–14
Estate map
Sponsor, operators, systems list, retire list, pilot metrics.
Days 15–45
Pilot live
One pod or region on production-shaped config.
Days 46–75
Expand wave
Second region or adjacent product family.
Days 76–90
Hypercare exit
Metrics review, owner map, BAU support path.
In the room
Software that respects the people running it
Harbor Retail Group runs day-to-day work in CEDX - the same apps you can open from every product page.
FAQ
About this scenario
Is this a named public case study?
It is an illustrative composite on live demo tenants. Named case studies publish with customer approval.
Can we pilot the same products?
Yes. Products in this scenario: CRM, POS, Inventory, Desk, Payments.
Who implements?
CEDX, a consulting partner, or both - decided in the estate map.
Architecture notes
What sits under the pilot
Identity
One login for operators in scope. Entitlements follow the product graph.
Data
Customer graph ownership is written in the estate map before imports.
Integrations
Prefer estate events and first-party connectors over brittle nightly CSV.
Commercial shape
How programs are usually bought
Pilot SKU
Products in the first motion, seats for the pod, success metrics, hypercare window.
Expansion
Wave-two products and regions after metrics clear - written in the original map when possible.
Related scenarios
More operators on the estate
Northline Systems
Northline collapsed six tools into one estate so new reps quote, sign, and bill on the same customer graph.
LogisticsAtlas Cold Logistics
Atlas put field jobs, parts, and contact-center voice on one desk record across three provinces.
Food & beverageCedar Foods
Cedar connected campaigns to CRM to orders so marketing and ops stop arguing about whose number is real.
Map a retail program like Harbor Retail Group
Thirty minutes. Products, retire list, and pilot metrics - before anyone configures.