A chain of forty owned stores tracks lease dates in inboxes, audit findings in binders and its next opening in a spreadsheet. This case is a design-partner pilot: the records first, on the core that already runs, and the site decision brief as the phase its models are being built for.
Where to open this quarter, and which lease dates cannot slip
Duration
Twelve weeks, records first
Runs in
An EU region, agreed before any data moves
> The starting point
Every Record Somewhere,No Decision Anywhere
Each lease's renewal and indexation dates sit in someone's reminders. Audit findings are recorded, but no corrective action is tracked to proof, so the same finding returns on the next visit.
The next opening is scored by hand, one spreadsheet per candidate. When the board asks why site A and not site B, the answer is rebuilt from scratch.
How the decision is made today
Who
Operations, real estate and finance, separately
On what
Spreadsheets, inboxes, accounting exports
Critical dates
Tracked from reminders
Board questions
Answered by hand, each time
# What could not move
What The NetworkRequired First
Settled before any record moved. Three are guarantees of the platform; the fourth is about what Basis is today.
Each network's records are its own
A chain's leases and finances are nobody else's business.
So the system isolates every tenant at row level, enforced in the database.
Nothing is overwritten
Finance and audit records have to stand up later.
So the system keeps financial records append-only; a correction is a reversing entry.
The ERP stays the system of record
Accounting and stock already live in the ERP.
So the system works beside it, from the records it keeps.
Honest about what exists
Basis is in development: the core runs, the site models do not yet.
So the system starts on the built core; the site brief is a later phase, not a promise.
% The architecture
The Records First,The Decision On Top
The pilot is built in the order Basis is: records that hold up, then decisions read from them.
Records
Stores, leases and their critical dates
Finance records, append-only
Documents, versioned
Control
Audits, with corrective actions tracked to closure
Tasks, roles and an audit trail
Weigh
Candidate sites against rent, distance and opening date
Scoring models scoped, not built
Brief
A decision with its reasoning, for the board
The decision stays with the network
Runs inAn EU region, with encryption at rest planned; the model is agreed before any data moves.
= Twelve weeks
Records First,The Brief Last
The first nine weeks run on the core that already exists. The last phase depends on the site models, and moves if they do.
Week0123456789101112You leave with
ScopingWeek 0You leave with: Scope and success criteria
The network
Store list, lease set, audit forms
Ecarus
Roles, tenant setup, import plan
RecordsWeeks 1–3You leave with: Every lease date in one place
The network
Leases, stores and documents
Ecarus
Import, critical-date checks, roles
AuditsWeeks 4–6You leave with: Findings closed, or visibly open
The network
Two audit rounds
Ecarus
Corrective actions tracked to proof
Opening projectWeeks 7–9You leave with: One opening, one record
The network
The candidate sites and their constraints
Ecarus
The opening run as a project, with documents and tasks
Site briefWeeks 10–12You leave with: A brief, or a dated plan for one
The network
Review against the network's best stores
Ecarus
The decision brief, if the site models are ready
Success criteria, agreed in week 0
Critical lease dates surfaced before their notice period
Audit findings closed with proof, against the last round
Time to answer a board question about the opening
Share of network records held in one place
= The decision, as delivered
The Brief TheBoard Would Read
The brief the pilot's last phase delivers, in the format the site-selection models are being built to produce.
Site decision brief
Weeks 10–12
Network
Own chain, 40 stores
Decision
One opening this quarter
Constraints
Rent ceiling, distance to existing stores, opening date
Candidate sites, ranked
Rank
Candidate
Potential
Cannibalisation
Rent against a reasonable range
Why this rank
1
Site A, high street
High
Low
Inside the range
Strongest match to the network's best-performing stores
2
Site B, shopping centre
Good
Notable
Above the range
Shares a catchment with an existing store
3
Site C, retail park
Moderate
Low
Below the range
Opening date at risk from the permit timeline
Why site A ranks first
Its footfall pattern and catchment resemble the network's strongest stores, it draws on no existing store's area, and the rent offer sits inside the range comparable leases support.
Designed so every score opens onto its explanation and the records behind it. The brief recommends; the decision stays with you.
~ What gets measured
On The Network'sOwn Record
The measures agreed in week 0, on the network's own record. Results are published with the network's approval.
Measure
Baseline
How it is measured
Result
Missed or late critical dates
The last twelve months of lease events
Dates surfaced before their notice period, per quarter
Published on approval
Repeat audit findings
The last audit round
Findings that return on the next visit
Published on approval
Time to answer the board
The last opening question
Hours from question to brief
Published on approval
Records in one place
The inventory taken in week 0
Share of leases, audits and documents held in Basis
Published on approval
What stays with the network
Every lease, store and document in one record
Audit history, each corrective action with its proof
An opening run as a project, ready for the site brief
Close to yours if
Lease dates live in reminders
Audit findings return on the next visit
Openings are scored in spreadsheets
You want to shape the product as a design partner
Your operation, in the same format
The technical briefing is where your specific constraints get discussed, including which of these four your operation resembles. Results are published with each customer's approval, never before.
