Skip to content

Delivery

MSP exit plan: the handover pack your IT provider should deliver

The nine sections of the pack we commit to hand over at exit, and the two decisions that decide whether any pack is worth having.

Computing America IT. Help desk support, backups and network infrastructure.

Send the pack
4 min readComputing America
A neatly packed archive box of tabbed folders with a USB drive and a sealed envelope on top.

In short

  • The cost of leaving a managed service is set at signing and onboarding, by the contract terms and by whose name holds the tenants, domains, licenses and circuits.
  • "Reasonable cooperation during transition" is not a deliverable. A deliverable has a name and a table of contents.
  • A pack assembled in the notice period is thinnest where the next provider needs it most: open issues and the reasons behind changes. Maintain it as the work happens.
  • Ask a prospective provider what they hold that you would have to get back, and ask to see the table of contents of today's pack.

A handover pack is a named set of documents, and a good one has nine sections. Contract wording often reads "reasonable cooperation during transition." That is an attitude with no table of contents, so it can't be incomplete.

The difference shows up when you give notice and need to know who holds the account for the site-to-site circuit. Below is the pack we commit to as a deliverable, and the two design decisions that decide whether any pack is worth having.

The index

The index is a directory. Your next provider should be able to read it in an afternoon and work from it on day one. If your contract says only "reasonable cooperation," ask to attach it as a schedule, with a refresh interval and a named owner for each section.

handover/pack index
handover/  00-README.md          what this is, who to call, and what is missing  01-inventory/         devices, accounts, licenses, circuits, contracts  02-identity/          tenants, domains, admin roles, break-glass accounts  03-network/           topology, addressing, firewall rules, circuit details  04-backup/            regime, retention, and the last verified test restore  05-runbooks/          the twenty things that break here, and the fix for each  06-vendors/           who to ring, account numbers, renewal and notice dates  07-open-issues/       what is broken today, and what we were about to do next  08-changes/           what we changed while we ran it, and why  09-credentials.md     where every credential lives, not what it is

What each section saves the next provider

Each section answers a question the incoming provider would otherwise answer by discovery, at your cost, in their first month.

SectionWithout it
01 InventoryDiscovery at your cost, and a license bill that can't be reconciled.
02 IdentityA tenant or domain in somebody else's name makes the exit a migration, not a handover.
03 NetworkFirewall rules that stay forever, because the risk of deleting one is unknown.
04 BackupA job that reports success and a restore that produces a working system are different things. You find out which you had at the worst moment.
05 RunbooksYour staff teach the new provider your estate one incident at a time, and are billed for it.
06 VendorsAn auto-renewal that slipped through, or a circuit outage where the carrier won't talk to you because you aren't the account holder.
07 Open issuesProblems transferred silently, and a new provider blamed for finding them.
08 ChangesDecisions whose rationale left with the person who made them, so touching anything is a gamble.
09 CredentialsShared passwords in a spreadsheet, and no way to tell which are still live.
Each section, and what its absence costs you

Section 09 records where credentials are held and how custody transfers. The credentials stay out, because the pack travels by email and file share.

The first decision: write it as the work happens

The pack that fails is assembled during the notice period. People who are leaving write it against a deadline, from memory, about a system they no longer have a reason to explain. It's thinnest in sections 07 and 08, where the incoming provider needs it most, and "as part of offboarding" promises a document that doesn't exist yet. What was broken and what we were about to do is live state that is gone a month later. Why we changed something is a rationale, and a rationale written eighteen months on is a guess in a confident tone.

The second decision: hold nothing that has to come back

What leaving costs is settled at onboarding, by whose name everything is in. Contract the tenants, domains, license subscriptions and circuits in your name and keep administrative access, and a handover is a copy.

Fig. 1
  1. Group: Stays where it is

    In your name

    Nothing moves: tenants, domains, licenses, circuits

  2. Group: Leaves at exit

    In our systems

    Agents, ticket history, documentation, your data

    • To Exported to you
    • To Certified destroyed
    • To Agents removed
  3. Group: Leaves at exit

    Exported to you

    Tickets readable, documentation as content, data neutral-format

  4. Group: Leaves at exit

    Agents removed

  5. Group: Leaves at exit

    Certified destroyed

    Our copy of your data, after export

At exit, nothing in your name moves; everything in our systems is exported to you or removed, and our copy of your data is certified destroyed.

SourceOur published leaving terms (process page, IT division).

A provider can have ordinary reasons for the opposite arrangement. Reselling licenses earns margin and eases procurement, registering the domain helps a client with no IT staff, and holding the firewall under the provider's management account fits its tooling. Each choice is defensible, and together they decide how hard a change of provider is.

Holding nothing costs us. Resold licenses are recurring margin we don't take, and procurement is work we do without billing for it. We think that is the right trade for a client who may someday leave.

What we do hold, and what happens to it

"We hold nothing of yours" would not be true for us either. Our monitoring and management agents run on your endpoints under our licenses, our ticketing system holds your request history, and our documentation lives in our systems while we run the estate.

  • Section 01 identifies everything we deploy, and whether it stays useful to you without us. Saying early what won't stay useful is the point.
  • Agents come off on a schedule you set. An endpoint that loses its agent on the last day and gains nothing is worse off than one that keeps it for a couple of weeks.
  • Ticket history is exported to you in a readable format. It records what your staff struggle with, which an inventory doesn't.
  • Documentation is handed over as content you keep. A link into a system you'll lose access to doesn't count.
  • Your data is returned in a platform-neutral format, then certified destroyed against NIST SP 800-88, with the certificate handed over as a document.

Four questions for whoever runs your IT today

You don't need to be leaving to ask these, and the answers are more useful when you aren't.

  1. 1.Which of our tenants, domains, license subscriptions and circuits are contracted in your name? Ask for the list itself, not an answer in principle.
  2. 2.If we gave notice on Monday, what document would we receive, and can we see the table of contents of today's version?
  3. 3.When did you last perform a restore, as distinct from the last successful backup job, and what was restored?
  4. 4.What have you deployed into our estate that stops working the day you stop being our provider?

Next step

Send us what you were handed.

Forward whatever the last provider actually left you, even if it is one spreadsheet and a list of passwords. We will tell you which items on the pack are missing and which of those you can recover without their cooperation.

Reply
A person replies, not a sequence: within one business day, from someone who would be on the engagement.

Packages

Start with a package

Four packages, each with its price on the page. Pick one and buy it online.

Billing