Moving in is not the same as losing everything
Most teams never migrate at all: JiraVision sits on top of your Jira, so your issues, history and sprints stay exactly where they are. If you're coming from Asana, Linear, Trello or Notion, this page is the honest version — what carries, what gets rebuilt, and what you should leave read-only for a while.
- Jira teams: no export, no import — JiraVision reads your existing project
- Build a step-by-step plan for your own workspace and download it
- Effort ranges are labelled illustrative, never a commitment
The problem
Nobody fears a new tool. They fear losing five years of history.
The hard part of adopting anything is the suspicion that you'll trade a known workspace for an empty one — and find out three months later that the context lived in comments nobody copied.
The frame is usually wrong
Vendor migration pages describe ripping the old tool out. That's the moment people freeze: an irreversible write on a workspace full of other people's work. We don't ask for it.
JiraVision sits on Jira
If your issues already live in Jira, there is no import. JiraVision reads your Jira project — issues, history, sprints, story points — and writes back only through the approval gate. Your source of truth doesn't move.
For everyone else, it's a project
Coming from Asana, Linear, Trello or Notion is real work with real sequencing. So it gets a plan with phases, exit criteria and an archive step — not a promises page.
Not a data-loss event
What actually comes with you
Read this before the planner. The scariest question in any migration is what you can't get back, so here is the split written down: what is copied, what stays where it is, and what we deliberately do not touch.
Copied into your workspace
- Issues or tasks: title, description, assignee, due date
- Comment threads and history where the source exposes them
- Status, priority and estimate fields that have a numeric target
- Sub-tasks and checklists, with their completion state
Stays where it is
- For Jira teams: everything. Jira remains the source of truth
- Attachments remain in the original system unless copied on purpose
- Archived and closed items usually stay archived — by decision
- Documentation pages stay documentation, not issues
Never copied
- Credentials, tokens and SAML configuration
- User accounts — people are invited, not imported
- Embeds, synced blocks and cross-workspace page widgets
- Nothing is deleted from the old tool by us, ever
Attachment scope is the one thing we won't quote a number for on a page: ceilings depend on your workspace and on source-side export limits we have not pinned down. Confirm scope with us before committing — we'll tell you where your workspace lands.
Ask about attachment scopeThe sequence
Discovery, parallel run, cutover, archive — in that order
Every phase has an exit criterion, and the archive phase is a phase. That ordering is the reason a migration can be paused: you always know which gate you're standing in front of.
- Step 1
Discovery
Read-only. We inventory issues, fields, automations and the dashboards people depend on. Nothing is written anywhere.
- Step 2
Parallel run
Two sprints with both tools open. The old tool stays writable precisely because we expect it to win a few arguments.
- Step 3
Cutover
A named date at a sprint boundary, announced with the archive location attached. One switch, then a daily watch period.
- Step 4
Archive
The old workspace goes read-only and stays there. Deletion is a separate, much later decision — usually never taken.
Migration planner
Pick your source, set your size, read the plan
Two inputs and two switches produce a phase-by-phase plan with exit criteria you can hold us to. Tick items off as you go, then download the whole thing as a text file for the team channel.
Coming from
Nothing to import. JiraVision reads your Jira project and writes back only when you approve a plan, so migration here is configuration, not data movement.
Team size is the scope proxy: 4 boards at roughly 5 people per board. It is not a delivery promise.
For Jira teams this changes nothing: JiraVision never copies files, Jira stays the store.
Two sprints with both tools writable. Turning it off deletes a phase and moves that risk into cutover week.
- Scope proxy: 20 seats ≈ 4 boards at 5 people per board. Team size is the only scope input.
- Phase hours = base + (hours per seat × team size) × the source multiplier. Both tables are printed in the “How these numbers are made” section below.
- Displayed band is 70%–135% of the mid figure — a good week and a bad one.
- Elapsed weeks assume 6–10 focused hours per week, because nobody migrates full-time.
- Attachments on: +0 h per seat folded into workspace build (zero for Jira — nothing is copied).
- Parallel run included: two sprints with both tools writable.
Jira · 20 seats
~33–62 hover 5–8 weeks elapsed
Effort ranges are typical, based on workspace size — illustrative, not a commitment.
01Discovery
~4–7 h · mid 5 h
Read the workspace before anybody makes a decision about it.
Done when: A written scope list exists and both named owners have agreed to it.
02Field mapping
~3–5 h · mid 4 h
Every source field gets a target, a fallback, or an honest “stays behind”.
Done when: A mapping sheet where every row reads carries, rebuilt, or stays behind — no “we'll see”.
03Workspace build
~5–9 h · mid 7 h
Boards, permissions and automations exist before anyone depends on them.
JiraVision never copies attachments — Jira stays the store, so there is nothing to do in this phase. Your history and files stay exactly where audit already points.
Done when: A pilot team runs a full sprint end to end without opening the old tool.
04Parallel run
~18–35 h · mid 26 h
Two sprints with both tools open, JiraVision as the working copy.
For Jira teams the parallel run is free: Jira stays authoritative the whole time, so “both tools open” is just the default state.
Done when: Zero blocking gaps across two consecutive sprints.
05Cutover
~2–4 h · mid 3 h
A date, not a mood: one announcement, one switch, one watch period.
Done when: One full sprint closes with nobody needing write access to the old workspace.
06Archive & cleanup
~1–2 h · mid 2 h
The old workspace goes read-only. Nothing is deleted until you say so.
Done when: Old tool is read-only, the backup location is written down, and no team member has write access.
Jira field map
Shape, not a guarantee: the importer confirms or refuses each row against your actual workspace during discovery.
| In Jira | In JiraVision | Verdict |
|---|---|---|
| Issue type | Issue type Read straight from Jira — no rename, no remap. | carries |
| Status / workflow | Board column Statuses map one-to-one; the workflow itself is untouched. | carries |
| Sprint | Sprint Jira sprints are read, never recreated. | carries |
| Story points field | Story points Needs to be numeric and on the screen scheme — a config check, not a conversion. | carries |
| Comments & history | Activity timeline Read from Jira. Nothing is re-entered or duplicated. | carries |
01
Field mapping is where a migration is won or lost
Importers rarely fail loudly. They fail by putting your sizing field in the wrong place, or by turning a grouping section into a status column that nobody can get work out of. So the mapping sheet is the artefact: every source field gets a verdict before anything moves.
- Statuses become columns on purpose, one decision each
- One numeric field is nominated as the sizing field — no exceptions
- Fields with no equivalent become labels or are dropped in writing
- Anything the importer will refuse is listed before it runs
Asana → JiraVision
| TaskIssue | Name, description and assignee come across. | carries |
|---|---|---|
| SectionBoard column or epic | Needs a human decision per section: status or grouping? | rebuilt |
| Number custom fieldStory points | Only when the field is numeric and consistent across the project. | carries |
| ProjectJira project | One Asana project per Jira project keeps the mapping legible. | carries |
| CommentsActivity | Text and @mentions as text. Nested file links need the attachment pass. | carries |
Asana as the example — its sections are the classic “is this a status or a grouping?” argument.
See your source in the planner02
Two sprints with both tools open
The parallel run is the phase people want to skip and the reason the rest survive contact with reality. The old tool stays writable, JiraVision becomes the working copy, and every place the old tool wins gets logged instead of argued about.
- Committed vs delivered compared at each retro, in both tools
- New dashboards and reports are frozen in the old tool
- Exit criterion is concrete: zero blocking gaps across two sprints
- For Jira teams it is the default state, not an extra project
03
Rollback is a setting, not an apology
Every phase here leaves a way out, which is why saying yes is low-stakes. We never delete anything in the old workspace, and archive means read-only — a flag somebody with admin rights can turn back off.
- Read-only archive, indefinite, unless you ask us to help delete
- Backup location is written down during the archive phase
- Jira teams can uninstall the layer and lose nothing
- The plan pauses anywhere; there is no point of no return
During discovery
Nothing to undo — the integration is read-only.
During the parallel run
Keep writing in the old tool. It was never turned off.
After cutover
Remove the read-only flag, carry on as before.
After archive
Still nothing deleted by us. The old workspace is a normal account.
Your side of the table
Four decisions only you can make
We can run the mechanics. These four calls belong to the team, and every one of them is a conversation that gets cheaper before the migration than after it.
What each status means
In the old tool, “In Review” probably means three different things to three different people. Pick one meaning per column, or the column becomes a parking bay.
Which field is the sizing field
One numeric field, and it means effort. Teams that carry a ticket ID or an hours-original-estimate column in as story points get a capacity model that lies to them.
What stays documentation
Especially from Notion and Asana: pages that describe work are not work items. Decide the boundary, or inherit a backlog of meeting notes.
When the date is, and where the archive lives
Cutover needs a named sprint boundary and a written location for the read-only workspace. “Soon” is not a plan; nobody has to sign anything on the day.
How these numbers are made
The estimate model, with the working shown
The planner above is arithmetic you can audit. Here is every input it uses — including the ones we'd rather not have to explain, like why we won't promise you a date.
- One input: team size
- Every phase costs base + (hours per seat × your team size). Team size is a proxy for workspace size; it is not a measure of how disciplined you are.
- One weight: which tool you're leaving
- Each source carries a multiplier per phase. Jira is well under 1 because there is no data movement; Notion is over 1 because documentation triage is slow.
- A band, not a point
- We show 70% to 135% of the mid figure. A single number implies a precision nobody has earned, and the low end is what a good week looks like.
- Elapsed weeks ≠ effort
- Weeks are the mid-range hours divided by 6–10 focused hours per week. Nobody migrates a workspace full-time, and pretending otherwise is how dates get missed.
- Two switches change the shape
- Carrying attachments adds scope-review hours to the build phase. Turning off the parallel run deletes a whole phase — and moves its risk into your cutover week.
- What the model refuses to claim
- No migration counts, no “typical team finishes in X weeks”, no adoption-rate promises. We have not logged enough migrations yet to publish any of those honestly.
- Where the numbers come from
- Our own phase checklist, priced by hand. The inputs we haven't confirmed — attachment ceilings and competitor field semantics — are the first things to replace once real migrations are instrumented. Until then they are labelled as model weights, never as facts.
Straight talk
When a migration is not worth it
Works well if
- You're on Jira and want planning, load and fairness signals — there's no migration to do
- One to four teams that can run a real parallel sprint without a programme plan
- Your old tool's reporting is the actual reason for leaving, and it's documented
- Somebody can own the mapping sheet for two weeks
Not for teams that
- You're on Jira Data Center with a compliance regime that blocks a cloud integration — talk to us first
- The tool isn't the problem: nobody estimates, or the roadmap is decided elsewhere
- A 400-seat single workspace where no team can name their own custom fields
- You need the old licence cancelled by a date that is already next week
Per-source notes
What changes by source, and what doesn't
The sequence is the same for everyone. What changes is which phase eats your time, and how much of your old workspace's meaning survives the trip.
Jira-only teams
You are not migrating. You are connecting a read-only integration, approving a scope list, and turning the planning layer on sprint by sprint.
- No export file, no import job, no double-keying
- History and attachments stay in Jira where audit already points
- The worst case is uninstalling an app
Asana teams
Strongest fit when Asana is already being used as a sprint board rather than a task list — the section structure maps cleanly once somebody rules on what each section means.
- Sections become columns or epics on purpose, not by accident
- Number fields pick up story-point meaning
- Comment threads come with the task
Linear teams
Usually a motivation migration rather than a data migration: you want capacity, load and fairness signals Linear doesn't produce. Plan the sprints first, the import second.
- Estimates carry as real numbers, not converted labels
- Sub-issue structure survives the move
- Cycle → sprint is the only workflow decision that matters
Trello teams
The migration is easy and the discipline is the project. Budget most of your effort into the first two sprints, where estimation is a new habit rather than a new tool.
- Lists and checklists translate without interpretation
- Capacity and workload answers arrive immediately
- Colour-taxonomy labels get deliberately pruned, not silently dropped
Notion teams
Run the discovery phase as a triage: docs stay in Notion, work items become issues. Teams that skip this end up with a Jira full of meeting notes.
- Documentation stops pretending to be a backlog
- Select-driven workflows map to real board columns
- Relations become traceable links instead of page titles
What early teams said afterwards
“The part that convinced our lead engineer was the archive step. Read-only, not deleted — so saying yes wasn't irreversible.”
“We came from Notion and spent longer deciding which pages were work than we did importing anything. They warned us, in writing, before we started.”
Example quotes from early-access conversations. We publish attributed testimonials only with permission.
Keep looking
Neighbouring pages
Jira, with the layer on top
What JiraVision adds by reading your Jira — and what it deliberately never replaces.
Asana vs JiraVision
Where task-list thinking stops answering sprint questions, feature by feature.
Linear vs JiraVision
Closest shape of the four sources — the gap is capacity and fairness signals.
Notion vs JiraVision
Docs versus backlog: the triage that decides what a migration actually means.
AI Scrum Master
The reason most teams switch on the layer first and worry about the tool later.
Pricing
Per-seat, 14-day trial, no card — and no onboarding fee bolted onto a migration.
Bring your history with you — or don't move yet
Start with a read-only discovery pass and a scope list you can argue with. If the answer is “not worth it”, we'll say so in writing, because the fit section above isn't decoration.