Peak decides your calendar. Everything else negotiates around it.
Executive strategy outlines and status discipline for retail and consumer-goods programs. The planning constraint in this sector is not complexity, it is the calendar: a trading peak that cannot move, a freeze around it that is longer than anyone budgets, and a store estate that has to be ready before any of it. Walk the five decisions, find your seat, then run the two-minute check.
Five calls that decide a retail program
Set early, tested at peak, and visible in store labor and margin within a season. Pick a decision.
The schedule gets built forwards from readiness and then somebody notices it lands inside the trading peak, or inside the freeze around it.
Fix the freeze window first, agreed with trading, and build the whole plan backwards from it.
The freeze is always wider than the peak itself, because stabilizing before it and unwinding after it both take time nobody schedules. Programs that plan forwards discover this late, at which point the only options are a rushed go-live before the freeze or a delay past it that costs a full season.
What is our agreed freeze window, and does the plan work backwards from it?
Readiness gets reported as a percentage across the estate, which averages a well-prepared flagship with a store that has had no training and a manager who started last month.
Assess readiness per store against defined criteria, and sequence activation by readiness rather than by geography or convenience.
An average hides the tail, and the tail is what generates the tickets. Sequencing by readiness costs a more complicated rollout plan and buys a support model that can reach the stores in trouble, which is the thing that decides whether the estate trusts the system.
Can we name the ten least ready stores, and what is the plan for them?
Point of sale, merchandising, planning and supply chain systems already run. They generate the numbers the business manages by, and the ERP receives them.
Name a client-side owner for each interface and agree posting granularity before design rather than during testing.
Get the granularity wrong and either the ledger drowns in transaction detail or you lose the ability to answer a store-level or category-level question. Both surface at the first close after go-live, which is precisely when nobody has capacity to redesign an interface.
What granularity crosses from merchandising and point of sale into the ledger, and who agreed it?
Scheduling, availability and premium patterns get treated as a workforce management detail while the ERP program focuses on finance and merchandising.
Treat store labor as a first-class workstream with its own owner, because it is the largest controllable cost in the estate and the most visible to staff.
Store colleagues check their hours and their pay personally, in the first cycle after go-live. A scheduling or pay error there becomes a trust problem across the estate at once, and it arrives at the same moment store managers are learning a new system. Add the platform clock: on-premises Workforce Central reaches end of life in March 2027, and moving off it is a reimplementation.
Who owns store labor design, and how many parallel payroll cycles cover a full scheduling pattern?
The support model is designed around corporate users and business hours. Stores trade seven days and the people who need help are serving customers.
Design support around the store, and get day-after-hypercare staffing in writing before the first wave.
If a store cannot get an answer quickly it invents one, and the invented answer spreads through the district. Ticket volume also peaks after the senior people roll off, which is the most predictable staffing failure in this kind of rollout and the easiest to prevent by writing it down early.
What does support look like for a store on a Saturday, the week after hypercare ends?
What those five mean for the chair you sit in
Retail programs get judged on trading, labor cost and margin within a season. Each seat's exposure, the early sign, and the question worth asking this quarter.
The estate averages out the tail
Readiness reported as a percentage hides the stores that will generate most of the tickets. Store teams also solve problems locally and quickly, so any process slower than the workaround loses. The combination produces an estate that appears ready and then behaves inconsistently for a season.
Readiness is reported as an estate percentage rather than per store.
Can we name our ten least ready stores and what happens to them?
Margin truth crosses systems you do not own
Category and store margin read from merchandising and point of sale data at a granularity somebody chose early. Wrong, and you can produce a consolidated number but not a credible category-level one, which is what the business runs on. It surfaces at the first close after go-live.
Nobody can state posting granularity from the merchandising systems without checking.
Can we answer a category margin question on day one after cutover?
Store colleagues check their own hours
Scheduling, availability and premium patterns are the most personally visible part of the program. An error is noticed by thousands of people in one pay cycle. Generic templates assume one schedule and one manager, and a store estate is neither. The workforce management platform clock adds a hard date in March 2027.
Hypercare exit is written in weeks rather than completed payroll cycles covering a full scheduling pattern.
How many parallel payroll cycles cover a complete store scheduling pattern?
The freeze sets the plan, and it is wider than the peak
Decision one, drawn out. Programs that plan forwards discover the freeze late, when the only options are a rushed go-live or a lost season.
Fix the freeze window with trading before anything else in the plan is agreed. Every other date in the program is downstream of it.
Four items already on the calendar
None of these are retail-specific, and all four land on a retail program. Verify each against your own estate before the next steering meeting.
Engineering stopped at the end of 2025. Moving off it is a reimplementation rather than an upgrade, and any shift-based workforce is in scope.
Extended maintenance runs to 2030 for a fee. If the target platform is a lift and shift of ECC, it arrives with a published expiry date attached.
Oracle support runs past 2036. When an integrator sells urgency on that basis, the pressure is customization debt and scarce skills, not vendor abandonment. Knowing the difference is a negotiating position.
Not a deadline, a treadmill. Two mandatory regression cycles annually, permanently, and the item most reliably missing from a post-go-live staffing plan.
Five questions worth more than a readiness assessment
Answerable from memory, scored on this page, nothing captured and nothing emailed.
These five are the start of the instrument. A full review also covers wave sequencing, inventory and cost model design, the merchandising interface inventory and hypercare exit criteria. Or skip the tooling and book the program review.
Rolling out across a store estate?
Pre-SOW, mid-build, or stabilizing after a rough wave with peak approaching. I sell no software and staff no builds, so these questions get asked out loud. Tell me where the program is and I will tell you what I see.