Team planning a migration to Google Workspace
Google Workspace Migration Partner

Google Workspace Migration

A migration moves your mail, contacts, calendars and files onto Google Workspace while the system you use today stays live — then switches your domain over in a planned window, rather than in one irreversible moment.

The data is rarely the hard part. What decides whether a migration lands is the discovery done before it, the cutover planned around it, and whether anyone is still answering questions the week after. This page covers all three honestly, including what does not migrate.

How a migration runs

  1. 1Discovery and designScope, structure and the cutover window
  2. 2Pilot groupProblems found cheaply, at small scale
  3. 3Bulk and delta syncYour current system stays fully live
  4. 4Cutover and hypercareMX switched in a quiet window, then support
The honest version

What actually moves — and what does not

Most migration pages say “everything transfers seamlessly.” It does not, and the gap between that promise and the first Monday afterwards is where the complaints come from. Here is the real breakdown.

WhatStatusWhat that means in practice
Email messages and foldersMoves cleanlyFolders become Gmail labels, and nested folders become nested labels. The hierarchy survives; the metaphor changes, which is a training point rather than a technical one.
ContactsMoves cleanlyPersonal contacts come across. A shared or global address list usually has to be rebuilt as a Google Group or a shared directory — it is not a file you can copy.
Calendars and eventsMoves, with checksRecurring events and meeting-room bookings are the usual casualties. Both are migrated and then spot-checked, because a silently broken recurring meeting is not noticed until it fails to appear.
Files and foldersMoves cleanlyThe catch is ownership. Files land owned by the account that migrated them, which quietly recreates the problem you are trying to escape unless they are moved into Shared Drives owned by the organisation.
File sharing permissionsMoves partiallyExplicit shares on a file usually carry across. Permissions inherited from an old folder tree frequently do not, so access is verified after the move rather than assumed.
Distribution lists and groupsRebuiltRecreated as Google Groups with their members and addresses. Worth treating as an audit — most organisations find lists that have not been correct for years.
Aliases and forwardingRebuiltRecreated in the Admin console. Copying them blindly carries forward every forward to an ex-employee, so this is another audit point rather than a copy job.
Server-side rules and filtersUsually not movedThey rarely translate between platforms. In practice users recreate the two or three they actually rely on, and quietly abandon the rest.
Email signaturesNot automaticNormally re-applied centrally at cutover, which is the natural moment to standardise a set that has drifted across the company.
Shared mailboxes and public foldersRedesignedThese become delegated mailboxes, Google Groups or Shared Drives depending on how they were genuinely used — which often differs from how they were originally intended.
Archives and journaled mailDepends on sourceRecoverable if it sits in a mailbox or an exportable file. A journaling or compliance archive usually needs its own export and is scoped separately.

The three rows worth reading twice are file ownership, shared mailboxes and permissions. Each of them can migrate “successfully” in a technical sense and still leave you with the problem you were trying to solve.

Source systems

Where businesses migrate from

The destination is the same every time. What changes the scope, the timeline and the price is what you are leaving — and specifically which parts of it have no direct Google equivalent.

Microsoft 365 / Exchange Online

The most common source, and the most predictable. Mail, contacts and calendars migrate well. The work is in shared mailboxes, distribution lists and public folders, which have no direct Google equivalent and have to be redesigned rather than copied.

Read morearrow_forward

cPanel, IMAP or shared-hosting webmail

Usually the easiest data to move and the hardest environment to leave tidily. Mail comes across over IMAP; the difficulty is the forwarding rules, catch-all addresses and old aliases nobody has documented in a decade.

Read morearrow_forward

Consumer Gmail accounts

A business running on personal @gmail.com addresses. Mail and Drive content migrate, but the real work is untangling which files belong to the business and which belong to the individual — and doing it before someone resigns.

Read morearrow_forward

On-premise Exchange Server

Scoped individually. Mailbox sizes are usually far larger than anyone expects, the server is often several versions behind, and the cutover needs a coexistence period rather than a single switch.

Another Google Workspace tenant

Demergers, acquisitions and domain consolidations. Technically clean because both ends speak the same API, but the ownership and Shared Drive structure need deciding before anything moves, not after.

Read morearrow_forward

Zoho Mail or another business suite

We implement Zoho as well, so this is a conversation about fit rather than a sales pitch. If the honest answer is that you should stay where you are, we will say so before you spend anything.

Read morearrow_forward

Migrating from something not listed here is usually still straightforward — what matters is how the data comes out of it, which is confirmed during discovery before anything is quoted.

The process

Seven phases, and where each one earns its keep

Nothing here is unusual — any competent migration runs roughly this way. What varies is how seriously the first phase and the last one are taken, and those are the two that decide the outcome.

Design overlaps heavily with implementation on most projects, since there is little point moving data into a tenant whose structure has not been settled.

fact_check

1. Discovery

Mailbox count and sizes, total Drive footprint, what the DNS looks like today, which shared addresses exist, and what has to be retained for compliance. This is where the scope is actually set — everything quoted later depends on getting it right.

route

2. Design

User and group structure, Shared Drive layout, sharing policy, security baseline and the cutover window. Decided before anything is created, because the structure is far harder to change once data is in it.

sync

3. Pilot

A small group moved first, usually including at least one person who will complain loudly if something is wrong. Problems surface here, at a scale where fixing them is cheap.

cloud_done

4. Bulk sync

The main body of data copied across while your existing system stays fully live. Nobody is disrupted, and nothing depends on this finishing at a particular hour.

published_with_changes

5. Delta sync

Everything that arrived during the bulk sync is caught up. Repeated as needed, so the gap between the two systems is minutes rather than days by the time you cut over.

dns

6. Cutover

MX records switched at a low-traffic window, with forwarding in place so nothing bounces while DNS propagates. A final delta sync follows the switch.

support_agent

7. Hypercare

The period immediately after, when the questions actually arrive. Signatures, missing shares, a rule someone relied on. Short, intensive, and the difference between a migration that lands and one that is merely finished.

Tech Geum team running a Google Workspace migration
The cutover

What happens when the MX record changes

This is the moment everyone is nervous about, and the one most often described vaguely. It is worth understanding, because the nervousness is mostly misplaced — a cutover is not a switch that either works or loses your mail. It is a window during which two systems are both valid destinations.

schedule

Days before: the TTL comes down

The time-to-live on your MX record is lowered — often to a few minutes. That tells the world's DNS resolvers to stop caching the old answer for hours, so when the change comes it propagates in minutes rather than a day.

published_with_changes

Hours before: a final delta sync

Everything that arrived since the bulk sync is caught up, so the two systems are effectively level when the switch happens.

dns

The switch itself

The MX records are repointed at Google. This takes seconds to do and minutes to hours to propagate, depending on how well the TTL reduction took.

forward_to_inbox

During propagation: nothing bounces

Some senders will still resolve the old MX for a while. Forwarding or dual delivery is left in place so that mail is delivered rather than rejected — this is the single step that makes a cutover safe.

mark_email_read

After: one more delta sync

Anything that landed on the old system during propagation is pulled across. Only once that is clean does the old MX stop receiving.

lock

Later: decommission, deliberately

The old system stays read-only for a period before anyone cancels it. This costs one more billing cycle and removes an entire category of risk.

Deliverability

The part that fails two weeks later

Migrations rarely fail on the day. They fail a fortnight afterwards, when someone mentions that their quotes have stopped reaching a customer. The cause is almost always authentication that was never completed, because the migration was judged finished the moment mail started arriving.

verified

SPF

Declares which servers may send as your domain. Google's mechanism has to be included — and if the old provider's include is left in place during coexistence, watch the ten-lookup limit, which silently invalidates the whole record when exceeded.

key

DKIM

Cryptographically signs your outgoing mail. This is the one that gets missed: Google Workspace does not sign by default. The key has to be generated in the Admin console and published in DNS as a deliberate step, and until it is, your mail is authenticated more weakly than before you moved.

policy

DMARC

Tells receivers what to do when SPF and DKIM fail, and reports back on who is sending as you. Start at monitoring only, read the reports for a few weeks to find the legitimate senders nobody remembered — the CRM, the invoicing tool, the booking system — then tighten to quarantine and reject.

All three are configured and verified as part of the migration rather than left as a follow-up task. Going straight to DMARC enforcement without reading the reports first is the reliable way to stop your own invoices from being delivered.

What goes wrong

The seven failures worth planning for

None of these are exotic. They are the same handful of problems every time, which is precisely why they are avoidable — and why a migration that skips discovery tends to meet several of them at once.

RiskWhat actually happensHow it is prevented
Mail lost at the switchMessages keep arriving at the old server for hours after the MX change, because DNS propagation is not instant and some senders cache aggressively.Forwarding or dual delivery stays up until the old MX genuinely stops receiving, and the TTL is lowered days in advance so propagation is fast.
Deliverability collapsesMail starts sending from a new source with no matching authentication, and it lands in spam at exactly the moment everyone is watching.SPF, DKIM and DMARC are published and verified before cutover, not after. DKIM in particular is not enabled by default — it has to be generated and published deliberately.
Files nobody can findEverything migrated into individual Drive accounts. Two months later someone leaves and their files leave with them.Organisational content goes into Shared Drives owned by the company from the start, rather than being tidied up afterwards.
Calendars quietly breakRecurring meetings and room bookings do not survive intact, and nobody notices until a meeting fails to appear.Calendars are migrated and then spot-checked against the source, with rooms recreated as bookable resources rather than left as text.
Storage runs out immediatelyYears of archived mail and abandoned project folders land in a pooled allowance sized for current usage.The real footprint is measured at discovery. Sometimes the answer is a higher edition; often it is a clear-out that removes the need for one.
Shared addresses stop workinginfo@ or accounts@ was a mailbox several people signed into. Migrated as a user account, it breaks the way the team actually worked.Shared addresses are redesigned at discovery as groups or delegated mailboxes, matching how they are used rather than how they were set up.
Everyone reverts to old habitsPeople keep emailing attachments because nobody showed them Shared Drives, and you have paid for storage that stays empty.Short, role-specific sessions at cutover covering what changed for that person — not a generic product tour.
Timing

How long a migration takes

Typical ranges from the migrations we run. Headcount is the weakest predictor of the four — what genuinely moves these numbers is the number of source systems, the size of the shared-folder estate, and how quickly decisions get made on your side.

SituationTypical spanWhat decides it
Under 20 users, one source systemOne to two weeksUsually limited by how fast you can free up a cutover window, not by the data.
20 to 100 usersTwo to four weeksShared mailboxes, group structure and the size of the Drive footprint are what move this range.
100 to 300 users, multiple sourcesFour to eight weeksPhased by department or site, with several cutover windows rather than one.
On-premise Exchange, or a heavy shared-folder estateScoped individuallyMailbox sizes, server version and the length of the coexistence period decide it. Estimating this without discovery would be guesswork.

These are indicative, not a quotation. The span for your organisation is fixed at discovery, once the mailbox sizes, source systems and retention requirements are actually known.

Preparation

Six things worth doing before anything moves

You can do all of these yourself, and doing them makes the discovery call far shorter. None of them require technical knowledge beyond knowing how your own business works.

inventory_2

Count what you actually have

Mailbox sizes, total Drive or file-server footprint, and how many of your user accounts are still real people. The last one changes the licence count more often than anything else.

alternate_email

List every address in use

Shared addresses, aliases, catch-alls and forwards. There are almost always more than anyone remembers, and finding one after cutover is a bad way to find it.

dns

Confirm who controls the DNS

The single most common cause of a delayed cutover is nobody having login access to the domain registrar. Establish this in week one, not on the night of the switch.

gavel

Settle the retention question

How long mail has to be kept, and whether anyone needs to search it later. That answer affects the edition, so it belongs at the start rather than after the licences are bought.

schedule

Pick the quiet window

Every business has one — a weekend, a holiday, the end of a season. Cutting over into your busiest week is a decision you make once.

group

Name someone on your side

One person who can answer questions about how the business works and make small decisions quickly. Migrations stall waiting for answers far more often than they stall on data.

After the switch

The week nobody plans for

A migration is usually declared finished at the cutover. In practice the cutover is the point at which the remaining work becomes visible, and the organisations that handle this period well are the ones that still think the project went smoothly a year later.

task_alt

Keep the old system read-only for a while

Do not cancel it the moment the switch is made. A short overlap costs one more billing cycle and removes the entire category of panic about something that was not copied.

task_alt

Verify before you decommission

Spot-check mailboxes, shared folders and calendars against the source. Decommissioning is the one step that cannot be undone, so it goes last and deliberately.

task_alt

Tighten DMARC gradually

Start at monitoring only, read the reports for a few weeks to find the legitimate senders you forgot about, then move to quarantine and reject. Going straight to enforcement is how invoices stop arriving.

task_alt

Re-check the licence count

Once the dust settles, the true number of active users is visible. It is usually lower than the count you migrated, and that is a real saving.

task_alt

Hand over the admin documentation

Who has admin rights, how groups are structured, where the DNS records live and why they are what they are. Without it you have swapped one undocumented system for another.

Google Workspace migration — common questions

help

How long does a Google Workspace migration take?

expand_more

For a team under 20 users moving from a single source, one to two weeks is typical, and the limiting factor is usually finding a cutover window rather than the data itself. Between 20 and 100 users it is more often two to four weeks. Larger organisations with multiple source systems run in phases over four to eight weeks. On-premise Exchange and heavy shared-folder estates are scoped individually, because mailbox sizes and the length of the coexistence period decide the answer more than headcount does.

help

Will we lose email during the migration?

expand_more

Not if the cutover is planned properly. Your existing system stays fully live while data is copied across, so nothing depends on a single switchover moment. The DNS time-to-live is lowered days beforehand so the MX change propagates quickly, and forwarding or dual delivery is kept in place until the old server genuinely stops receiving. Mail that arrives during propagation is delivered, not bounced, and a final delta sync after the switch catches anything still in transit.

help

What actually moves, and what does not?

expand_more

Mail, contacts, calendars and files move. File sharing permissions move partially — explicit shares usually carry, but permissions inherited from an old folder tree often do not. Distribution lists, aliases and shared mailboxes are rebuilt rather than copied, because Google has no direct equivalent for some of them. Server-side rules, filters and signatures generally do not transfer, and in practice people recreate the handful they genuinely use.

help

Can we keep working during the migration?

expand_more

Yes. The bulk of the data is copied while the old system is still running normally, so there is no period where the business is offline waiting. The only moment that needs coordinating is the MX switch itself, which is scheduled for a quiet window and takes effect over a few hours rather than instantly.

help

Do we need to tell everyone in advance?

expand_more

Yes, and it matters more than most people expect. The technical migration can be flawless and still be judged a failure if users arrive on Monday to an interface they were not warned about, a signature that has changed and no idea where the shared folder went. A short, role-specific briefing at cutover prevents most of the support load that follows.

help

What happens to email that was already archived?

expand_more

If it sits in a mailbox or an exportable file it comes across with everything else. A dedicated journaling or compliance archive is a different matter — those usually need their own export and are scoped separately, and the retention requirement behind them may also decide which Google Workspace edition you need.

help

Why does DKIM keep being mentioned?

expand_more

Because it is the most commonly missed step in a Google Workspace migration. DKIM signing is not switched on by default — the key has to be generated in the Admin console and published in your DNS deliberately. Skip it and your mail is authenticated more weakly than it was before you moved, which tends to show up as deliverability problems a week or two after cutover when everyone has stopped watching.

help

Can you migrate from a system you have not worked with before?

expand_more

Usually, yes. Most business mail systems can export over IMAP or a standard file format, and that is what a migration actually depends on. What we would not do is quote a fixed scope for an unfamiliar platform without first confirming how the data comes out of it — so that verification happens during discovery, before you commit to anything.

help

Should we migrate everything, or start clean?

expand_more

It is worth asking, and the honest answer is sometimes not everything. Migrating fifteen years of mail that nobody will ever open costs storage and can push you to a higher edition for no benefit. A common middle path is to bring across a defined period in full, archive the rest in an accessible form, and keep the old system read-only for a while. That decision belongs at discovery, because it affects the edition you buy.

help

What does a migration cost?

expand_more

It is scoped on volume, source platform and how much has to be preserved — a dozen mailboxes off cPanel is a different job from 200 users and fifteen years of shared folders off Microsoft 365. We quote it as an explicit line alongside the licences rather than folding it into a rate, so you can see what each part costs. The requirement call and the quote are free.

Plan your Google Workspace migration

Tell us what you are running today and roughly how many users. You will get back a migration plan with a scope, a cutover window and a fixed-scope figure in INR — and an honest answer about anything in your setup that will not migrate cleanly.

  • check_circle

    Discovery before the quote, not after the invoice

  • check_circle

    Cutover planned around your trading hours

  • check_circle

    SPF, DKIM and DMARC configured as part of the work

  • check_circle

    Support through the week after, when the questions actually arrive

Prefer to talk first? Call +91 99467 89916 or email admin@techgeum.com.

call