# programme delivery · cutover

Someone has to own the date.

A migration with a date on it is not an engineering problem with a project manager attached to it. It is a delivery problem: a rehearsal that has to run before anyone is entitled to be confident, a freeze somebody senior has to hold, a go/no-go that gets called by a person rather than by a mood, and a rollback that has been executed at least once by someone who timed it.

This is what you buy when the estate is not the risk — the weekend is. It applies to an AWS migration and to an S/4HANA conversion, because it is the same cutover. What it is not is an SAP technical practice. No ABAP, no Basis, no functional configuration, no HANA sizing. The boundary is one screen down, and it is not behind a tap.

You leave with a written spek — what your roadmap is blocked on, a fixed scope, and a timeline for your estate. Yours whether or not you use us.

$ ./scope --unit
unit
One cutover. The event with the date on it.
from
The first rehearsal.
to
The written close of hypercare.
keep
The runbook, the criteria and the dashboards — in your repositories and your accounts, written to be re-run.
# one discipline, two estates

Same weekend. Different people in the room.

A cutover is a cutover: rehearse it, freeze it, time it, decide it, be able to undo it, then staff the noise afterwards. What changes between an AWS migration and an S/4HANA conversion is not the discipline. It is who else is in the room, and which of them owns the parts we do not.

Pick the one with the date on it.

estate/ aws

An AWS migration

Workloads, databases, the DNS flip. Your build or a partner's — we sequence the move.

  • We own — the plan; the rehearsal being run and timed; the timed runbook; the go/no-go call; the rollback plan, its rehearsal, and the decision to stop if it has not been proven; and hypercare through to its written exit. And the estate underneath it: accounts, guardrails, the network paths between environments, restores that actually restore, access granted and revoked on a schedule, and the reporting layer a steering committee reads.
  • You own — your application code. We do not write it. We sequence its move, and we own the weekend it moves on.
Book the 30-min review → $ ./book --estate=aws
estate/ sap

An S/4HANA conversion

A conversion with a date, on AWS or not. Your Basis team and your SAP partner keep their work.

  • We own — the plan; the rehearsal being run and timed; the timed runbook; the go/no-go call; the rollback plan, its rehearsal, and the decision to stop if it has not been proven; and hypercare through to its written exit. Plus the programme layer a conversion demands: concurrent workstreams sequenced against one date, executive reporting, risk forecasting and the contingency planning that follows from it. And the AWS estate underneath, where there is one.
  • They own — ABAP, SAP Basis, functional configuration, HANA sizing, and every SAP-certified infrastructure call. We plan and run alongside your Basis team and your SAP partner; we do not replace them. There is no SAP partnership here and no AWS SAP Competency. If either is a requirement on your programme, we are not the firm — and it is cheaper to learn that here than three calls in.
Book the 30-min review → $ ./book --estate=sap

$ ls ./estates — 2 · same runbook, different room

# what one cutover buys

Scoped to one cutover. Named, dated, and closed in writing.

unit — one cutover: the event with the date on it, from the first rehearsal to the written close of hypercare. The date is yours, not ours; what we scope against is the event, not a calendar we do not control.

What lands — 6
01
A timed runbook

Every task with an owner, a predecessor, and a duration taken from the rehearsal rather than estimated at a desk.

02
A rehearsal report

What ran long, what broke, and what changed in the plan because of it. The first pass runs long; that is the point of running it.

03
Signed go/no-go criteria

Written before the rehearsal, by named owners, while everyone was still rested.

04
A rollback runbook

Its own steps, its own measured duration, and the last point at which it can still be taken.

05
A hypercare exit

Triage staffing, a defect path agreed before the first defect, an owner per business area, and the written criteria that end it.

06
The reporting line

One report a steering committee reads without being walked through it — and where the estate is ours to build, the dashboards behind it in your accounts, as code.

What you keep

All six, in your systems: the runbook in your repository, the criteria in your document store, the dashboards in your accounts. They are written to be re-run rather than to be filed.

The deliverable is not the weekend. It is the pattern your own PMO runs the next one from.

fixed scope · you own the output

# the doubt, named

Everyone says they will call no-go.

You would normally test that with a reference call. We do not have one to give you — no client names, no logos, and none coming. So the test gets written down instead, before the engagement starts, and it is yours to enforce.

01
The criteria are signed before the rehearsal.

Not during the weekend. If we ask you to approve go/no-go criteria after the freeze, you have grounds to remove us.

02
The rollback is executed at least once, timed, against a copy of production.

If it has not been executed, we call no-go — including when calling it costs us the rest of the engagement.

03
The freeze is enforced, or it is lifted in writing.

There is no third state. We will not run a cutover against a plan that describes a system that no longer exists.

04
Every red goes in the same report the steering committee reads.

In the reporting cycle we raise it, not at the end. You will not hear about a slip from us once it is too late to act on it.

05
Hypercare has written exit criteria before it starts.

When they are met we leave, and our access to your accounts is revoked while you watch — the same condition the rest of this site runs on.

Five conditions, all checkable against what we are actually doing rather than against what someone says we did. That is the whole substitute for a reference.

$ ls ./conditions — 5 · yours to enforce

# the record

Ask the obvious question.

You should want to know whether anyone here has actually stood in this. The discipline above did not come from a conference talk — it came from SAP S/4HANA conversion programmes run end to end, into real production, where the business was offline if the weekend went wrong. That record, and the exact line where our SAP scope stops, is set out on one page rather than summarised into a bullet here.

The adjacent block of work is Migration Factory — discovery, wave planning and repeatable cutover across an estate. This page owns the date; that block moves the estate.

read: what a rehearsed cutover looks like, from people who have run the weekend → $ ls ./accelerators →
# next step

Whose name is on the go/no-go?

Thirty minutes with an engineer. Bring the cutover you are not looking forward to — the conversion, the database move, the DNS flip — and we will tell you what we would rehearse first, and where your rollback is currently a sentence rather than a path.

You leave with a written spek — what your roadmap is blocked on, a fixed scope, and a timeline for your estate. Yours whether or not you use us.

Book the 30-min review → $ ./book --30min --no-pitch